A national laboratory of the U.S.

Department of Energy
Office of Energy Efficiency & Renewable Energy

National Renewable Energy Laboratory

Innovation for Our Energy Future

Technical Report
FAST User’s Guide NREL/TP-500-38230
October 2005
Updated August 2005
J.M. Jonkman
M.L. Buhl Jr.

This guide is frequently updated and revised.

For the latest online version see
http://wind.nrel.gov/designcodes/simulators/fast/FAST.pdf

NREL is operated by Midwest Research Institute ● Battelle Contract No. DE-AC36-99-GO10337

 Technical Report
FAST User’s Guide NREL/TP-500-38230
October 2005
Updated August 2005
J.M. Jonkman
M.L. Buhl Jr.
Prepared under Task Nos. WER1.1220, WER2.1220, WER3.1910,
WER4.3102, and WER5.3115

National Renewable Energy Laboratory

1617 Cole Boulevard, Golden, Colorado 80401-3393
303-275-3000 • www.nrel.gov
Operated for the U.S. Department of Energy
Office of Energy Efficiency and Renewable Energy
by Midwest Research Institute • Battelle
Contract No. DE-AC36-99-GO10337
 NOTICE

This report was prepared as an account of work sponsored by an agency of the United States government.
Neither the United States government nor any agency thereof, nor any of their employees, makes any
warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or
usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not
infringe privately owned rights. Reference herein to any specific commercial product, process, or service by
trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement,
recommendation, or favoring by the United States government or any agency thereof. The views and
opinions of authors expressed herein do not necessarily state or reflect those of the United States
government or any agency thereof.

Available electronically at http://www.osti.gov/bridge

Available for a processing fee to U.S. Department of Energy

and its contractors, in paper, from:
U.S. Department of Energy
Office of Scientific and Technical Information
P.O. Box 62
Oak Ridge, TN 37831-0062
phone: 865.576.8401
fax: 865.576.5728
email: mailto:reports@adonis.osti.gov

Available for sale to the public, in paper, from:

U.S. Department of Commerce
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
phone: 800.553.6847
fax: 703.605.6900
email: orders@ntis.fedworld.gov
online ordering: http://www.ntis.gov/ordering.htm

Printed on paper containing at least 50% wastepaper, including 20% postconsumer waste
 DOCUMENT REVISION RECORD

Revision Date Description

1.00 09-Jul-2002 First publication. Supports FAST v4.0.
1.10 04-Sep-2002 Inserted an additional output channel (TipSpdRat or TSR) and additional prefixes (“-”, “_”,
“m”, or “M”) for reversing the sign of the selected output channels.
Still supports FAST v4.0.
1.20 28-Feb-2003 Clarified the descriptions of several input variables.
1.21 31-Mar-2003 Replaced drivetrain references from “rotational” to “torsional.”
1.22 03-Apr-2003 Fixed the labels on the drawings of the coordinate systems.
1.30 08-Sep-2003 Added description of the FAST-to-ADAMS preprocessor and associated inputs.
Added placeholders for the inputs of the currently undocumented FAST linearization and
noise prediction routines.
Clarified the descriptions of several input variables.
Renamed input switch TeetDMod to TeetMod and altered its functionality so that routine
UserTeet() can implement user-defined teeter spring and damper models.
Added supplementary output names for the blade root in-plane, out-of-plane, flap, and edge
bending moments.
1.31 03-Oct-2003 Added description of the FAST linearization analysis capability.
Clarified the descriptions of several input variables.
1.32 07-Oct-2003 Updated the example summary, AeroDyn, and linearization output files.
Had the Linearization chapter type edited by NREL communications.
1.40 28-Oct-2003 Changed the extension of the input file names from .fad or .inp to “.fst”
Added Alpha, PrecrvRef, and PreswpRef to the FAST-to-ADAMS preprocessor.
Supports FAST v4.4.
1.41 07-Nov-2003 Inserted additional output channels for the blade and tower torsional deflections.
Moved the origin of the nacelle/yaw coordinate system to the tower-top.
1.42 11-Dec-2003 Documented that the dummy supplied UserGen() routine now calls the sample
UserVSCont() routine.
2.00 05-Mar-2004 Added description of new FAST furling capability and associated I/O.
Added additional output channels for blade tip-to-tower clearances.
Clarified the descriptions of several input variables.
Supports FAST v5.0.
2.10 01-Oct-2004 Added a chapter on how to compile FAST.
Documented upgrades relating to turbine control, including yaw and high-speed shaft brake
control.
Added a description on the interface between FAST and Simulink and on the interface
between FAST and Bladed-style DLL master controllers.
Clarified the descriptions of several input variables.
Added supplementary output names for the angular speed and acceleration of the high-speed
shaft and generator.
Supports FAST v5.1.
3.00 09-Jun-2005 Added description of new platform motion and loading functionality.
Documented upgrades relating to turbine control and output capabilities.
Added additional output channels and names for the angular (rotational) deflections of the
tower-top and blade tips.
Clarified the descriptions of several input and output parameters.
Supports FAST v6.0.
3.01 12-Aug-2005 Incorporated editorial changes from NREL communications.
Added reference to the certification report from Germanischer Lloyd WindEnergie.
Added a Sample Input Files section in the Input Files chapter.
Added flowcharts for explaining how the progam uses the blade pitch, nacelle yaw, variable-
speed, generator, and HSS brake control input parameters during runtime.
Modified the description of how Bladed-style DLL controllers are interfaced.
Added supplementary output names for blade pitch angles and nacelle yaw motions and
moment and modified the description of output CThrstRad.
Still supports FAST v6.0.

FAST User's Guide i Last updated August 2005 for version 6.0
FAST User's Guide ii Last updated August 2005 for version 6.0
 TABLE OF CONTENTS

Document Revision Record ..................................................................................................i

Table of Contents.................................................................................................................iii

Figures ...................................................................................................................................v

Tables.....................................................................................................................................v

Foreword..............................................................................................................................vii

Acknowledgements ...........................................................................................................viii

Upgrading to FAST v6.0 from v5.1 .....................................................................................ix

Upgrading to FAST v5.1 from v5.0 ......................................................................................x

Upgrading to FAST v5.0 from v4.4 ....................................................................................xii

Using This Manual .............................................................................................................xiv

Modes of Operation ..............................................................................................................1

Running FAST .......................................................................................................................3

Compiling FAST ....................................................................................................................5

Model Description .................................................................................................................7

General Description ................................................................................................................7

Coordinate Systems................................................................................................................7

Inertial Frame Coordinate System .................................................................................... 8

Tower-Base Coordinate System ....................................................................................... 8

Tower-Top/Base-Plate Coordinate System....................................................................... 8

Nacelle/Yaw Coordinate System....................................................................................... 8

Shaft Coordinate System .................................................................................................. 9

Azimuth Coordinate System.............................................................................................. 9

Hub Coordinate System .................................................................................................... 9

Coned Coordinate Systems .............................................................................................. 9

Blade Coordinate Systems................................................................................................ 9

Turbine Layout ......................................................................................................................10

Flexible Tower and Blades....................................................................................................10

Drivetrain...............................................................................................................................11

Generator ..............................................................................................................................11

Nacelle Yaw ..........................................................................................................................12

Rotor-Furl ..............................................................................................................................12

Tail-Furl .................................................................................................................................13

Rotor-Teeter..........................................................................................................................14

Support Platform ...................................................................................................................15

Rotor Aerodynamics..............................................................................................................16

Tail Fin Aerodynamics...........................................................................................................16

Controls ...............................................................................................................................25

General Description ..............................................................................................................25

Blade Pitch Control ...............................................................................................................25

Variable-Speed Torque Control ............................................................................................26

FAST User's Guide iii Last updated August 2005 for version 6.0
 HSS Brake Control................................................................................................................27

Nacelle Yaw Control..............................................................................................................28

Master Controllers and the Bladed-Style DLL Interface........................................................30

Tip Brakes .............................................................................................................................32

Simulating Special Events.....................................................................................................33

Turbine Startup ............................................................................................................... 33

Normal Pitch-to-Feather Shutdown ................................................................................. 33

Shutdown Where One Blade Fails to Feather................................................................. 33

One Blade Feathers Accidentally .................................................................................... 33

HSS Brake Shutdown after Loss of Grid ......................................................................... 33

HSS Brake Shutdown with Generator Brake .................................................................. 33

Normal Tip Brake Shutdown ........................................................................................... 33

Tip Brake Shutdown after Loss of Grid ........................................................................... 33

Accidental Deployment of a Tip Brake ............................................................................ 33

Idling Turbine .................................................................................................................. 33

Parked Turbine................................................................................................................ 34

Simulink Interface ...............................................................................................................35

General Description ..............................................................................................................35

Getting Started ......................................................................................................................35

Specific Input File Options for the FAST S-Function.............................................................36

Customizing the Simulink Model ...........................................................................................37

Linearization ........................................................................................................................39

General Description ..............................................................................................................39

Periodic Steady State Solution..............................................................................................39

Model Linearization ...............................................................................................................41

Post Processing ....................................................................................................................43

ADAMS Preprocessor.........................................................................................................45

General Description ..............................................................................................................45

Compiling and Linking ADAMS .............................................................................................45

Guidelines for Creating ADAMS Datasets ............................................................................46

Running ADAMS ...................................................................................................................48

Description of the Extracted ADAMS Datasets .....................................................................49

Input Files ............................................................................................................................53

Sample Input Files ................................................................................................................53

Primary Input File ..................................................................................................................53

Tower Input File ....................................................................................................................54

Blade Input Files ...................................................................................................................54

AeroDyn Input Files...............................................................................................................55

Platform Input File .................................................................................................................55

Furling Input File ...................................................................................................................55

ADAMS-Specific Input File....................................................................................................55

Linearization Control-Input File .............................................................................................55

Output Files .........................................................................................................................93

References.........................................................................................................................125

FAST User's Guide iv Last updated August 2005 for version 6.0
 FIGURES

Figure 1. Modes of Operation. ......................................................................................... 1

Figure 2. Example display output..................................................................................... 3

Figure 3. Tower-base coordinate system......................................................................... 8

Figure 4. Tower-top/base-plate coordinate system.......................................................... 8

Figure 5. Nacelle/yaw coordinate system. ....................................................................... 8

Figure 6. Shaft coordinate system. .................................................................................. 9

Figure 7. Hub coordinate system. .................................................................................... 9

Figure 8. Coned coordinate system. ................................................................................ 9

Figure 9. Blade coordinate system. ............................................................................... 10

Figure 10. Tower mode shapes. .................................................................................... 10

Figure 11. Blade layout. ................................................................................................. 11

Figure 12. Simple-induction-generator torque/speed curve. .......................................... 11

Figure 13. Thevenin-equivalent-induction-generator torque/speed curve. .................... 12

Figure 14. Layout of a downwind, two-bladed turbine (a) and a close-up of its hub (b). 17

Figure 15. Layout of a two-bladed rotor illustrating δ3. ................................................... 18

Figure 16. Layout of a conventional, upwind, three-bladed turbine. .............................. 19

Figure 17. Layout of a three-bladed, upwind, furling turbine: furl axes. ......................... 20

Figure 18. Layout of a three-bladed, upwind, furling turbine: rotor-furl structure ........... 21

Figure 19. Layout of a three-bladed, upwind, furling turbine: tail-furl structure. ............. 22

Figure 20. Support platform / foundation layout............................................................. 23

Figure 21. Flowchart of Blade Pitch Control Runtime Options....................................... 26

Figure 22. Torque/speed curve for simple variable-speed control. ................................ 26

Figure 23. Flowchart of Variable-Speed, Generator, & HSS Brake Control Runtime

Options. ........................................................................................................ 28

Figure 24. Flowchart of Nacelle Yaw Control Runtime Options. .................................... 30

Figure 25. Interface to a Bladed-Style Master Controller DLL. ...................................... 31

Figure 26. FAST Wind Turbine Block. ........................................................................... 35

Figure 27. Simulink Model OpenLoop.mdl. .................................................................... 36

Figure 28. Periodic Steady State Computation.............................................................. 41

Figure 29. Input and Output Files. ................................................................................. 54

Figure 30. Sample output file. ...................................................................................... 120

Figure 31. Sample linearization model file. .................................................................. 121

Figure 32. Sample summary file. ................................................................................. 122

Figure 33. Sample AeroDyn options file. ..................................................................... 124

TABLES
Table 1. FAST Source Files............................................................................................. 5

Table 2. User-Specified Routines. ................................................................................... 6

Table 3. Parameter Settings to be Used With Bladed-Style Master Controller DLLs. ... 31

Table 4. Differences Between FAST’s and Bladed’s Interface to Master Controller

DLLs................................................................................................................. 32

Table 5. Control Input Settings. ..................................................................................... 43

Table 6. Wind Input Disturbance Settings...................................................................... 43

Table 7. Sample Models Provided with the FAST Archive. ........................................... 53

FAST User's Guide v Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters......................................................................... 56

Table 9. Tower-Input-File Parameters. .......................................................................... 70

Table 10. Blade-Input-File Parameters. ......................................................................... 73

Table 11. AeroDyn-Input-File Parameters. .................................................................... 77

Table 12. Platform-Input-File Parameters...................................................................... 80

Table 13. Furling-Input-File Parameters. ....................................................................... 82

Table 14. ADAMS-Specific-Input-File Parameters......................................................... 88

Table 15. Linearization Control-Input-File Parameters. ................................................. 90

Table 16. Output Parameters for Wind Motions............................................................. 94

Table 17. Output Parameters for Blade 1 Tip Motions................................................... 95

Table 18. Output Parameters for Blade 2 Tip Motions................................................... 96

Table 19. Output Parameters for Blade 3 Tip Motions................................................... 98

Table 20. Output Parameters for Blade 1 Local Span Motions.................................... 100

Table 21. Output Parameters for Blade Pitch Motions................................................. 101

Table 22. Output Parameters for Teeter Motions. ....................................................... 101

Table 23. Output Parameters for Shaft Motions. ......................................................... 102

Table 24. Output Parameters for Nacelle Inertial Measurement Unit Motions............. 103

Table 25. Output Parameters for Rotor-Furl Motions................................................... 104

Table 26. Output Parameters for Tail-Furl Motions...................................................... 104

Table 27. Output Parameters for Nacelle Yaw Motions............................................... 104

Table 28. Output Parameters for Tower-Top, Yaw-Bearing Motions. .......................... 105

Table 29. Output Parameters for Local Tower Motions. .............................................. 107

Table 30. Output Parameters for Platform Motions. .................................................... 108

Table 31. Output Parameters for Blade 1 Root Loads................................................. 110

Table 32. Output Parameters for Blade 2 Root Loads................................................. 111

Table 33. Output Parameters for Blade 3 Root Loads................................................. 112

Table 34. Output Parameters for Blade 1 Local Span Loads. ..................................... 113

Table 35. Output Parameters for Hub and Rotor Loads. ............................................. 114

Table 36. Output Parameters for Shaft Strain-Gage Loads......................................... 115

Table 37. Output Parameters for Generator and HSS Loads. ..................................... 115

Table 38. Output Parameters for Rotor-Furl Bearing Loads. ....................................... 116

Table 39. Output Parameters for Tail-Furl Bearing Loads. .......................................... 116

Table 40. Output Parameters for Tail Fin Aerodynamic Loads.................................... 116

Table 41. Output Parameters for Tower-Top, Yaw-Bearing Loads.............................. 117

Table 42. Output Parameters for Tower Base Loads. ................................................. 117

Table 43. Output Parameters for Local Tower Loads. ................................................. 118

Table 44. Output Parameters for Platform Loads. ....................................................... 119

FAST User's Guide vi Last updated August 2005 for version 6.0
 FOREWORD

The FAST (Fatigue, Aerodynamics, Structures, turbine loads for design and certification" (3).
and Turbulence) Code is a comprehensive aeroelastic Additional features were also added to the Codes.
simulator capable of predicting both the extreme and These include new nacelle inertial measurement unit
fatigue loads of two- and three-bladed horizontal-axis and tower strain gage outputs, upgrades to the simple
wind turbines (HAWTs). This document covers the variable-speed control model, and new support
features of FAST and outlines its operating procedures. platform motion and loading functionality. Despite the
The FAST Code is the result of the marriage of addition of six new platform motion degrees of
three distinct codes; the FAST2 Code for two-bladed freedom, the Code was also better-optimized so that it
HAWTs; the FAST3 Code for three-bladed HAWTs; runs 15% faster than previous versions (or faster,
and the AeroDyn (1) aerodynamics subroutines for depending on the options being modeled).
HAWTs. While combining these three codes, changes This manual is an updated subset of one originally
were made in the computational loops and in the written at Oregon State University (OSU) (4). The
kinematic calculations of the FAST codes. An original manual included a detailed discussion of the
intermediate version of FAST, called FAST_AD, used theory behind FAST_AD (an earlier incarnation of
different executable files for two- and three-bladed FAST) and a validation of the code. For these two
turbines. The version of FAST documented in this topics, please refer to the original. Also available is
report, which was developed in 2002, uses a single Buhl and others (5), which is a structural verification
executable for both types of turbines. These changes of FAST_AD against ADAMS. Both FAST and
resulted in a code that runs very quickly, so the code is ADAMS use the AeroDyn subroutine set, so the
indeed, fast. structural-verification study did not provide any
In 2003, additional features were added to the verification of the aerodynamics of FAST_AD. A
FAST Code, including the ability to develop periodic more recent verification of FAST against ADAMS is
linearized state matrices for controls design and the provided in Jonkman and Buhl (6).
ability to use FAST as a preprocessor for generating The Modes of Operation chapter describes the
ADAMS® datasets of wind turbine models (“ADAMS” different types of analysis available in FAST and a
is used to imply “ADAMS®” throughout this brief description on how to run the code is provided in
document). Aeroacoustic noise prediction algorithms the Running FAST chapter. If you want to recompile
have also been introduced. FAST, you can find the information you’ll need in the
Additional features were added to the FAST Code Compiling FAST chapter. The Model Description
again in 2004. New model features added include a chapter discusses the degrees of freedom for both two-
lateral offset and skew angle of the rotor shaft, rotor- and three-bladed HAWTs. The Controls chapter
furling, tail-furling, tail inertia and aerodynamics, yaw documents methods for actively controlling many
control, and high-speed shaft (HSS) brake control. An aspects of the turbine operation during simulation.
interface has been developed between FAST and a Active controls can also be implemented in Simulink
master controller implemented as a dynamic-link- as described in the Simulink Interface chapter. The
library (DLL) in the style of Garrad Hassan's Bladed Linearization chapter documents how to extract
wind turbine software package (2). An interface has linearized wind turbine models out of FAST. The
also been developed between FAST and Simulink® functionality of using FAST as a preprocessor for
with MATLAB® (“Simulink” and “MATLAB” are creating ADAMS datasets is documented in the
used to imply “Simulink®” and “MATLAB®” ADAMS Preprocessor chapter. The Input Files chapter
throughout this document), enabling users to describes the various program input files. Finally, the
implement advanced turbine controls in Simulink’s Output Files chapter lists the possible output
convenient block diagram form. parameters. It also describes the optional summary and
In 2005, FAST and ADAMS with AeroDyn were element output files.
evaluated by Germanischer Lloyd WindEnergie and
found suitable for "the calculation of onshore wind

FAST User's Guide vii Last updated August 2005 for version 6.0
 ACKNOWLEDGEMENTS

Funding for FAST development came from the the AeroDyn routines that we link with the FAST
U.S. Department of Energy under contract No. DE- structural-dynamics routines and for writing an
AC36-83CH10093 to the National Renewable Energy example pitch-control routine. Tim McCoy of Global
Laboratory (NREL). Later improvements were funded Energy Concepts and Craig Hansen of Windward
by the U.S. Department of Energy under contract No. Engineering took time from their busy schedules to
DE-AC36-98-GO10337 to NREL. review drafts of this manual. Our editors, Kathy
The authors would like to thank Tim Weber, Lisa O’Dell, Ruth Baranowski, and Janie Homan, helped a
Freeman, and Ross Harman of OSU; Norm Weaver of lot by converting our twisted prose into readable
InterWeaver Consulting; and Kirk Pierce, formerly of English. Another tip of the hat goes to the many folks
NREL, for their past developments of FAST and in the wind-energy industry who tested FAST and
FAST_AD. We would also like to thank the folks at provided us with their valuable feedback.
Windward Engineering for developing and interfacing

FAST User's Guide viii Last updated August 2005 for version 6.0
 UPGRADING TO FAST V6.0 FROM V5.1

This section describes how to update input files the tower-base was stationary in v5.1. See Table 16 of
created previously for FAST v5.1 so that they are the Output Files chapter for more information.
compatible with FAST v6.0. New users can skip to the The names of the output channels pertaining to the
section entitled Using This Manual. blade tip accelerations were also changed since they
FAST v6.0 contains an improvement to the simple are now output in the local blade coordinate system
variable-speed control model, a few upgrades and instead of the undeflected coordinate system—see
modifications to the output capabilities, and new Table 17, Table 18, and Table 19 of the Output Files
support platform motion and loading functionality. chapter for more information. The names of the output
The simple variable-speed control model has been channels pertaining to the tower-top / yaw bearing
upgraded to include a linear transition Region 2½, in angular (rotational) velocities and accelerations were
addition to the previously-available Regions 2 and 3. also changed since they are now output in the tower-
See the updated Variable-Speed Torque Control top / base-plate coordinate system instead of the tower
section of the Controls chapter and the Turbine Control base coordinate system—see Table 28 of the Output
section of Table 8 for more information. Files chapter for more information. These changes
You can now specify up to 5 strain gage locations were made so that the associated outputs are in
along the tower for examining local tower motions and coordinate systems that are easier to measure in the
loads output, similar to what is available for blade 1. “real world”.
You can also specify the location in the nacelle of an Updating to FAST v6.0 from v5.1 requires a few
inertial measurement unit (IMU)—this location is used modifications to FAST’s primary input file, even if you
for new nacelle motion outputs. See the Output section want to keep your turbine model configuration
of Table 8 and the Output Files chapter for more unchanged. Additionally, to take advantage of FAST’s
information. new support platform motion functionality, a new file
The new support platform motion and loading of inputs must be assembled. In addition to the
functionality represents a major expansion in the changes listed below, please be aware that all of the
number of degrees of freedom and loading options inputs that had a lower limit restriction of –180 degrees
available in FAST. Detailed information on these new in v5.1 where changed to greater than –180 degreees
features and associated inputs are presented throughout in FAST v6.0. This change was made since the –180
this manual where appropriate. In particular, see the degrees (inclusive) restriction caused problems in
Support Platform section and Figure 20 of the Model ADAMS where the ATAN2() FUNCTION is used to
Description chapter, the Platform Input File section of initialize variables.
the Input Files chapter, the Platform Model section of The changes to the primary input file are as
Table 8, and Table 12. follows (in the order they appear in the file):
Despite the addition of six new platform motion • Replace inputs RatGenSp and Reg2TCon
DOFs (translational surge, sway, and heave and with inputs VS_RtGnSp, VS_RtTq,
rotational roll, pitch and yaw), the Code was also VS_Rgn2K, and VS_SlPc in the turbine
better-optimized so that it runs 15% faster than control section. Inputs VS_RtGnSp and
previous versions (or faster, depending on the options VS_Rgn2K are simply renamed versions of
being modeled). inputs RatGenSp and Reg2TCon.
With the addition of support platform motion VS_RtTq and VS_SlPc are new inputs used
functionality, it made sense to add more information to to specify the rated generator torque in Region
the FAST summary (.fsm) file. See the Output Files 3 and the rated generator slip percentage in
chapter, especially Figure 32, for more information Region 2½, respectively. These new inputs
It also made made sense to rename some of the are needed to specify the characteristics of the
output parameters. The output channels WindVxt, improved simple variable-speed generator
WindVyt, and WindVzt in FAST v5.1 were renamed to controller, which now includes Region 2½ in
WindVxi, WindVyi, and WindVzi in v6.0, respectively, addition to Regions 2 and 3.
since the wind speeds relative to the inertia frame (i) • Add a new platform model section including a
are now more important than the wind speeds relative header plus inputs PtfmModel and PtfmFile
to the tower-base frame (t), which can now move between the Thevenin-equivalent induction
relative to the inertia frame. WindVxi, WindVyi, and generator and tower sections. PtfmModel is a
WindVzi in FAST v6.0 will give the same results as switch used to indicate the type of support
WindVxt, WindVyt, and WindVzt gave in v5.1, since platform as follows: {0: none, 1: onshore, 2:

FAST User's Guide ix Last updated August 2005 for version 6.0
 fixed bottom offshore, 3: floating offshore}. VS_SlPc = 9999.9E-9
If PtfmModel is not 0, FAST will read in an (a very small don’t care > 0.0)
additional file of inputs for defining the model
properties of the support platform. PtfmFile PtfmModel = 0
is the name of this file. In FAST v6.0, all
nonzero PtfmModel options will work the PtfmFile = <may be left blank>
same way by reading in PtfmFile. In future
versions, the format of PtfmFile will depend NcIMUxn = 0.0 (a don’t care)
on which PtfmModel option is selected.
• Add inputs NcIMUxn, NcIMUyn, and NcIMUyn = 0.0 (a don’t care)
NcIMUzn between inputs SttsTime and
ShftGagL in the output section. These three NcIMUzn = 0.0 (a don’t care)
new inputs define the distance from the tower-
top to the nacelle inertial measurement unit in NTwGages = 0
the downwind, lateral, and vertical directions,
respectively. TwrGagNd = <may be left blank>
• Add inputs NTwGages and TwrGagNd
between inputs ShftGagL and NBlGages in Finally, if you use the FAST-to-ADAMS
the output section. Inputs NTwGages and preprocessor to create ADAMS wind turbine datasets,
TwrGagNd define the tower strain gage upgrading from FAST v5.1 to v6.0 also requires you to
locations like inputs NBlGages and upgrade from v12.17 to v12.18 of the ADAMS to
BldGagNd do for blade 1. AeroDyn (A2AD) source files and to recompile the
If you want to leave your model unchanged when ADAMS user-created dynamic-link-library (DLL).
converting to FAST v6.0, use the following This is because ADAMS datasets generated using
equivalency relationships when defining the new inputs FAST v6.0 must be simulated with an ADAMS user-
from the old, now obsolete, inputs: created DLL compiled using the source files from
A2AD v12.18. All of the new features for FAST v6.0
VS_RtGnSp = RatGenSp listed above are also available in the FAST-to-ADAMS
preprocessor.
VS_RtTq = Reg2TCon • ( RatGenSp^2 )

VS_Rgn2K = Reg2Tcon

UPGRADING TO FAST V5.1 FROM V5.0

This section describes how to update input files convenient block diagram form. An interface has also
created previously for FAST v5.0 so that they are been developed between FAST and a master controller
compatible with FAST v5.1. New users can skip to the dynamic-link-library (DLL) implemented in the style
section entitled Using This Manual. of Garrad Hassan’s Bladed wind turbine software
FAST v5.1 contains many upgrades relating to package (this interface is not linked with the distributed
turbine control. Yaw control features have been added executable, but is available as a source file containing a
to the simulation and linearization analysis modes and set of subroutines, which can be compiled with FAST
the ADAMS preprocessor. “Hooks” for user-defined in place of the built-in example control routines; the
high-speed shaft brake models have also been added to same set of routines can be used to interface FAST-
FAST and the FAST-generated ADAMS datasets. generated ADAMS datasets with Bladed DLL
Within user-defined routines, you now have the option controllers). Finally, the ramp-up of aerodynamic
of switching DOFs on-or-off at runtime and you now loads, which occurred over the first two seconds of
have the ability of accessing the current value of any simulation in previous versions, has been eliminated;
available output parameter without changing the thus, trim solutions and/or start-up transients may be
number of arguments passed to the routines. The user- different than in previous versions.
defined pitch control routine written by Craig Hansen, Detailed information on these new features and
which is linked with the executable version of FAST, their associated input parameters are presented
has also been upgraded. An interface has been throughout this manual where appropriate. In
developed between FAST and Simulink, so that you particular, a description of yaw control is provided in
can implement advanced turbine controls in Simulink’s the new Nacelle Yaw Control section of the Controls

FAST User's Guide x Last updated August 2005 for version 6.0
 chapter, user-defined high-speed shaft brake control is If you want to leave your turbine model
described in the HSS Brake Control section of the configuration unchanged when converting to FAST
Controls chapter, master controller DLLs are described v5.1, use the following equivalency relationships when
in the Master Controllers and the Bladed-Style DLL defining the new inputs from the old, now obsolete,
Interface section of the Controls chapter, upgrades to inputs:
Craig’s pitch controller should be apparent by
YCMode = 0
examining the example Pitch.ipt input file located in
FAST’s CertTest folder, and a description of FAST’s
new interface to Simulink is given in the new Simulink TYCon = 9999.9 (a don’t care)
Interface chapter.
Updating to FAST v5.1 from v5.0 requires a few PCMode = 0 if PCMode was 0 in v5.0
modifications to FAST’s primary and linearization = 1 if PCMode was 1 or 2 in v5.0
control-input files, even if you want to keep your
turbine model configuration unchanged. The changes VSContrl = same value as VSContrl in v5.0
to the primary input file are as follows (in the order
they appear in the file): HSSBrMode = 1
• Add inputs YCMode and TYCOn before
PCMode. Inputs YCMode and TYCOn TYawManS = 9999.9 (a don’t care > TMax)
define yaw control options like inputs
PCMode and TPCOn do for pitch control. TYawManE = 9999.9 (a don’t care ≥

• Change pitch control mode input PCMode so TYawManS)

that 0 means none, 1 means user-defined from

routine PitchCntrl(), and 2 means user- NacYawF = 0.0 (a don’t care)
defined from Simulink.
• Change variable speed control mode input TrimCase = 2 if TrimCase was 1 in v5.0
VSContrl so that 0 means none, 1 means = 3 if TrimCase was 2 in v5.0
simple variable speed control model, 2 means
user-defined from routine UserVSCont(), and NInputs = 0 if CntrlInpt was 0 in v5.0
3 means user-defined from Simulink. = 1 if CntrlInpt was 1 or 2 in v5.0
• Add input HSSBrMode between TimGenOf = NumBl if CntrlInpt was 3 in v5.0
and THSSBrDp. Input HSSBrMode = 2 if CntrlInpt was 4 in v5.0
provides a switch between the simple and = 1 + NumBl if CntrlInpt was 5 in v5.0
user-defined from routine UserHSSBr() high-
speed shaft brake models. CntrlInpt = <may be left blank> if CntrlInpt
• Add inputs TYawManS, TYawManE, and was 0 in v5.0
NacYawF between TBDepISp(3) and = 3 if CntrlInpt was 1 in v5.0
TPitManS(1). Inputs TYawManS, = 4 if CntrlInpt was 2 in v5.0
TYawManE, and NacYawF define override = 5,6 if CntrlInpt was 3 and
yaw control options like inputs TPitManS, NumBl = 2 in v5.0
TPitManE, and BlPitchF do for pitch control. = 5,6,7 if CntrlInpt was 3 and
The changes to the linearization control-input file NumBl = 3 in v5.0
are as follows (in the order they appear in the file): = 3,4 if CntrlInpt was 4 in v5.0
= 3,5,6 if CntrlInpt was 5 and
• Change trim case input TrimCase so that 1
NumBl = 2 in v5.0
means find nacelle yaw, 2 means find
= 3,5,6,7 if CntrlInpt was 5 and
generator torque (region 2 linearization), and
NumBl = 3 in v5.0
3 means find collective blade pitch (region 3
linearization). If you compile FAST yourself, please note that
• Add input NInputs before CntrlInpt. new source files FAST_Prog.f90, UserVSCont_KP.f90,
NInputs defines the number of control inputs and BladedDLLInterface.f90 have been added to, and
in the output linearized state matrices. source file PitchCntrl.f90 has been removed from, the
• Change input CntrlInpt so that it is a list of Source folder in the FAST archive. Please see the new
control inputs from 1 to NInputs where 1 is Compiling FAST chapter for more information.
nacelle yaw angle, 2 is nacelle yaw rate, 3 is Finally, if you use the FAST-to-ADAMS
generator torque, 4 is collective blade pitch, 5 preprocessor to create ADAMS wind turbine datasets,
is individual pitch of blade 1, 6 is individual upgrading from FAST v5.0 to v5.1 also requires you to
pitch of blade 2, and 7 is individual pitch of upgrade from v12.16 to v12.17 of the ADAMS to
blade 3. AeroDyn (A2AD) source files and to recompile the

FAST User's Guide xi Last updated August 2005 for version 6.0
 ADAMS user-created dynamic-link-library (DLL). A2AD v12.17. All of the new features for FAST v5.1
This is because ADAMS datasets generated using listed above are also available in the FAST-to-ADAMS
FAST v5.1 must be simulated with an ADAMS user- preprocessor.
created DLL compiled using the source files from

UPGRADING TO FAST V5.0 FROM V4.4

FAST v5.0 contains a major expansion in the tilt angle, replacing what used to be input
range of turbine configurations available relative to NacTilt.
those available in FAST v4.4. This section describes • Add input YawBrMass before NacMass.
how to update input files created previously for FAST YawBrMass defines the point mass of the
v4.4 so that they are compatible with FAST v5.0. New yaw bearing.
users can skip to the section entitled Using This • Remove input NacTIner.
Manual. • Replace the entire nacelle-tilt section, which
New to FAST v5.0 is the availability of a lateral includes the header plus inputs TiltSpr,
offset and skew angle of the rotor shaft, rotor-furling, TiltDamp, TiltSStP, TiltHStP, TiltSSSp, and
tail-furling, and tail inertia and aerodynamics. These TiltHSSp, with a furling section, which
new features support the analysis of most small wind includes a header plus inputs Furling and
turbine configurations. A few new mass and inertia FurlFile. Furling is a flag used to tell FAST
terms are also available for conventional turbine whether or not to read in an additional file of
configurations including a yaw bearing point mass, a inputs for defining the model configuration of
lateral offset for the nacelle mass, and a hub inertia for a furling turbine. FurlFile is the name of this
3-bladed rotors (the hub inertia was previously file.
available only for 2-bladed rotor configurations). The changes to the ADAMS-specific input file are
While upgrading FAST, we tried to minimize the as follows (in the order they appear in the file):
number of changes to the input files as a courtesy to • Add input LSSLength between inputs
our users; nevertheless, some changes were HSSLength and GenRad. LSSLength
unavoidable. Updating to FAST v5.0 from v4.4 defines the length of the low-speed shaft
requires a few modifications to FAST’s primary and cylinder used for LSS graphical output in
ADAMS-specific input files, even if you want to keep ADAMS. This is in contrast to how the LSS
your turbine model configuration unchanged. previously extended from the hub to the yaw
Additionally, to take advantage of FAST’s new model axis.
configuration properties for small wind turbines, a new • Remove inputs TetPnLngth and
file of inputs must be assembled. Detailed information TeetPinRad. These inputs were deemed
on the new features and associated inputs are presented unnecessary graphical output in ADAMS.
throughout this manual where appropriate. In • Add input BoomRad after input
particular, a description of the input file for specifying ThkOvrChrd at the end of the file.
additional model properties for a furling turbine is BoomRad defines the radius of the tail boom
provided in Table 13. cylinder used for tail boom graphical output in
The changes to the primary input file are as ADAMS.
follows (in the order they appear in the file): If you want to leave your turbine model
• Remove input TiltDOF. configuration unchanged when converting to FAST
• Remove input NacTilt. v5.0, use the following equivalency relationships when
• Replace inputs ParaDNM and PerpDNM defining the new inputs from the old, now obsolete,
with inputs NacCMxn, NacCMyn, and inputs:
NacCMzn. These three new inputs define the
distance from the tower-top to the nacelle NacCMxn = ParaDNM • COS( NacTilt ) -
mass center in the downwind, lateral, and PerpDNM • SIN( NacTilt )
vertical directions, respectively. This is in
contrast to how ParaDNM and PerpDNM NacCMyn = 0.0
previously located the nacelle mass center
relative to the rotor shaft. NacCMzn = ParaDNM • SIN( NacTilt ) +
• Add input ShftTilt between inputs TwrRBHt PerpDNM • COS( NacTilt ) + Twr2Shft
and Delta3. ShftTilt defines the rotor shaft
ShftTilt = NacTilt

FAST User's Guide xii Last updated August 2005 for version 6.0
 YawBrMass = 0.0

Furling = False

FurlFile = <may be left blank>

LSSLength = ABS( OverHang )

BoomRad = 0.0
Note that the above equations are only applicable
if your existing FAST v4.4 model had the nacelle-tilt
degree of freedom disabled (TiltDOF = False). If your
existing FAST v4.4 model had the nacelle-tilt degree of
freedom enabled (TiltDOF = True)*, you will now need
to assemble the FurlFile in order to define the model
properties of your tilting turbine. This is because the
nacelle-tilt degree of freedom has been replaced with
the more general rotor-furl degree of freedom.
Finally, if you use the FAST-to-ADAMS
preprocessor to create ADAMS wind turbine datasets,
upgrading from FAST v4.4 to v5.0 also requires you to
upgrade from v12.15 to v12.16 of the ADAMS to
AeroDyn (A2AD) source files and to recompile the
ADAMS user-created dynamic-link-library (DLL).
This is because ADAMS datasets generated using
FAST v5.0 must be simulated with an ADAMS user-
created DLL compiled using the source files from
A2AD v12.16. All of the new features for FAST v5.0
listed above are also available in the FAST-to-ADAMS
preprocessor.

*
The only example, known by the authors, of a wind
turbine with a tilting-nacelle degree of freedom is the Wind
Eagle 300 turbine from Cannon Wind Eagle Corporation.

FAST User's Guide xiii Last updated August 2005 for version 6.0
 USING THIS MANUAL
We use several typographic conventions in this
manual to make it easy to distinguish various entities.
Most titles and headings are formatted with the Arial
bold typeface. This manual uses the Times New
Roman typeface for body text. To make it easy to spot
Variable Names within the body text, we formatted
them with the Arial typeface. We did the same for
routine names but appended a pair of parentheses to the
end of the name (for example, Routine()). We
formatted file names with Times New Italic so that we
wouldn’t have to deal with the awkward situation of
having to include punctuation within the quote marks,
which might cause confusion. Examples are formatted
with the Letter Gothic typeface.

FAST User's Guide xiv Last updated August 2005 for version 6.0
 MODES OF OPERATION

FAST has two different forms of operation or simulations using FAST as a DLL interfaced with
analysis modes (see Figure 1). Switch AnalMode in Simulink.
the primary input file is used to control this mode. The second form of analysis provided in FAST is
The first analysis mode is time-marching of the linearization. FAST has the capability of extracting
nonlinear equations of motion—that is, simulation. linearized representations of the complete nonlinear
During simulation, wind turbine aerodynamic and aeroelastic wind turbine modeled in FAST. This
structural response to wind-inflow conditions is analysis capability is useful for developing state
determined in time. Active controls for determining matrices of a wind turbine “plant” to aid in controls
many aspects of the turbine operation may be design and analysis. It is also useful for determining
implemented during simulation analyses as described the full system modes of an operating or stationary
in the Controls chapter. Outputs of simulations include HAWT through the use of a simple eigenanalysis. The
time-series data on the aerodynamic loads as well as Linearization chapter documents how to extract
loads and deflections of the structural members of the linearized wind turbine models out of FAST. The
wind turbine as described in the Output Files chapter. linearization capability is only available in the
These outputs can be used, for example, to predict both Windows executable version (not the DLL interface
the extreme and fatigue loads of HAWTs. The with Simulink).
aerocoustic signature of an operating turbine is another Another feature available in FAST is the ADAMS
output that can be obtained from simulation. preprocessor. The ADAMS preprocessor feature is
Simulation analyses can be run using the separate from the two analysis modes available in
distributed Windows executable program file or as a FAST. It is not considered an analysis mode of FAST,
dynamic-link-library (DLL) interfaced with Simulink. because it does not make use of the aeroelastic wind
When running the executable version of FAST, active turbine model available in FAST. Instead, the
controls must be implemented through user-defined ADAMS preprocessor uses the input parameters
routines that have been linked with FAST during available in the FAST input files to construct an
creation of the executable or as a master controller ADAMS dataset of a complete aeroelastic wind
implemented as a DLL in the style of Garrad Hassan's turbine. ADAMS then becomes the code in which
Bladed wind turbine software package. When running different wind turbine analyses (simulation or
FAST as a DLL interfaced with Simulink, active linearization) are performed. The ADAMS
controls can be implemented in the Simulink preprocessor feature of FAST is documented, not
environment in addition to the implementations surprisingly, in the ADAMS Preprocessor chapter of
available with the FAST executable. Most of the this manual and is controlled by switch ADAMSPrep
contents of this manual relate to simulation using the in FAST’s primary input file. The ADAMS
FAST executable; there is no chapter in this manual preprocessor capability is only available in the
devoted specifically to this mode of operation. The Windows executable version (not the DLL interface
Simulink Interface chapter documents how to run with Simulink).

System User-Defined
Properties Routines

FAST AeroDyn ADAMS

Input Files Input Files Datasets

FAST AeroDyn AeroDyn ADAMS

Simulation FAST-to-ADAMS
(exe or Linearization Preprocessor Simulation
Simulink DLL) (exe only) (exe only) (ADAMS Solver)
Time-Series Periodic State Time-Series
Data Matrices Data

Figure 1. Modes of Operation.

FAST User's Guide 1 Last updated August 2005 for version 6.0

 RUNNING FAST

This section documents how to run the FAST echoes the title line from the input file. If aerodynamic
Windows executable program file that we distribute in calculations are requested in your input file, the
the FAST archive available at our Web page AeroDyn routines then print out some startup
http://wind.nrel.gov/designcodes/simulators/fast/. For messages. If running a time-marching analysis, next,
a description on how to run FAST with Simulink, see you will see one line that is repeatedly overwritten
the Simulink Interface chapter. telling you what the status of the simulation is. It will
Before you run FAST, you will want to install it in update this line periodically.
such a way that you can run it from any folder. For At the end of the simulation, FAST prints out
instructions on installing codes such as FAST, please some run-time statistics as seen in Figure 2. The Total
read Buhl (7). Real Time is the amount of time passed from the time
To run the executable, open a command prompt you started the program to the time it completes. The
window in the directory in which you want to work. Total CPU Time is a measure of the computer time
The command-line syntax is: used for the entire FAST run, which includes the time
it takes to read in the input files and set up the model.
fast [/h] [<input file>]

The difference between these two times is the amount

of time your computer was busy with other things
where: while running FAST. The Simulation Time is the
amount of time simulated. The Simulation CPU Time
/h prints a help message. is the amount of computer time use during that time-
<input file> is the name of the primary input file. marching part of the simulation. The Simulation Time
The default name is primary.fst). Ratio is the ratio of the amount of time simulated to the
simulation CPU time. The bigger this number is, the
When FAST runs, it first prints out a line faster your computer is. If the value is greater than 1,
informing you of the version and compile date of the then FAST can simulate an event in less time than it
code. If it cannot find the input file, it aborts with an would take in real life. If the value is less than 1, then
error message. If it finds a valid input file, FAST it might be time to upgrade your computer ☺.

Running FAST (v4.00, 09-Jul-2002).

FAST certification test #1 for AWT-27CR2 with many

DOFs.

Heading of the aerodyn.ipt file :

AWT-27CR aerodynamic parameters for FAST certification

test #01.

Detected hub-height wind file:

"Wind/Shr12_30.wnd"

Aerodynamics loads calculated using AeroDyn(12.46, 23-

May-2002)

Total Real Time: 7.141 seconds

Total CPU Time: 7.1406 seconds
Simulation Time: 20 seconds
Simulation CPU Time: 7.1094 seconds
Simulation Time Ratio: 2.8132

FAST completed normally.

Figure 2. Example display output.

FAST User's Guide 3 Last updated August 2005 for version 6.0

FAST User's Guide 4 Last updated August 2005 for version 6.0
 COMPILING FAST
You should not need to compile FAST unless you
want to create and link a user-defined routine, make
changes to the source code, or port FAST to an Table 1. FAST Source Files.
operating system other than Microsoft Windows. The Source File Description
FAST Windows executable program file that we FAST_Prog.f90 Contains PROGRAM
distribute in the archive can be used for all other FAST(), which guides the
purposes. program’s execution
You must include both FAST’s and AeroDyn’s FAST_Mods.f90 Contains MODULEs that
source files in a workspace in order to compile FAST. store variables used by
AeroDyn’s source code, which is available in the FAST’s routines
AeroDyn archive, is available for download from our FAST_IO.f90 Contains routines related to
Web page program input and output
http://wind.nrel.gov/designcodes/simulators/aerodyn/. FAST.f90 Contains routines that make­
All of the FAST source code resides in the Source up the “guts” of FAST,
folder of the FAST archive. The FAST Windows including the equations of
executable program file that we distribute in the FAST motion and their solution
archive is compiled using the Compaq Visual Fortran AeroCalc.f90 Contains the interface routines
(CVF) Standard Edition compiler version 6.6.B. Table between FAST and AeroDyn
1 lists the FAST source files used to compile this FAST2ADAMS.f90 Contains routines that make
program file. up the FAST-to-ADAMS
When compiling using CVF, we have the preprocessor
Debugging Level set to "Minimal", the Warning Level
FAST_Lin.f90 Contains routines used during
set to "Normal Warnings", and the Optimization Level
a linearization analysis
set to "Full Optimizations". We also use Project
SetVersion.f90 Contains a routine that sets the
Options /assume:byterecl, /compile_only, /nologo,
program version number
/stand, /traceback, and /warn:nofileopt. We
GenUse.f90 Contains general-purpose
recommend that you use the same compiler and project
routines
options when compiling FAST using the CVF
compiler. NoiseMods.f90 Contains a MODULE that
All of the CVF compiler-dependent code in FAST stores variables used by the
resides in the files called SysCVF.f90 and aeroacoustic routines
ModCVF.f90. (AeroDyn also contains some CVF NoiseSubs.f90 Contains routines related to
compiler-dependent code) If you want to port FAST to aeroacoustics
another platform or compiler, you should have to ModCVF.f90 Contains a MODULE that
change only these two files. Also included in the stores compiler-dependent
Source folder are files SysLL.f90, ModLL.f90, variables
SysLU.f90, and ModLU.f90. Files SysLL.f90 and SysCVF.f90 Contains compiler-dependent
ModLL.f90 should be used when compiling FAST in routines
Lahey Linux (LL). Files SysLU.f90 and ModLU.f90 UserSubs.f90 Contains dummy placeholders
should be used when compiling FAST in Lahey Unix of all available user-specified
(LU). In the FAST archive, we also distribute a file routines
named make_LL. This is a makefile that was PitchCntrl_ACH.f90 Contains an example pitch
developed by Hugh Currin who used it to compile an control routine written by A.
older version of FAST (v4.03) in LL. Please be aware Craig Hansen
that the NWTC does not have an LL or LU compiler, UserVSCont_KP.f90 Contains an example variable-
and as such, we have not been able to update the LL speed torque control routine
and LU source files and LL makefile to accommodate written by Kirk Pierce
upgrades to the program. Nevertheless, these sample
files should be a useful starting point if you need to FAST includes “hooks” for ten user-specified
port FAST to another operating system. routines as summarized in Table 2. Dummy
placeholder versions of these routines are all contained
within source file UserSubs.f90.

FAST User's Guide 5 Last updated August 2005 for version 6.0
 Table 2. User-Specified Routines. in Table 1, the executable version of FAST that is
distributed with the archive is linked with the example
Routine Description
PitchCntrl() routine contained in source file
PitchCntrl() User-specified blade pitch control PitchCntrl_ACH.f90 and the example UserGen() and
(either independent or rotor- UserVSCont() routines contained in source file
collective) model UserVSCont_KP.f90. Thus, the dummy placeholders
UserGen() User-specified generator torque for routines PitchCntrl(), UserGen(), and
and power model UserVSCont() are commented out within source file
UserHSSBr() User-specified high-speed shaft UserSubs.f90. The example pitch controller was
brake model written by A. Craig Hansen (ACH) and the example
UserPtfmLd() User-specified platform loading generator and variable speed controllers were written
model by Kirk Pierce (KP). Please see the aforementioned
UserRFrl() User-specified rotor-furl spring / source files for additional information on these
damper model example user-specified routines.
UserTeet() User-specified rotor-teeter spring / Also contained in the Source folder is a file named
damper model BladedDLLInterface.f90. This source file contains
UserTFin() User-specified tail fin example PitchCntrl(), UserHSSBr(), UserVSCont(),
aerodynamics model and UserYawCont() routines that may be used to
UserTFrl() User-specified tail-furl spring / interface FAST with a master controller implemented
damper model as a dynamic-link-library (DLL) in the style of Garrad
UserVSCont() User-specified variable-speed Hassan's Bladed wind turbine software package (2). In
torque and power control model order to compile FAST with these routines, you must
UserYawCont() User-specified nacelle-yaw comment-out the dummy placeholder versions of
control model routines PitchCntrl(), UserHSSBr(), UserVSCont(),
and UserYawCont() contained in source file
In order to interface FAST with your own user- UserSubs.f90 and recompile FAST with the addition of
specified routines, you can develop your own logic source file BladedDLLInterface.f90. The executable
within these dummy placeholders and recompile version of FAST that is distributed with the FAST
FAST, or comment out the appropriate dummy archive is not linked with the routines contained within
placeholders, create your own routines in their own source file BladedDLLInterface.f90. Please see the
source files, and recompile FAST while linking in Master Controllers and the Bladed-Style DLL Interface
these additional source files. For example, as implied section of the Controls chapter for more information.

FAST User's Guide 6 Last updated August 2005 for version 6.0
 MODEL DESCRIPTION

The next two DOFs arise from the first flapwise

General Description bending mode of each blade. Two more DOFs
originate from the second flapwise bending modes.
The FAST code can model the dynamic response
Blade edgewise motion accounts for the next two
of both two- and three-bladed, conventional,
DOFs. The blades are rigidly attached to the hub
horizontal-axis wind turbines. The wind turbine
through a cantilever connection. Motion of the blades
configuration may optionally include rotor-furling, tail-
is along the local principal axes. See the discussion of
furling, and tail aerodynamics—features useful in the
blade mode shapes in the Flexible Tower and Blades
analysis of most small wind turbines. The code was
section on page 10 for details.
evaluated by Germanischer Lloyd WindEnergie and
The last two DOFs are associated with furling of
found suitable for "the calculation of onshore wind
the rotor and tail about the yawing-portion of the
turbine loads for design and certification" (3).
structure atop the tower. The rotor-furl DOF can also
The FAST model employs a combined modal and
be used to model torsional flexibility in the gearbox
multibody dynamics formulation. The model for two-
mounting if you align the rotor-furl axis with the rotor
bladed turbines relates nine rigid bodies (earth, support
shaft axis. The amount of furling motion can be
platform, base plate, nacelle, armature, gears, hub, tail,
restricted with springs, dampers, or a combination of
and structure furling with the rotor) and four flexible
both.
bodies (tower, two blades, and drive shaft) through 22
The FAST code can also model a three-bladed
degrees of freedom (DOFs). Accounted for in the
HAWT with 24 DOFs. The first six DOFs originate
degrees of freedom are platform translation and
from the translational (surge, sway, and heave) and
rotation (6 DOF), tower flexibility (4 DOF), nacelle
rotational (roll, pitch, and yaw) motions of the support
yaw (1 DOF), variable generator and rotor speeds (2
platform relative to the inertia frame. The next four
DOF), blade teetering (1 DOF), blade flexibility (6
DOFs account for tower motion; two are longitudinal
DOF), rotor-furl (1 DOF), and tail-furl (1 DOF).
modes, and two are lateral modes. Yawing motion of
Flexibility in the blades and tower are characterized
the nacelle provides another DOF. The next DOF is
using a linear modal representation that assumes small
for the generator azimuth angle, and another DOF is
deflections. The three rotational DOFs of the support
the compliance in the drivetrain between the generator
platform (roll, pitch, and yaw) also employ a small
and hub/rotor. These DOFs account for variable rotor
angle approximation. The remaining DOFs may
speed and drive-shaft flexibility. The next three DOFs
exhibit large displacements without loss of accuracy.
are the blade flapwise tip motion for the first mode.
The DOFs are further described below.
Three more DOFs give the tip displacement for each
The first six DOFs (the most recent additions)
blade for the second flapwise mode. The next three
originate from the translational (surge, sway, and
DOFs are for the blade edgewise tip displacement for
heave) and rotational (roll, pitch, and yaw) motions of
the first edgewise mode. The last two DOFs are for
the support platform relative to the inertia frame.
rotor- and tail-furl.
Two DOFs originate from the first bending mode
For both the two- and three-bladed wind turbine
of the tower in the longitudinal and transverse
configurations, you can enable any combination of the
directions. Two more DOFs model the second bending
available DOFs and features during your analysis. The
mode in the same directions. The tower is rigidly
DOFs and features most applicable to you are dictated
attached to the support platform through a cantilever
by the configuration of the wind turbine you are
connection.
analyzing.
Another DOF accounts for the nacelle yaw motion,
which can be free or fixed with a torsional yaw spring.
The rotor can be either upwind or downwind with the Coordinate Systems
rotor providing yaw loads. Figure 3 through Figure 9 show the coordinate
The next DOF accounts for variations in generator systems used for input and output parameters.
speed. Another DOF accounts for drivetrain flexibility Coordinate systems t, n, h, and b conform to the
associated with torsional motion between the generator International Electrotechnical Commission (IEC)
and the hub/rotor. standard for wind turbines (8). Additional coordinate
Another DOF accounts for teeter motion of the systems i, p, a, s, and c are necessary for interpreting
blades about a pin located on the hub. Dampers, some of the output parameters. Some of the coordinate
springs, or a combination of both can restrict teeter systems used internally by FAST differ from these.
motion. FAST takes care of these conversions for you.

FAST User's Guide 7 Last updated August 2005 for version 6.0
 Inertial Frame Coordinate System and the tower bends, but it does not yaw with the
nacelle.
Origin The point about which the translational
motions of the support platform (surge, Origin A point on the yaw axis at a height of
sway, and heave) are defined. TowerHt above ground level [onshore or
mean sea level [offshore] (see Figure
xi axis Pointing in the nominal (0°) downwind
14(a), Figure 16, or Figure 20).
direction.
xp axis When the tower is not deflected, it is
yi axis Pointing to the left when looking in the
aligned with the xt axis.
nominal downwind direction.
yp axis When the tower is not deflected, it is
zi axis Pointing vertically upward opposite to
aligned with the yt axis.
gravity.
zp axis When the tower is not deflected, it is
Tower-Base Coordinate System aligned with the zt axis. It is also the yaw
This coordinate system is fixed in the support axis.

platform so that it translates and rotates with the

platform.

Origin Intersection of the center of the tower and

the tower base connection to the support

platform.

xt axis When the support platform has no pitch

or yaw displacement, it is aligned with

Zp
Yp
the xi axis (pointing horizontally in the

nominal downwind direction).

Xp
yt axis When the support platform has no roll or
yaw displacement, it is aligned with the yi
axis (pointing to the left when looking in Figure 4. Tower-top/base-plate
the nominal downwind direction). coordinate system.
zt axis Pointing up from the center of the tower.
When you request output of motions or loads for
various locations along the tower with the TwrGagNd Nacelle/Yaw Coordinate System
array, a local coordinate system similar to the standard This coordinate system translates and rotates with
tower system is used, but the local coordinate systems the top of the tower, plus it yaws with the nacelle.
orient themselves with the deflected tower.
Origin The origin is the same as that for the
tower-top/base-plate coordinate system.
xn axis Pointing horizontally toward the
0° wind nominally downwind end of the nacelle.
yn axis Pointing to the left when looking toward
the nominally downwind end of the
nacelle.
zn axis Coaxial with the tower/yaw axis and
pointing up.

zt yt zn yn
xt xn

Figure 3. Tower-base coordinate system.

Figure 5. Nacelle/yaw coordinate system.
Tower-Top/Base-Plate Coordinate System
This coordinate system is fixed to the top of the

tower. It translates and rotates as the platform moves

FAST User's Guide 8 Last updated August 2005 for version 6.0
 Shaft Coordinate System
The shaft coordinate system does not rotate with zh yh
the rotor, but it does translate and rotate with the tower
and it yaws with the nacelle and furls with the rotor. xh
The nacelle inertial measurement unit uses this
coordinate system for all of its motion outputs. Shaft
bending moments at the hub and at the position
denoted by ShftGagL use this coordinate system or the
rotating hub coordinate system shown below.
Origin Intersection of the yn-/zn-plane and the
rotor axis. Figure 7. Hub coordinate system.
xs axis Pointing along the (possibly tilted) shaft
in the nominally downwind direction.
ys axis Pointing to the left when looking from the Coned Coordinate Systems
tower toward the nominally downwind There is a coned coordinate system for each blade
end of the nacelle. that rotates with the rotor. The coordinate system does
zs axis Orthogonal with the xs and ys axes such not pitch with the blades and it also teeters in two-
that they form a right-handed coordinate bladed models. For three-bladed rotors, blade 3 is
system. ahead of blade 2, which is ahead of blade 1, so that the
order of blades passing through a given azimuth is 3-2-
1-repeat.
Origin The origin is the same as that for the
zs ys hub coordinate system.
Xc,i axis Orthogonal with the yc,i and zc,i axes
such that they form a right-handed
xs coordinate system. (i = 1, 2, or 3 for
blades 1, 2, or 3, respectively)
Yc,i axis Pointing towards the trailing edge of
blade i if the pitch and twist were zero
and parallel with the chord line. (i = 1,
Figure 6. Shaft coordinate system. 2, or 3 for blades 1, 2, or 3,
respectively)
Zc,i axis Pointing along the pitch axis towards
Azimuth Coordinate System the tip of blade i. (i = 1, 2, or 3 for
The azimuth, or a, coordinate system is located at blades 1, 2, or 3, respectively)
the origin of the shaft coordinate system, but it rotates Zc,i
with the rotor. When Blade 1 points up, the azimuth Yc,i
and shaft coordinate systems are parallel. For three-
bladed rotors, blade 3 is ahead of blade 2, which is
ahead of blade 1, so that the order of blades passing Xc,i
through a given azimuth is 3-2-1-repeat.
Hub Coordinate System
The hub coordinate system rotates with the rotor.
It also teeters in two-bladed models.
Origin Intersection of the rotor axis and the plane Figure 8. Coned coordinate system.
of rotation (non-coned rotors) or the apex
of the cone of rotation (coned rotors).
xh axis Pointing along the hub centerline in the Blade Coordinate Systems
nominal downwind direction. These coordinate systems are the same as the
yh axis Orthogonal with the xh and zh axes such coned coordinate systems, except that they pitch with
that they form a right-handed coordinate the blades and their origins are at the blade root. For
system. three-bladed rotors, blade 3 is ahead of blade 2, which
zh axis Perpendicular to the hub centerline with is ahead of blade 1, so that the order of blades passing
the same azimuth as Blade 1. through a given azimuth is 3-2-1-repeat.

FAST User's Guide 9 Last updated August 2005 for version 6.0
 Origin Intersection of the blade’s pitch axis Flexible Tower and Blades
and the blade root.
FAST models flexible elements, such as the tower
xb,i axis Orthogonal with the yb and zb axes such
and blades, using a linear modal representation. The
that they form a right-handed coordinate
reliability of this representation depends on the
system. (i = 1, 2, or 3 for blades 1, 2,
generation of accurate mode shapes, which are input
or 3, respectively)
into FAST. You can use a program called Modes (9)
yb,i axis Pointing towards the trailing edge of
to generate these shapes and copy its output to your
blade i and parallel with the chord line
FAST input file. Modes uses essentially the same
at the zero-twist blade station. (i = 1, 2, structural data as FAST. Although the tower and blade
or 3 for blades 1, 2, or 3, respectively) input files include flags to calculate the mode shapes
zb,i axis Pointing along the pitch axis towards internally, we have not implemented this feature in the
the tip of blade i. (i = 1, 2, or 3 for code.
blades 1, 2, or 3, respectively) For the tower, you will need to know the tower-top
When you request output of motions or loads for mass to run Modes. If you do not know the tower-top
various span locations along the blade with the mass, you can obtain it by first running FAST with a
BldGagNd array, a local coordinate system similar to rigid tower and with dummy mode shapes, and then
the standard blade system, but the x-axis and y-axis are reading the summary output file, which includes the
aligned with the local principal axes and the local tower-top mass (see Figure 32 on page 122). FAST
coordinate systems orient themselves with the allows you to specify four different mode shapes for
deflected blade. the tower. The two fore-aft modes are defined
separately from the two side-to-side modes. The mode
shapes take the form of a sixth-order polynomial with
the zeroth and first terms always being zero. This is
because the mode shapes are cantilevered at the base so
they must have zero deflection and slope there. At the
zb,i top of the tower, where the normalized height is 1, the
Wind yb,i deflection must have a normalized value of 1. This
xb,i means the sum of the polynomial coefficients must add
to 1. See Figure 10 for a graphic example of tower
mode shapes.
Tower Height

Figure 9. Blade coordinate system.

Turbine Layout
Figure 14 and Figure 15 show the layout of a First Mode
conventional, downwind, two-bladed turbine and
Second Mode
Figure 16 shows the layout of a conventional, upwind,
three-bladed turbine. Figure 17 through Figure 19
Deflection
show the layout of an upwind turbine with both rotor-
and tail-furling. Figure 20 shows the layout of the
support platform regardless of the above ground Figure 10. Tower mode shapes.
[onshore] or above water [offshore] configuration.
These figures also include some of the important input The blade mode shapes are defined in a way
dimensions. For definitions of these parameters, please similar to that of the tower. For the blades, FAST can
see the Turbine Configuration section of Table 8 on use two flapwise modes and one edgewise mode. The
page 61 for nonfurling turbines, the same section of modes are defined with respect to the local structural
Table 13 on page 82 for furling turbines, and the same twist, that is, the shapes twist with the blade, are three-
section of Table 12on page 81 for the support platform. dimensional, and do not lie within a single plane. In
the case of a twisted blade, the tip will deflect in both
the in-plane and out-of-plane directions due to a pure
FAST User's Guide 10 Last updated August 2005 for version 6.0
 flapwise deflection. The edgewise mode works in a Otherwise, the generator torque will be zero until the
similar fashion. When generating blade modes for a generator speed reaches SpdGenOn.
variable-speed turbine, you should choose a typical You can control when to stop the generator with
rotor speed for the cases you will simulate when the GenTiStp flag in conjunction with TimGenOf. If
generating the mode shapes. Usually, the rotor speed GenTiStp is True, the generator torque will be set to
has little effect on the mode shapes, but it will have a zero after TimGenOf. Otherwise, the generator will
significant effect on the frequency of vibration. Still, stay on until its power reaches zero. Once the
you may want to generate multiple mode shapes for generator is turned off by either method, it will stay off
different rotor speeds to see whether there is a until the end of the simulation. If you are not going to
significant impact on the results. simulate a shutdown or a loss of grid, set GenTiStp to
Hub Centerline True and TimGenOf to a value greater than TMax.
Please see the Simulation Special Events section in the
RNodes5 Controls chapter for more information about this
Pitch
Axis
subject.
Enabling GenDOF will also invoke one of several
generator models. The choice of the model is
HubRad determined by the setting of the GenModel switch or
Node 5 the VSContrl switch. Unless the VSContrl switch is 0,
DRNodes5 GenModel will be ignored. Please see the Variable-
Speed Torque Control section in the Controls chapter
Figure 11. Blade layout. for more information on the variable-speed control
options.
If you set VSContrl to 0 and GenModel to 1,
Drivetrain FAST will use the simple induction generator model.
This model uses just four parameters: rated generator
The drivetrain is modeled as an equivalent shaft slip percentage (SIG_SlPc), the synchronous (zero­
separating the generator from the hub. The shaft can torque) generator speed (SIG_SySp), the rated torque
have a linear torsional spring and a linear torsional (SIG_RtTq), and the pullout ratio (SIG_PORt). This
damper. Use the drivetrain DOF flag, DrTrDOF, to results in the torque/speed curve seen in Figure 12. In
enable this feature. The equation governing the the chart, the rated rotor speed, ΩR, is derived from the
restoring torque of the spring/damper is: synchronous speed and the slip percent:
Tres = DTTorSpr•( RotorPos – GboxPos ) +
DTTorDmp•( RotorSpeed – GboxSpeed ) ΩR = SIG_SySp• ( 1 + 0.01•SIG_SlPc )

The constants DTTorSpr and DTTorDmp are the

–
equivalent torsional stiffness and damping constants SIG_SySp (Ω0)
Generator Torque

for the combined low-speed shaft (LSS), gearbox, and ΩR

high-speed shaft (HSS). All values used in this
equation are cast on the LSS side of the gearbox.
You can simulate losses of the torque being
transmitted through the gearbox by setting the gearbox SIG_RtTq SIG_RtTq•SIG_PO
efficiency, GBoxEff, to some value less than 100%.
+
When generating power, FAST will multiply the LSS
Generator Speed
torque by the efficiency and divide by the gearbox ratio
to determine HSS torque. When motoring, FAST will
multiply the HSS torque by the efficiency and gearbox
ratio to compute the torque on the LSS. Figure 12. Simple-induction-generator
torque/speed curve.
Generator
The generator flag, GenDOF, also governs the The simple model is really too simple to use for a
behavior of the drivetrain, with several options turbine startup. Instead, set VSContrl to 0 and
available. Disabling it will force the generator side of GenModel to 2 to invoke the more-accurate generator
the shaft to turn at a constant speed. model that uses the Thevenin Equivalent Circuit
You can control when to start the generator with equations for a three-phase induction generator. This
the GenTiStr flag in conjunction with either model uses eight input parameters. These values are
SpdGenOn or TimGenOn. If GenTiStr is True, the input in engineering units instead of using the per-unit
generator torque will be zero until TimGenOn. values (normalized by base values) often found in

FAST User's Guide 11 Last updated August 2005 for version 6.0
 generator specification sheets. FAST’s Thevenin­ The flowchart provided in Figure 23 of the
equivalent equations assume a Y-connected, three- Controls chapter explains how the program uses the
phase-generator configuration. If you have a delta- generator model input parameters during runtime, as
connected configuration, you must divide your described above. In this flowchart, GenTq is the
impedances by three and your voltage by 3 to convert instantaneous electrical generator torque, GenPwr is
the values to a Y-connected configuration. Table 8 the instantaneous electrical generator power, and
includes a detailed description of the input parameters GenSpeed is the instantaneous HSS (generator)
and an example torque/speed curve can be seen in speed. The additional logic presented in the flowchart
Figure 13. explains how the program uses the variable-speed
torque and HSS brake control input parameters during
runtime.
Torque

Nacelle Yaw
FAST can model nacelle yaw as a perfect hinge
with no resistance forces by setting YawDOF to True
and the yaw spring constant, YawSpr, and the yaw
damping constant, YawDamp, to zero. You can also
Generator Speed model a free-yaw machine with yaw damping by
setting YawDamp to a nonzero value.
Figure 13. Thevenin-equivalent-induction- You can model the flexibility and damping in the
generator torque/speed curve. yaw drive of a yaw-driven turbine whose commanded
yaw position is held constant, by setting YawDOF to
True, YCMode to 0, and YawSpr and YawDamp to a
Users can create their own generator model by
nonzero value. FAST will use input parameter
modifying the supplied dummy subroutine UserGen()
YawNeut as the neutral yaw position (i.e., constant
available (but commented out) in the UserSubs.f90
yaw command) and NacYaw as the initial yaw angle.
source file. To use your own generator model, it will
In this case, the torque transmitted through the yaw
be necessary to compile the modified file and link it
bearing, YawMom, is:
with the rest of the code. FAST will call UserGen() if
you set VSContrl to 0 and GenModel to 3. The YawMom = YawSpr • ( YawPos – YawNeut )
UserGen() routine linked with the distributed + YawDamp•YawRate
executable version of FAST, which is supplied in
source file UserVSCont_KP.f90, currently calls where YawPos is the instantaneous yaw position.
subroutine UserVSCont(), so that setting GenModel For a fixed-yaw simulation, set YawDOF to False,
to 3 causes FAST to behave as if VSContrl is set to 2. YCMode to 0, TYawManS greater than TMax, and
The routine that calls UserGen() passes the HSS speed NacYaw to the fixed nacelle yaw angle.
and expects the electrical generator torque and You can also actively control the nacelle-yaw
electrical power to be returned. But within routine motion during a simulation. Please see the Nacelle
UserGen(), you have the ability to access the current Yaw Control section in the Controls chapter for
value of any output parameter available from FAST information on active yaw control options.
without changing the number of arguments passed to
the routine. Also, you have the option of switching the Rotor-Furl
generator DOF on-or-off at runtime within UserGen() The rotor-furl DOF allows you to model the
by overriding input GenDOF. Please see the supplied unusual configuration of a bearing that permits the
dummy routines in UserSubs.f90 and the Controls rotor and drivetrain to rotate about the yawing-portion
chapter for further details. of the structure atop the tower. The rotor-furl DOF can
You can simulate generator losses by setting the alternatively be used to model torsional flexibility in
generator efficiency, GenEff, to some value less than the gearbox mounting if you align the rotor-furl axis
100%. When generating power, FAST will multiply with the rotor shaft axis. In order to include rotor-
the mechanical generator power by the efficiency to furling in your model, you must designate the turbine
determine electrical generator power. When motoring, as a furling machine by setting input Furling from the
FAST will multiply the electrical generator power by primary input file to True. Then you must assemble
the efficiency to compute the mechanical generator the furling input file, FurlFile, and use the rotor-furl
power. FAST does not use the generator efficiency for flag, RFrlDOF, to enable this feature.
the Thevenin model since the Thevenin model The angular rotor-furl motion takes place about the
incorporates a more complex expression for the rotor-furl axis defined by inputs RFrlPntxn,
electrical power based on the input circuit resistances. RFrlPntyn, RFrlPntzn, RFrlSkew, and RFrlTilt

FAST User's Guide 12 Last updated August 2005 for version 6.0
 available in FurlFile. Inputs RFrlPntxn, RFrlPntyn, the complications involved in having to define
and RFrlPntzn locate an arbitrary point on the rotor- geometries differently, depending on whether or not a
furl axis relative to the tower-top. Inputs RFrlSkew rotor-furl assembly exists separately from the nacelle,
and RFrlTilt then define the angular orientation of the which depends on whether rotor-furl is present or
rotor-furl axis passing through this point. See Figure absent in the turbine. The developers of FAST also
17 for a schematic. believe that defining geometry relative to the tower-top
The rotor-furl bearing can be an ideal bearing with is the most standard convention. For instance, analysts
no friction by setting RFrlMod to 0; by setting usually think of the rotor shaft offset as the lateral
RFrlMod to 1, it also has a standard model that distance between the rotor shaft axis and the yaw axis
includes a linear spring, linear damper and Coulomb (input Yaw2Shft in FAST), not as a distance relative
damper, as well as up- and down-stop springs, and up- to some coordinate system in the structure furling with
and down-stop dampers. FAST models the stop the rotor.
springs with a linear function of rotor-furl deflection. Since the component geometry of the furling-rotor
The rotor-furl stops start at a specified angle and work assembly is defined relative to the tower-top, this
as a linear spring based on the deflection past the stop geometry naturally changes with the rotor-furl angle.
angles. The rotor-furl dampers are linear functions of In order to avoid having to define different geometries
the furl rate and start at the specified up-stop and for different rotor-furl positions (for example,
down-stop angles. These dampers are bidirectional, variations in the initial rotor-furl angle), FAST expects
resisting motion equally in both directions once past the component geometry of the furling-rotor assembly
the stop angle. to be defined/input at a rotor-furl angle of zero. As
A user-defined rotor-furl spring and damper model such, the initial rotor-furl angle does not affect the
is also available. To use it, set RFrlMod to 2 and specification of any other rotor-furl geometry. Stated
create a subroutine entitled UserRFrl() with the another way, the input geometries for the rotor-furl
parameters RFrlDef, RFrlRate, DirRoot, ZTime, and assembly components define the rotor configuration
RFrlMom: when the rotor-furl angle is zero regardless of initial
rotor-furl position. Users should be clear of this
RFrlDef: Current rotor-furl angular deflection
convention when assembling their furling input file.
in radians (input)
Defining the geometry of the rotor-furl structure
RFrlRate: Current rotor-furl angular rate in
relative to the tower-top instead of in some coordinate
rad/sec (input)
system inherent in the furling-rotor assembly also has
ZTime: Current simulation time in sec (input) some undesirable consequences. The following
DirRoot: Simulation root name including the example will highlight a drawback to the input
full path to the current working convention used in FAST and, at the same time,
director (input) illustrate how the convention works. Consider the case
RFrlMom: Rotor-furl moment in N·m (output) of a small wind turbine company who has settled on
The source file UserSubs.f90 contains a dummy the rotor-furl assembly configuration, including the
UserRFrl() routine; replace it with your own and location of the rotor-furl bearing attachment point on
rebuild FAST. Within routine UserRFrl() you have this assembly, but has yet to determine the best
the option of switching the rotor-furl DOF on-or-off at location of the rotor-furl axis with respect to the
runtime by overriding input RFrlDOF. You can also yawing portion of the structure atop the tower. If the
access the current value of any output parameter design analyst wants to test the rotor-furl response at
available from FAST without changing the number of several different rotor-furl axis locations, this will
arguments passed to the routine. Please see the dummy require him/her to alter not just one input parameter
UserRFrl() routine for a description of how to take (i.e., the rotor furl axis point) but several input
advantage of these incredibly flexible features. parameters collectively. For instance, if he/she wants
Parameter DirRoot may be used to write a record of to alter the lateral (yn) location of the rotor-furl axis,
what is done in UserRFrl() to be stored with the this will require him/her to shift inputs RFrlPntyn,
simulation results. RfrlCMyn, and Yaw2Shft by the same amount since
The geometries of the hub and rotor-furl structure shifting the rotor-furl axis relative to the tower-top also
mass center, which are both components of the furling- shifts the rotor-furl assembly.
rotor assembly, are defined relative to the tower-top as
shown in Figure 18. This definition was chosen in Tail-Furl
order to avoid having to define a coordinate system in The tail-furl DOF allows you to model the unusual
the furling-rotor assembly since such a coordinate configuration of a bearing that permits the tail to rotate
system would most likely have an obscure orientation, about the yawing-portion of the structure atop the
making it difficult for users to input configuration tower. In order to include tail-furling in your model,
information relative to it. This definition also avoids you must designate the turbine as a furling machine by

FAST User's Guide 13 Last updated August 2005 for version 6.0
 setting input Furling from the primary input file to which are all components of the furling-tail assembly,
True. Then you must assemble the furling input file, are defined relative to the tower-top as shown in Figure
FurlFile, and use the tail-furl flag, TFrlDOF, to enable 19. This definition was chosen in order to avoid
this feature. having to define a coordinate system in the furling-tail
The angular tail-furl motion takes place about the assembly since such a coordinate system would most
tail-furl axis defined by inputs TFrlPntxn, TFrlPntyn, likely have an obscure orientation, making it difficult
TFrlPntzn, TFrlSkew, and TFrlTilt available in for users to input configuration information relative to
FurlFile. Inputs TFrlPntxn, TFrlPntyn, and it. This definition also avoids the complications
TFrlPntzn locate an arbitrary point on the tail-furl axis involved in having to define geometries differently,
relative to the tower-top. Inputs TFrlSkew and depending on whether or not a tail-furl assembly exists
TFrlTilt then define the angular orientation of the tail- separately from the nacelle, which depends on whether
furl axis passing through this point. See Figure 17 for tail-furl is present or absent in the turbine.
a schematic. Since the component geometry of the furling-tail
The tail-furl bearing can be an ideal bearing with assembly is defined relative to the tower-top, this
no friction by setting TFrlMod to 0; by setting geometry naturally changes with the tail-furl angle. In
TFrlMod to 1, it also has a standard model that order to avoid having to define different geometries for
includes a linear spring, linear damper and Coulomb different tail-furl positions (for example, variations in
damper, as well as up- and down-stop springs, and up- the initial tail-furl angle), FAST expects the component
and down-stop dampers. FAST models the stop geometry of the furling-tail assembly to be
springs with a linear function of tail-furl deflection. defined/input at a tail-furl angle of zero. As such, the
The tail-furl stops start at a specified angle and work as initial tail-furl angle does not affect the specification of
a linear spring based on the deflection past the stop any other tail-furl geometry. Stated another way, the
angles. The tail-furl dampers are linear functions of input geometries for the tail-furl assembly components
the furl rate and start at the specified up-stop and define the tail configuration when the tail-furl angle is
down-stop angles. These dampers are bidirectional, zero regardless of initial tail-furl position. Users
resisting motion equally in both directions once past should be clear of this convention when assembling
the stop angle. their furling input file. Further clarification on this
A user-defined tail-furl spring and damper model furling geometry convention is provided in the Rotor-
is also available. To use it, set TFrlMod to 2 and Furl section above.
create a subroutine entitled UserTFrl() with the
parameters TFrlDef, TFrlRate, ZTime, DirRoot, and Rotor-Teeter
TFrlMom:
For two-bladed turbines, FAST can model a
TFrlDef: Current tail-furl angular deflection in teetering rotor. To enable the teeter DOF, set
radians (input) TeetDOF to True.
TFrlRate: Current tail-furl angular rate in The teeter bearing can be an ideal bearing with no
rad/sec (input) friction by setting TeetMod to 0; by setting TeetMod
ZTime: Current simulation time in sec (input) to 1, it also has a standard model that includes a spring,
DirRoot: Simulation root name including the stop, and damper. FAST models the spring with a
full path to the current working linear function of teeter deflection. The teeter stop
director (input) starts at a specified angle and works as a linear spring
TFrlMom: Tail-furl moment in N·m (output) based on the deflection past the stop angle. The teeter
damper is a linear function of teeter rate that starts at a
The source file UserSubs.f90 contains a dummy specified angle.
UserTFrl() routine; replace it with your own and A user-defined teeter-spring and damper model is
rebuild FAST. Within routine UserTFrl() you have the also available. To use it, set TeetMod to 2 and create
option of switching the tail-furl DOF on-or-off at a subroutine entitled UserTeet() with the parameters
runtime by overriding input TFrlDOF. You can also TeetDef, TeetRate, ZTime, DirRoot, and TeetMom:
access the current value of any output parameter
available from FAST without changing the number of TeetDef: Current teeter deflection in radians
arguments passed to the routine. Please see the dummy (input)
UserTFrl() routine for a description of how to take TeetRate: Current teeter rate in rad/sec (input)
advantage of these incredibly flexible features. ZTime: Current simulation time in sec (input)
Parameter DirRoot may be used to write a record of DirRoot: Simulation root name including the
what is done in UserTFrl() to be stored with the full path to the current working
simulation results. director (input)
The geometries of the tail boom mass center, tail TeetMom: Teeter moment in N·m (output)
fin mass center, and tail fin aerodynamic surface,

FAST User's Guide 14 Last updated August 2005 for version 6.0
 The source file UserSubs.f90 contains a dummy external load acting on the platform other than loads
UserTeet() routine; replace it with your own and transmitted from the wind turbine. For example, these
rebuild FAST. Within routine UserTeet() you have loads should contain contributions from foundation
the option of switching the rotor-teeter DOF on-or-off stiffness and damping [not floating] or mooring line
at runtime by overriding input TeetDOF. You can restoring and damping [floating], as well as hydrostatic
also access the current value of any output parameter and hydrodynamic contributions [offshore]. The
available from FAST without changing the number of platform loads will be applied on the platform at the
arguments passed to the routine. Please see the dummy instantaneous platform reference position (located by
UserTeet() routine for a description of how to take input PtfmRef).
advantage of these incredibly flexible features. To use this feature, set PtfmLdMod to 1 and create
Parameter DirRoot may be used to write a record of a subroutine entitled UserPtfmLd() with the
what is done in UserTeet() to be stored with the parameters X(6), XD(6), ZTime, DirRoot,
simulation results. PtfmAM(6,6), and PtfmFt(6):
FAST also allows you to specify a δ3 angle for the
X(6): A vector of size 6 containing the 3
teeter hinge. By teetering about an angle that is not
components of the current platform
perpendicular to the blades, you can introduce
translational displacement in meters
flap/pitch coupling to your rotor. This is thought to
and the 3 components of the current
add aerodynamic restoring forces to the blade. Positive
platform rotational displacement in
δ3 will cause the leading edge of the downwind-most radians (input)
blade to feather into the wind. This is illustrated in XD(6): A vector of size 6 containing the 3
Figure 15. See Malcolm’s paper (10) for an analysis of components of the current platform
δ3. translational velocity in m/sec and
the 3 components of the current
Support Platform platform rotational (angular) velocity
You can model the support platform in an onshore in rad/sec (input)
foundation, fixed bottom offshore foundation, or ZTime: Current simulation time in sec (input)
floating offshore configuration by setting the value of DirRoot: Simulation root name including the
input switch PtfmModel from the primary input file to full path to the current working
1, 2, or 3, respectively. Setting PtfmModel to 0 director (input)
disables the platform models—in this case, FAST will PtfmAM(6,6): A symmetric matrix of size 6 X 6
rigidly attach the tower to the inertia frame (ground) containing the current added mass
through a cantilever connection. matrix of the platform with units of
The support platform model properties are kg, kg·m and kg·m2 (output)
designated using the input parameters available in the PtfmFt(6): A vector of size 6 containing 3
platform input file, PtfmFile. In FAST v6.0, all translational and 3 rotational
nonzero PtfmModel options work the same way by components of the current portion of
reading in PtfmFile. In future versions, the format of the platform load, with units of N and
this file will depend on which PtfmModel option is N·m, associated with everything but
selected. the added mass effects (output)
A layout of the configuration properties available As implied by the outputs above, the routine
for the support platform is given in Figure 20. The assumes that the platform loads are transmitted through
platform reference point, located by input parameter a medium like soil [foundation] and/or water
PtfmRef, is the origin in the platform about which the [offshore], so that added mass effects are important.
translational (surge, sway, and heave) and rotational Consequently, the routine assumes that the total
(roll, pitch, and yaw) motions of the support platform platform load can be written as:
are defined. It is also the point at which external
loading is applied to the platform. PtfmF(i) = SUM( -PtfmAM(i,j)•XDD(j), j = 1,2,…,6)
In FAST v6.0, only user-defined platform loading + PtfmFt(i) (for i = 1,2,…,6)
is available. For a value of 0 for PtfmLdMod where,
(available in PtfmFile), there will be no platform
loading and the support reactions normally produced PtfmF(i): he ith component of the total load
T
will be set to zero (causing the wind turbine to fall due applied on the platform; positive in
to gravity if PtfmHvDOF is True). the direction of positive motion of
If you set PtfmLdMod to 1, FAST will call a user the ith DOF of the platform
defined routine named UserPtfmLd() to compute the PtfmAM(i,j): The (i,j) component of the platform
platform loading. The platform loads returned by added mass matrix
UserPtfmLd() should contain contributions from any
FAST User's Guide 15 Last updated August 2005 for version 6.0
 XDD(j): he jth component of the platform
T Tail Fin Aerodynamics
acceleration vector
Your model can optionally include tail fin
PtfmFt(i): The ith component of the portion of
aerodynamic loads. In order to include them, you must
the platform load associated with
designate the turbine as a furling machine by setting
everything but the added mass
input Furling from the primary input file to True and
effects; positive in the direction of
then assemble the furling input file, FurlFile. A furling
positive motion of the ith DOF of the
model may also exclude tail fin aerodynamic loads by
platform
setting TFinMod in FurlFile to 0.
The order of indices in all arrays passed to and You can choose to invoke a simple tail fin
from routine UserPtfmLd() is asfollows: aerodynamics model built into FAST by setting
TFinMod to 1. By accessing information from
1 = Platform surge / xi-component of platform AeroDyn, this model computes the relative velocity of
translation the wind-inflow and its angle of attack relative to the
2 = Platform sway / yi-component of platform tail fin chordline and uses an AeroDyn airfoil table
translation chosen by the user (TFinNFoil) to determine the lift
3 = Platform heave / zi-component of platform and drag forces acting at the tail fin center-of-pressure.
translation Set SubAxInd to False if you want the wind velocity at
4 = Platform roll / xi-component of platform the tail fin to be unobstructed by the rotor wake. Set
rotation SubAxInd to True if you want FAST to decrease (i.e.,
5 = Platform pitch / yi-component of platform subtract) the wind velocity at the tail fin center-of-
rotation pressure by the average rotor induced velocity in the
6 = Platform yaw / zi-component of platform rotor shaft direction.
rotation You also have the option of implementing far
The source file UserSubs.f90 contains a dummy more sophisticated tail fin aerodynamics models by
UserPtfmLd() routine; replace it with your own and supplying your own routines that can easily be linked
rebuild FAST. Within routine UserPtfmLd() you have with the rest of FAST. To do this, set TFinMod to 2
the option of switching the platform DOFs on-or-off at and create a subroutine entitled UserTFin(). The
runtime by overriding inputs PtfmSgDOF, source file UserSubs.f90 contains a dummy
PtfmSwDOF, PtfmHvDOF, PtfmRDOF, UserTFin() routine; replace it with your own and
PtfmPDOF, and PtfmYDOF. You can also access the rebuild FAST. The routine that calls UserTFin()
current value of any output parameter available from passes the tail-furl angle and rate and tail-fin center-of-
FAST without changing the number of arguments pressure location and velocity and expects the angle of
passed to the routine. Please see the dummy attack, lift and drag coefficients, local dynamic
UserPtfmLd() routine for a description of how to take pressure, as well as the normal and tangential forces to
advantage of these incredibly flexible features. be returned. But within routine UserTFin(), you have
Parameter DirRoot may be used to write a record of the ability to access the current value of any output
what is done in UserPtfmLd() to be stored with the parameter available from FAST without changing the
simulation results. number of arguments passed to the routine. Please see
When using UserPtfmLd(), please note that the the supplied dummy routine in UserSubs.f90 for further
hydrostatic restoring contribution to the hydrodynamic details.
force returned by the routine should not contain the
effects of body weight, as is often done in classical
marine hydrodynamics. The effects of body weight are
included within FAST and ADAMS.

Rotor Aerodynamics
The AeroDyn aerodynamic subroutine library
supplies the aerodynamics algorithms for the rotor.
Although we include descriptions of the parameters in
the AeroDyn input file in Table 11, please refer to the
AeroDyn User’s Guide (1) for most of the details on
this package. Input flag CompAero can be used to
disable aerodynamics calculations while debugging a
model.

FAST User's Guide 16 Last updated August 2005 for version 6.0
 Teeter

Wind
NcIMUxn
TipRad
NacCMxn

Nacelle IMU

Nacelle C.M. OverHang

Teeter Pin
ShftTilt

NcIMUzn
Twr2Shft Rotor Axis
NacCMzn
Apex of Cone
of Rotation

Yaw Bearing
C.M. Yaw Axis

Pitch Axis
HubRad PreCone
TowerHt
Apex of Cone
of Rotation UndSling

Teeter Pin

Rotor Axis
HubCM
Hub C.M.

(a)
(b)
Figure 14. Layout of a downwind, two-bladed turbine (a) and a close-up of its hub (b).

FAST User's Guide 17 Last updated August 2005 for version 6.0
 Direction
of
Rotation

Looking
Downwind

Teeter
Axis
+δ3

Leading Edge

Figure 15. Layout of a two-bladed rotor illustrating δ3.

FAST User's Guide 18 Last updated August 2005 for version 6.0
 Precone
(negative as shown)

TipRad Pitch Axis NcIMUxn

Nacelle C.M.
HubCM
(negative as shown) NacCMxn Nacelle IMU
HubRad
Hub C.M. ShftTilt
(negative as shown)

NcIMUzn
OverHang
Rotor Axis (negative as shown)
Twr2Shft NacCMzn

Apex of Cone
of Rotation

Yaw Bearing
C.M.

Yaw Axis

Wind

TowerHt

Figure 16. Layout of a conventional, upwind, three-bladed turbine.

FAST User's Guide 19 Last updated August 2005 for version 6.0
Rotor-furl
axis
Tail-furl
Rotor-furl axis

Arbitrary point on Tail-furl

tail-furl axis
Yaw bearing
C.M.

zn

yn

Wind
xn

Arbitrary point on
rotor-furl axis

Figure 17. Layout of a three-bladed, upwind, furling turbine: furl axes.

FAST User's Guide 20 Last updated August 2005 for version 6.0
 C.M. of structure that
furls with the rotor [not
including rotor]

zn
Rotor
rotation yn

Rotor shaft

axis

xn
Wind

Yaw bearing
C.M.

Figure 18. Layout of a three-bladed, upwind, furling turbine: rotor-furl structure

FAST User's Guide 21 Last updated August 2005 for version 6.0
 Tail boom
C.M.

Tail fin
zn chordline
Tail fin
C.P. Tail fin z

yn Tail fin y

Tail fin x

xn
Yaw bearing
C.M.
Tail fin
Wind C.M.

Figure 19. Layout of a three-bladed, upwind, furling turbine: tail-furl structure.

FAST User's Guide 22 Last updated August 2005 for version 6.0
 Yaw Bearing
C.M.

Yaw Axis

Wind
TowerHt

Tower

Ground Level [onshore] or

Mean Sea Level [offshore]

TwrDraft
Tower Base

Support PtfmCM
Platform

Platform C.M. PtfmRef

Heave
Yaw
Sway

Pitch

Surge
Roll
Platform
Reference Point

Figure 20. Support platform / foundation layout.

FAST User's Guide 23 Last updated August 2005 for version 6.0
FAST User's Guide 24 Last updated August 2005 for version 6.0
 CONTROLS

rotor-collective. Within routine PitchCntrl() you have

General Description the ability to access the current value of any output
parameter available from FAST without changing the
During time-marching analyses, FAST makes it
number of arguments passed to the routine. Please see
possible to control your turbine and model specific
the dummy PitchCntrl() routine for a description of
conditions in many ways. Five basic methods of
how to take advantage of this incredibly flexible
control are available: pitching the blades, controlling
feature.
the generator torque, applying the HSS brake,
There is no pitch actuator model built into FAST
deploying the tip brakes, and yawing the nacelle. The
(though there is in ADAMS datasets generated by
simpler methods of controlling the turbine require
FAST); thus, you must implement your own actuator
nothing more than setting some of the appropriate input
model into routine PitchCntrl() if you want to include
parameters in the Turbine Control section of the
actuator dynamics effects.
primary input file. Methods of control that are more
When using the PitchCntrl() routine, you can
complicated require you to either write your own
delay the time it becomes effective by setting the
routines, compile them, and link them with the rest of
TPCOn parameter to a value greater than zero and
the program or implement your own routines in a
BlPitchi to the initial blade pitch angles. In this case,
Simulink model with which FAST can be interfaced to.
routine PitchCntrl() will not be called until time
For information on linking FAST with your own user-
TPCOn is reached.
defined controllers, please see the Compiling FAST
Setting PCMode to 2 causes FAST to accept pitch
chapter. For information on interfacing FAST with
demands externally from Simulink. In this case,
Simulink, please see the Simulink Interface chapter.
TPCOn must be set to zero since the authority to start
To aid in wind turbine controls design and
and stop the controller is reserved for the Simulink
analysis, linearization routines are also included in
model. You must be using FAST as a DLL interfaced
FAST. Please reference the Linearization chapter for
with Simulink in order to use this feature. Please see
documentation on linearization functionality.
the Simulink Interface chapter for further details.
Although the input file includes a parameter for
Blade Pitch Control partial-span pitch (PSpnElN), we have not yet
One of the most common forms of turbine control implemented this feature in the code.
is full-span blade pitch control. To disable active pitch With or without pitch control enabled, after time
control, set the PCMode switch to 0. TPitManSi, the ith blade will pitch to BlPitchFi using a
Setting PCMode to 1 will cause FAST to call a linear ramp from its current value at TPitManSi until
user-written routine called PitchCntrl() at every time TPitManEi. If pitch control is enabled when PCMode
step. is not 0, the pitch commands determined from inputs
A. Craig Hansen wrote a real pitch-control routine, TPitManSi, TPitManEi, and BlPitchFi override
and we supply that in the file PitchCntrl_ACH.f90. whatever commands come from the pitch controller.
This routine is linked with the executable version of You can use TPitManSi and TPitManEi to simulate a
FAST distributed in the archive. Craig’s routine pitch for startup, shutdown, or runaway fault pitch
controls either power (Region 2) or rotor speed event. By setting one blade different from the other(s),
(Region 3) with collective pitch control. The value of you can simulate a fault condition in which one blade
CntrlRgn, a parameter specified in an input file named unexpectedly pitches or fails to pitch.
Pitch.ipt, which Craig’s routine calls, determines the For a constant-pitch simulation, set PCMode to 0,
type of control used. An example Pitch.ipt file is TPitManSi greater than TMax, and BlPitchi to the
located in FAST’s CertTest folder. The data in this file fixed blade pitch angles.
are for the WindPACT 15A1001 model. Unless you The flowchart provided in Figure 21 explains how
are modeling that turbine, you will need to replace his the program uses the blade pitch control input
Pitch.ipt file with your own. Please contact Craig parameters during runtime, as described above. In this
Hansen for additional information on this pitch flowchart, BlPitchi is the instantaneous blade pitch
controller. angle and BlPitchComi is the instantaneous blade pitch
Additionally, A dummy version of routine angle command of blade i (i = 1, 2, or 3 for blades 1, 2,
PitchCntrl() is available (but commented out) in source or 3, respectively is implied).
file UserSubs.f90. You can write your own routine
here and link it with FAST, though this option requires
the use of a compiler. This user-defined pitch control
routine can act independently for each blade or be

FAST User's Guide 25 Last updated August 2005 for version 6.0
 Initialization: False Time < True
BlPitchCom i = BlPitch i TPCOn
?

Start:
Time = 0 PCMode 0
BlPitch i = BlPitch i (IC) setting?

Finish
False Time < True
TMax User-Written Subroutine:
? 1
BlPitchCom i defined by
PitchCntrl ()

2
FAST/Simulink Interface:
BlPitchCom i defined
Time Integration:
externally from Simulink
Time = Time + DT

False Time < True

Equations of Motion TPitManS i
?

Override Pitch Maneuver:

No Pitch Actuator: BlPitchCom i = F( TPitManS i,
BlPitch i = BlPitchCom i TPitManE i, BlPitchF i, Time,
BlPitch i@Time=TPitManS i)

Figure 21. Flowchart of Blade Pitch Control Runtime Options.

As shown, this simple variable-speed control

Variable-Speed Torque Control model distinguishes between Region 2 (maximum-
power control), Region 3 (constant-torque control), and
Variable-speed generator torque control is another
Region 2½ (linear transition). Region 2½ is a linear
common form of turbine control. To disable active
transition between Regions 2 and 3, with a torque slope
torque control, set the VS_Contrl switch to 0—in this
corresponding to the slope of an equivalent induction
case, FAST will use one of the generator models as
machine. Region 2½ is commonly needed since a wind
described in the Generator section of the Model
turbine does not typically reach rated torque at its rated
Description chapter.
speed using Region 2’s control law [i.e., the optimal
We supply a simple variable-speed control system
gain VS_Rgn2K is typically lower than that which
that uses input parameters VS_RtGnSp, VS_RtTq,
would make VS_RtTq = VS_Rgn2K • (
VS_Rgn2K, and VS_SlPc and results in the
VS_RtGnSp^2 ), since the rated speed, VS_RtGnSp,
torque/speed curve seen in Figure 22. You can enable
is generally limited from optimal in order to limit tip
this control system by setting VSContrl to 1.
speed for noise reasons]. If you want to effectively
eliminate Region 2½ from this model, set VS_RtTq =
VS_RtTq VS_Rgn2K • ( VS_RtGnSp^2 ) and VS_SlPc =
9999.9E-9 (a very small don’t care > 0.0).
Generator Torque

VS_SlPc
Region 3
A setting of 2 for VSContrl will tell FAST to call
a user-written routine named UserVSCont() at every
VS_Rgn2K•(GenSpd^2)
time step after the generator is turned on. Kirk Pierce
wrote an example routine when he worked at NREL
Cut In and this routine, which is contained in source file
Region 2 1/2 VS_RtGnSp UserVSCont_KP.f90, is linked with the executable
Region 2 version of FAST that is distributed in the archive. His
Generator Speed
routine uses a table lookup scheme with a built-in time
delay, which reads data from a file named Spd_Trq.dat,
an example of which is located in FAST’s CertTest
Figure 22. Torque/speed curve for simple folder. The data in this file are for the Small Wind
variable-speed control. Research Turbine (SWRT). Unless you are modeling
that turbine, you will need to replace his Spd_Trq.dat

FAST User's Guide 26 Last updated August 2005 for version 6.0
 file with your own. Please note that Kirk Pierce’s sliding friction. Once full brake torque is reached, the
routine only works when GBRatio is set to 1.0. magnitude of the torque is constant as long as the shaft
Additionally, A dummy version of routine speed is nonzero. When the speed is zero, the torque
UserVSCont() is available (but commented out) in takes on any value to prevent motion of the shaft (the
source file UserSubs.f90. You can write your own HSS can only move again if the external torque
routine here and link it with FAST, though this option exceeds the full braking torque).
requires the use of a compiler. The routine that calls A user-defined HSS brake model is also available.
UserVSCont() passes the HSS speed and expects the Set HSSBRMode to 2 to tell FAST to call the user-
electrical generator torque and electrical power to be written routine named UserHSSBr() at every time step
returned. But within routine UserVSCont(), you have after time THSSBrDp. A dummy version of routine
the ability to access the current value of any output UserHSSBr() is available in source file UserSubs.f90.
parameter available from FAST without changing the You can write your own routine here and link it with
number of arguments passed to the routine. Also, you FAST, though this option requires the use of a
have the option of switching the generator DOF on-or- compiler. The routine that calls UserHSSBr() passes
off at runtime within UserVSCont() by overriding the HSS speed and time and expects the fraction of
input GenDOF. Please see the supplied dummy full braking torque to be returned (0.0 = off – no brake
routine in UserSubs.f90 for further details. torque, 1.0 = full brake torque). As in the simple HSS
Setting VSContrl to 3 causes FAST to accept brake model, the magnitude of the full breaking torque
electrical generator torque and electrical power is specified in input HSSBrTqF. The fraction of full
demands externally from Simulink. In this case, the braking torque may continually vary, permitting you to
authority to start and stop the generator is reserved for continually switch the HSS brake on and off during the
the Simulink model. Thus, GenTiStr and GenTiStp simulation. Input HSSBrDT is ignored when
must be set to True, TimGenOn must be set to zero, HSSBRMode is set to 2.
and TimGenOf must be set greater than TMax. You Within routine UserHSSBr(), you have the ability
must be using FAST as a DLL interfaced with to access the current value of any output parameter
Simulink in order to use this feature. Please see the available from FAST without changing the number of
Simulink Interface chapter for further details. arguments passed to the routine. Also, you have the
The flowchart provided in Figure 23 explains how option of switching the generator DOF on-or-off at
the program uses the variable-speed torque control runtime within UserHSSBr() by overriding input
input parameters during runtime, as described above. GenDOF. Please see the supplied dummy routine in
In this flowchart, GenTq is the instantaneous electrical UserSubs.f90 for further details.
generator torque, GenPwr is the instantaneous The flowchart provided in Figure 23 explains how
electrical generator power, and GenSpeed is the the program uses the HSS brake control input
instantaneous HSS (generator) speed. The additional parameters during runtime, as described above. In this
logic presented in the flowchart explains how the flowchart, HSSBrFrac is the instantaneous fraction of
program uses the generator model and HSS brake full braking torque [limited to values between 0.0 and
control input parameters during runtime. 1.0 (inclusive)] and GenSpeed is the instantaneous
HSS (generator) speed. The additional logic presented
HSS Brake Control in the flowchart explains how the program uses the
variable-speed and generator model control input
By default, the HSS brake is disabled at the
parameters during runtime.
beginning of a run. At time THSSBrDp, the brake
will start to deploy. If you do not want the brake to
deploy during a given run, set THSSBrDp to a value
greater than TMax.
If you set HSSBrMode to 1, FAST will use a
simple HSS brake model in which the brake torque will
ramp linearly from zero at time THSSBrDp to full
brake torque of HSSBrTqF over HSSBrDT seconds.
The HSS brake is based on the Coulomb model of

FAST User's Guide 27 Last updated August 2005 for version 6.0
 Generator is offline
GenTq = 0
GenPwr = 0

Generator is online 0
VSContrl GenModel
setting? setting?

GenSpeed True
False False Time >= True False GenPwr True False Time >= True Simple Induction Generator:
>= 1
TimGenOn <= 0 TimGenO f
SpdGenOn
? GenTq & GenPwr = F(GenSpeed , SIG_SlPc ,
? ? ?
SIG_SySp , SIG_RtTq , SIG_PORt , GenEff )

Thevenin-Equivalent Induction Generator:

False GenTiStr True False GenTiStp True
GenTq & GenPwr = F(GenSpeed ,
Enabled ? Enabled ? 2
TEC_Freq , TEC_NPol , TEC_SRes ,
TEC_Rres , TEC_VLL , TEC_SLR,
TEC_RLR , TEC_MR )
Is the
False True
generator 3 User-Written Subroutine:
offline? GenTq & GenPwr defined
by UserGen()
Start: Has the
Time = 0 False generator True Simple Variable-Speed Control:
1
GenSpeed = ever been
GenTq & GenPwr = F(GenSpeed , VS_RtGnSp ,
GBRatio*RotSpeed (IC) online?
VS_RtTq , VS_Rgn2K , VS_SlPc , GenEff)

2 User-Written Subroutine:
Finish
False Time < True GenTq & GenPwr defined
TMax by UserVSCont()
?

3 FAST/Simulink Interface:
GenTq & GenPwr defined
Time Integration: externally from Simulink
Time = Time + DT
GenSpeed = F(GenSpeed , GenAccel )

False Time >= True HSSBrMode

Equations of Motion: HSSBrFrac = 0 THSSBrDp
Equations of Motion: ? setting?
GenAccel =
GenAccel = 0
F( GenTq+HSSBrTq , etc.)

Simple HSS Brake Control:

1
False GenDOF True HSSBrFrac = F( THSSBrDp ,
Enabled ? HSSBrDT , Time)

User-Written Subroutine:
Apply HSS Brake: 2
HSSBrFrac defined by
HSSBrTq = SIGN( HSSBrFrac *HSSBrTqF , GenSpeed )
UserHSSBr()

Figure 23. Flowchart of Variable-Speed, Generator, & HSS Brake Control Runtime Options.

YawError: Current nacelle-yaw error estimate

Nacelle Yaw Control (positive about the zt-axis) in radians
(input)
You can actively control the nacelle-yaw motion
ZTime: Current simulation time in sec (input)
during a simulation. To disable active yaw control, set
YawPosCom: Commanded nacelle-yaw angular
YCMode to 0.
position (demand yaw angle) in
Setting YCMode to 1 will cause FAST to call a
radians (output)
user-written routine called UserYawCont() at every
YawRateCom:Commanded nacelle-yaw angular
time step. A dummy version of routine
rate (demand yaw rate) in rad/sec
UserYawCont() is available in source file
(output)
UserSubs.f90. You can write your own routine here
and link it with FAST, though this option requires the As indicated, the yaw controller must always
use of a compiler. Though there are a few others, the specify a command (demand) yaw angle,
most important arguments of routine UserYawCont() YawPosCom, and command (demand) yaw rate,
are as follows: YawRateCom. Normally, you should correlate these
commands so that the commanded yaw angle is the
YawPos: Current nacelle-yaw angular position

integral of the commanded yaw rate, or likewise, the

in radians (input)

commanded yaw rate is the derivative of the

YawRate: Current nacelle-yaw angular rate in

commanded yaw angle. FAST will not compute these

rad/sec (input)

correlations for you and does not check to ensure that

WindDir: Current horizontal hub-height wind

they are correlated. In some situations, it is desirable

direction (positive about the zt-axis)

to set one of the commands (either yaw angle or yaw

in radians (input)

rate) to zero depending on the desired transfer function

of FAST's built-in actuator model (see below for a

FAST User's Guide 28 Last updated August 2005 for version 6.0
 discussion of FAST's built-in actuator model). In commanded yaw angle), then FAST's built-in second-
general, the commanded yaw angle and rate should order actuator model will have the following
never be defined independent of each other with both characteristic transfer function, T(s):
commands nonzero.
T(s) = YawDamp•s + YawSpr
Setting YCMode to 2 causes FAST to accept
YawIner•s2 + YawDamp•s + YawSpr
demand yaw angles and rates externally from
= 2•ζ•ωn•s + ωn2
Simulink. You must be using FAST as a DLL
s + 2•ζ•ωn•s + ωn2
2
interfaced with Simulink in order to use this feature.
Please see the Simulink Interface chapter for further where ωn = SQRT( YawSpr/YawIner ) is the yaw
details. actuator natural frequency in rad/sec and ζ =
The yaw controller's effect on the FAST model YawDamp / ( 2•SQRT( YawSpr•YawIner ) ) is the
depends on whether or not the yaw DOF is enabled. If yaw actuator damping ratio in fraction of critical.
the yaw DOF is disabled (YawDOF = False), then the If only the yaw angle is commanded, and
commanded yaw angle and rate from routine YawRateCom is zeroed, then the charecteristic
UserYawCont() or Simulink will be the actual yaw transfer function of FAST's built-in second-order
angle and yaw rate used internally by FAST (in actuator model simplifies to:
general, you should ensure these are correlated). In
this case, any desired actuator effects should be built T(s) = YawSpr
within the yaw control routine. Also in this case, YawIner•s2 + YawDamp•s + YawSpr
FAST will not compute the correlated yaw = ωn2
acceleration, but assume that it is zero. If the s + 2•ζ•ωn•s + ωn2
2

commanded yaw rate is zero while the commanded If only the yaw rate is commanded, and
yaw angle is changing in time, then the yaw controller's YawPosCom is zeroed, then the charecteristic transfer
effect on yaw angle is the identical to routine function of FAST's built-in second-order actuator
PitchCntrl()'s effect on pitch angle (i.e., routine model simplifies to:
PitchCntrl() commands changes in pitch angle with no
associated changes in pitch rate or pitch acceleration). T(s) = YawDamp
For yaw control, this situation should be avoided YawIner•s2 + YawDamp•s + YawSpr
however, since yaw-induced gyroscopic pitching loads = 2•ζ•ωn
on the turbine brought about by the yaw rate may be s2 + 2•ζ•ωn•s + ωn2
significant. Within routine UserYawCont() you have the
If the nacelle yaw DOF is enabled (YawDOF = option of switching the nacelle-yaw DOF on-or-off at
True), then the commanded yaw angle and rate from runtime by overriding input YawDOF. You can also
routine UserYawCont() or Simulink become the access the current value of any output parameter
neutral yaw angle, YawNeut, and neutral yaw rate, available from FAST without changing the number of
YawRateNeut, in FAST's built-in second-order arguments passed to the routine. Please see the dummy
actuator model defined by inputs YawSpr and UserYawCont() routine for a description of how to
YawDamp. In the time domain, the equation for the take advantage of these incredibly flexible features.
yaw DOF is then: When using the UserYawCont() routine, you can
YawIner•YawAccel + YawDamp•YawRate + delay the time it becomes effective by setting the
YawSpr•YawPos = TYCOn parameter to a value greater than zero and
YawDamp•YawRateNeut + NacYaw and YawNeut to the initial nacelle yaw angle
YawSpr•YawNeut + YawTq and neutral yaw position, respectively (the neutral yaw
rate, YawRateNeut, is always assumed zero until
where YawIner is the instantaneous inertia of the active yaw control is enabled). In this case, routine
nacelle, rotor, and tail about the yaw axis and YawTq UserYawCont() will not be called until time TYCOn
is the torque about the yaw axis applied by external is reached. TYCOn must be set to zero when
forces above the yaw bearing, such as wind loading. controlling yaw from Simulink, when YCMode is set
Thus, the torque transmitted through the yaw bearing, to 2, since the authority to start and stop the yaw
YawMom, is: controller is reserved for Simulink.
YawMom = YawSpr• ( YawPos – YawNeut ) + With or without yaw control or the yaw DOF
YawDamp• ( YawRate – YawRateNeut ) enabled, after time TYawManS, the nacelle will yaw
to NacYawF using a linear ramp from its current value
If the commanded yaw angle and rate are at TYawManS until TYawManE. If yaw control is
correlated (so that the commanded yaw angle is the enabled when YCMode is not 0, the yaw commands
integral of the commanded yaw rate, or likewise, the determined from inputs TYawManS, TYawManE,
commanded yaw rate is the derivative of the and NacYawF override whatever commands come

FAST User's Guide 29 Last updated August 2005 for version 6.0
 from the yaw controller. Also, the yaw commands You can also enable passive nacelle-yaw control
determined from inputs TYawManS, TYawManE, during a simulation. Please see the Nacelle Yaw
and NacYawF pass through FAST’s built-in second- section in the Model Description chapter for
order actuator model if the yaw DOF is enabled when information on passive yaw control options.
YawDOF is set to True. You can use TYawManS and The flowchart provided in Figure 24 explains how
TYawManE to simulate a yaw for startup, shutdown, the program uses the nacelle yaw control input
or runaway yaw event. parameters during runtime, as described above.
For a fixed-yaw simulation, set YawDOF to False,
YCMode to 0, TYawManS greater than TMax, and
NacYaw to the fixed nacelle yaw angle.

Initialization:
Initialization:
YawPosCom = YawPos
YawPosCom = YawNeut
False Time < True
YawRateCom = YawRate YawRateCom = 0 TYCOn
?

False YawDOF True

Enabled ? YCMode 0

Start: setting?
Time = 0
YawPos = NacYaw (IC)
YawRate = 0 User-Written Subroutine:
1
YawPosCom & YawRateCom
defined by UserYawCont()
Finish
False Time < True
TMax FAST/Simulink Interface:
? 2
YawPosCom & YawRateCom
defined externally from Simulink

Time Integration:

Time = Time + DT

YawPos = F(YawPos , YawRate )

YawRate = F(YawRate , YawAccel )

Equations of Motion: Equations of Motion:

YawAccel = 0 YawAccel = F(YawMom, etc.)

Yaw Actuator:
No Yaw Actuator: YawNeut = YawPosCom Time < True
False
YawPos = YawPosCom YawRateNeut = YawRateCom TYawManS
YawRate = YawRateCom YawMom = YawSpr*(YawPos-YawNeut ) ?
+ YawDamp *(YawRate -YawRateNeut )

Override Yaw Maneuver:

False YawDOF True YawPosCom & YawRateCom =
Enabled ? F(TYawManS , TYawManE , NacYawF ,
Time, YawPos@Time=TYawManS)

Figure 24. Flowchart of Nacelle Yaw Control Runtime Options.

wind turbine software package. All four routines call

Master Controllers and the Bladed- routine BladedDLLInterface(), which contains a call
Style DLL Interface to the Bladed-style DLL that evaluates as DISCON().
See Figure 25 for a schematic. Routine
In the Source folder of the FAST archive, we BladedDLLInterface() USEs a MODULE named
distribute a source file named BladedDLLInterface.f90. BladedDLLParameters(), which stores values of
This source file contains example PitchCntrl(), PARAMETER constants used in the interface.
UserHSSBr(), UserVSCont(), and UserYawCont() Routines BladedDLLInterface() and
routines that may be used to interface FAST with a BladedDLLParameters() are also contained in source
master controller implemented as a dynamic-link- file BladedDLLInterface.f90.
library (DLL) in the style of Garrad Hassan's Bladed

FAST User's Guide 30 Last updated August 2005 for version 6.0
 Control Measurements
HSS Brake
Control Interface to
Fraction of Full Master
UserHSSBr()
Brake Torque Controller
Variable-Speed BladedDLL
Aeroelastic
Control Interface()
Model Gen. Torque
UserVSCont()
(FAST, and Power IF (Time-LastTime
Nacelle Yaw >=DT) THEN
ADAMS)
Control LastTime=Time
Nacelle Yaw
UserYawCont()
Angle and Rate CALL DISCON
Blade Pitch ENDIF
Control
Blade Pitch
PitchCntrl()
Angles
Master Controller
Override
(Bladed-Style DLL)
Maneuvers
DISCON()

Figure 25. Interface to a Bladed-Style Master Controller DLL.

Source file BladedDLLInterface.f90 is useful if This interface is valid for DLLs of the style
you have a DLL controller created for a Bladed model specified in Appendices A and B of the Bladed User
and you want to use the same controller for your FAST Manual of Bladed version 3.6 (2). The documentation
model. This source file is also a useful template if you provided there is not repeated here. If you are running
prefer to control pitch, HSS brake torque, electrical FAST using a master controller DLL developed in
generator torque, and/or nacelle yaw with a single Bladed, please be aware of the differences indicated in
master controller, regardless of whether or not you use Table 4 between this interface and Bladed's interface.
the Bladed code and regardless of whether or not you
want to work with DLLs. As it is developed, the same
source file can be used to interface both FAST and Table 3. Parameter Settings to be Used
ADAMS to Bladed-style master controller DLLs. With Bladed-Style Master Controller DLLs.
In order to use these routines, you must first set the Parameter Setting Reason
values of the PARAMETERs contained in MODULE YCMode 1 Tells FAST to use routine
BladedDLLParameters() as required by your model. UserYawCont() for active
These PARAMETERs are model-specific inputs yaw control
available in the Bladed code, which are not available TYCOn 0.0 Tells FAST to start active
inputs in FAST, and are passed to the Bladed DLL in yaw control at the
this interface. You must then comment-out the dummy beginning of the simulation
placeholder versions of routines PitchCntrl(), PCMode 1 Tells FAST to use routine
UserHSSBr(), UserVSCont(), and UserYawCont() PitchCntrl() for active pitch
contained in source file UserSubs.f90 and recompile control
FAST with the addition of source file TPCOn 0.0 Tells FAST to start active
BladedDLLInterface.f90—see the Compiling FAST pitch control at the
chapter for more information. The executable version beginning of the simulation
of FAST that is distributed with the archive is not VSContrl 2 Tells FAST to use routine
linked with the routines contained within source file UserVSCont() for active
BladedDLLInterface.f90. After you have compiled variable-speed torque
FAST with the routines in BladedDLLInterface.f90, control
you must modify several input parameters from the GenTiStr True Tells FAST to start torque
primary input file in order to use the Bladed-style control based on time
controller. These parameters and the necessary settings TimGenOn
are listed in Table 3 (these conditions are not tested by
GenTiStp True Tells FAST to stop torque
these example routines). control based on time
TimGenOf

FAST User's Guide 31 Last updated August 2005 for version 6.0
 TimGenOn 0.0 Tells FAST to start torque of Record 62
control at the beginning of 64 The maximum number of characters which
the simulation can be returned in "OUTNAME" is always
TimGenOf >TMax Tells FAST not to stop set to 0 in accordance with the
controlling torque specification of Record 62
throughout the simulation 65 The number of variables returned for
HSSBrMode 2 Tells FAST to use routine logging returned by the DLL must be set to
UserHSSBr() for control 0 by the DLL indicating none in
of the HSS brake accordance with the specification of
THSSBrDp 0.0 Tells FAST to start HSS Record 62; the program will abort
brake torque control at the otherwise
beginning of the simulation 72 The generator start-up resistance is always
ignored
Table 4. Differences Between FAST’s and
79 The request for loads is ignored; instead,
Bladed’s Interface to Master Controller DLLs.
the blade, hub, and yaw bearing loads are
always passed to the DLL as if Record 79
Record Difference was set to 4
1 The status flag is not set to -1 for the final 80 The variable-slip current demand toggle
call at the end of the simulation switch is always ignored; instead, the
10 The pitch actuator type is always set to 0 generator torque demand from Record 47
by FAST, indicating pitch position is always used
actuator; as such, the returned value of 81 The variable-slip current demand is always
Record 46, demanded pitch rate ignored in accordance with the handling of
(Collective pitch), is always ignored Record 80
29 The yaw control type is always set to 0 by
FAST indicating yaw rate control; as such, We distribute a dummy placeholder version of the
the returned value of Record 41, demanded source file DISCON.f90 in the FAST archive. You
yaw actuator torque, is always ignored may use DISCON.f90 as a template for creating your
35 The generator contactor status, is own master controller DLL if you do not already have
initialized to 1 by FAST indicating main one created. Please refer to appendices A and B of the
(high speed) or variable speed generator; Bladed User Manual for further information.
the generator can be turned off in the DLL
by setting Record 35 to 0 or by setting
Record 47 to 0.0; if the DLL redefines Tip Brakes
Record 35 to something other than 0 or 1 The tip brakes can be controlled in two ways. You
(such as 2 = low speed generator), the can set a time at which each brake is deployed
program will abort (TTpBrDpi), or you can set a rotor speed at which each
41 The demanded yaw actuator torque is brake is deployed (TBDepISpi). The tip brakes for
always ignored in accordance with the different blades are controlled separately. If your
specification of Record 29 turbine does not have tip brakes, set the tip-brake drag
46 The demanded pitch rate (Collective pitch) terms, TBDrConN and TBDrConD, to zero. You
is always ignored in accordance with the should also set the deployment times and speeds to
specification of Record 10 values greater than those that are likely to occur during
55 The pitch override returned by the DLL the run so that FAST won’t waste time on unused
must be set to 0 indicating no override calculations.
(i.e., pitch demands come for the DLL); If you do use tip brakes, you will need to provide
the program will abort otherwise realistic values for the drag terms and the amount of
56 The torque override returned by the DLL time it takes to deploy them once they’ve started to
must be set to indicating no override (i.e., deploy (TpBrDT). The brakes take this long to deploy
torque demands come for the DLL); the for time- and speed-initiated deployments. Once the
program will abort otherwise brakes deploy, they remain so until the end of the run.
62 The maximum number of values which The interpolated drag term during the deployment
can be returned for logging is always set to follows an “S” curve from TBDrConN to TBDrConD.
0 by FAST indicating none FAST does not orient the tip-brake forces with
63 The record number for start of logging blade pitch. The tangential velocity of the blade tip,
output is always set to 0 (a don't care) by not taking into account wind motion, is used to
FAST in accordance with the specification calculate the dynamic pressure. Because of these

FAST User's Guide 32 Last updated August 2005 for version 6.0
 approximations, you should adjust TBDrConD so that (HSSBrMode = 1), set THSSBrDp to a short time
the rotor decelerates as expected. after TimGenOf and HSSBrDt to the amount of time
it takes to fully apply the brake (the brake torque will
Simulating Special Events ramp from zero to full in a linear fashion). When using
a user-defined HSS brake model (HSSBrMode = 2),
There are many special events that can be modeled
you may specify a nonlinear ramp. Input HSSBrTqF
with FAST. Although we will illustrate many of them
specifies the maximum full brake torque in both cases.
in this section, we cannot document all of them. We
hope that these examples will be sufficient so that you HSS Brake Shutdown with Generator Brake
can figure out how to model other cases. FAST can model a shutdown in which the
Turbine Startup generator acts as a dynamic brake until the rotor slows
enough that the HSS brake can stop the rotor, but for
There are several ways to start a turbine. One
now you must write the logic yourself by supplying a
common way is to pitch the blades from feather to the
UserGen() routine and linking it with the rest of the
run position and let the wind accelerate the rotor until a
code. We hope to find an easier way by adding a few
certain speed is reached. To model this case, set
input parameters.
PCMode to 0, GenTiStr to False, and SpdGenOn to
an appropriate value. For each blade, set TPitManS to Normal Tip Brake Shutdown
the time you want to start the maneuver, TPitManE to To model a normal shutdown, in which the tip
the end time for the maneuver, BlPitch to the feather brakes decelerate the rotor, set TTpBrDp to
position, and BlPitchF to the run position. appropriate values for each blade. As with the pitch-
You can also start a stall-regulated turbine by to-feather shutdown, set GenTiStp to False so that the
motoring. To perform a motor start, set PCMode to 0, generator will disengage when power drops to zero.
GenTiStr to True, and TimGenOn to an appropriate
value. You should also use either variable-speed Tip Brake Shutdown after Loss of Grid
control or the Thevenin-equivalent-induction-generator This case is similar to the previous case, but you’ll
model. Do not use the simple-induction-generator use the TBDepISp array to make FAST use rotor
model, because it does not have a realistic startup speed to deploy the tip brakes. You can also model the
torque. special case in which one brake deploys at a higher
speed than the other(s) or not at all by setting its
Normal Pitch-to-Feather Shutdown
deployment initiation speed to a higher speed.
To simulate this case, you’ll need to set the pitch
maneuver start and stop times (TPitManS and Accidental Deployment of a Tip Brake
TPitManE) and the initial and final pitch settings You can easily model the accidental deployment
(BlPitch and BlPitchF). Set TiGenOn to zero and, if of a tip brake. For one blade, set TTpBrDp to a value
you want to use the generator as a brake, set GenTiStp less than TMax. For the other blade(s), set TTpBrDp
to False. This will disengage the generator when the to value(s) greater than TMax. For all blades, set
turbine slows enough to drop the power to zero. TBDepISp to large numbers so that the brakes will
never deploy because of rotor speed. You will also
Shutdown Where One Blade Fails to Feather
need to set TBDrConN, TBDrConD, and TpBrDT to
This case is the same as the previous example, but appropriate values.
either set the times (TPitManS and TPitManE) for one
blade to values greater than TMax or set the final pitch Idling Turbine
value, BlPitchF, to the initial pitch value (BlPitch) or You can simulate an idling turbine by enabling the
some other value that is different from the other generator DOF (GenDOF) and setting GenTiStr to
blade(s). True and TimGenOn to a value greater than TMax to
ensure that the generator never goes online. You will
One Blade Feathers Accidentally
also want to initialize the rotor speed (RotSpeed) to a
This case is the same as the previous example, but small or zero value, set the generator inertia (GenIner)
either set the times (TPitManS and TPitManE) for the to a non-zero value, and set the gearbox efficiency
non-feathering blade(s) to values greater than TMax, or (GboxEff) to a value less than 100%. The torque
set the final pitch value(s), BlPitchF, to the initial pitch passing through the non-perfect gearbox to the
value(s) (BlPitch). generator-rotor inertia will tend to resist acceleration of
HSS Brake Shutdown after Loss of Grid the turbine rotor. You can also add some speed-
independent drag to the drivetrain by applying a light
To model an emergency shutdown where the HSS
brake load. To do so, tell the brake to always be on by
brake stops the rotor, set GenTiStp to True and
setting THSSBrDp and HSSBrDt to 0, and then set
TimGenOf to the time you want the grid to fail. When
the brake torque (HSSBrTqF) to a small value.
using the simple built-in HSS brake model

FAST User's Guide 33 Last updated August 2005 for version 6.0
 Parked Turbine drivetrain to ring. Another possibility is to enable
One of the standard IEC test cases is to model the GenDOF, but set GenTiStr and TimGenOn so the
turbine in high winds when the turbine is parked. If generator never starts. Set THSSBrDp and HSSBrDT
your turbine uses full-span pitch, set the values of to zero so the HSS brake is always on. You will need
BlPitch and BlPitchF to the feathered setting. Also set to set the HSSBrTqF to a realistic value.
TpitManS and TpitManE to 0 so the blades are For potential failure modes, you can model the
feathered during the entire simulation. case in which the brake torque is insufficient to hold
If you park your turbine by applying a HSS brake, the turbine. The generator DOF must be enabled for
you can model this condition by disabling the generator this case. You can examine another potential failure by
DOF (GenDOF) with RotSpeed set to zero and setting one of the blades so that it pitches to a non-
enabling the drivetrain DOF (DrTrDOF) to allow the feathered value at some time during the run.

FAST User's Guide 34 Last updated August 2005 for version 6.0
 SIMULINK INTERFACE
(http://www.mathworks.com/). A working knowledge
General Description of Simulink model development is also essential.
The FAST archive contains several files that are
Simulink is a popular simulation tool for controls
pertinent to FAST’s interface with Simulink as
design that is distributed by The Mathworks, Inc. in
described below:
conjunction with MATLAB. Simulink has the ability
to incorporate custom Fortran routines in a block FAST_SFunc.dll The FAST S-Function
called an S-Function. The FAST subroutines have compiled as a dynamic-link-
been linked with a MATLAB standard gateway library (DLL). This DLL
subroutine in order to use the FAST equations of contains the structural
motion in an S-Function that can be incorporated in a dynamic routines from FAST,
Simulink model. This introduces tremendous the aerodynamic routines from
flexibility in wind turbine controls implementation AeroDyn, and interfaces to
during simulation. Generator torque control, nacelle Simulink.
yaw control, and pitch control modules can be designed Simsetup.m This MATLAB script file
in the Simulink environment and simulated while prompts the user for the FAST
making use of the complete nonlinear aeroelastic wind primary input file name and
turbine equations of motion available in FAST. calls Read_FAST_Input.m,
The wind turbine block, as shown in Figure 26, which initializes model
contains the S-Function block with the FAST equations variables. It must be called
of motion. It also contains blocks that integrate the from the MATLAB workspace
DOF accelerations to get velocities and displacements. before you run a Simulink
Thus the equations of motion are formulated in the model with the FAST S-
FAST S-function but solved using one of the Simulink Function.
solvers. Read_FAST_Input.m This MATLAB script file is
called by Simsetup.m and
reads the FAST input files to
initialize parameters in a
Simulink model. Users should
not change this file.
OpenLoop.mdl An example Simulink model
containing the FAST S-
Function block, blocks that
integrate the DOFs, and
constant open loop control
input blocks.
Test01_SIG.mdl An example Simulink model
Figure 26. FAST Wind Turbine Block. containing the FAST S-
Function block, blocks that
The interface between FAST and Simulink is very integrate the DOFs, and the
similar to the interface developed for the Symbolic simple induction generator
Dynamics (SymDyn) code, which is a controls-oriented model for FAST certification
HAWT analysis tool developed by researchers at test #01 implemented within
NREL (11). The structural model of FAST, however, Simulink.
is of higher fidelity than that of SymDyn. To run a FAST model in Simulink, first transfer
files Simsetup.m and OpenLoop.mdl from the
Getting Started SimulinkSamples folder to the directory containing the
In order to build a Simulink model that uses the primary input file of a FAST model that you want to
FAST wind turbine dynamics in an S-Function, you use. If you want to use one of the certification test files
must purchase the commercial MATLAB software from the CertTest folder, you may have to make a few
with the additional Simulink package. MATLAB is minor changes to some of the input parameters in order
available from The Mathworks, Inc. to use the FAST S-Function in Simulink (refer to the
next section for the reason).

FAST User's Guide 35 Last updated August 2005 for version 6.0
 Now open a MATLAB command window. In Specific Input File Options for the
MATLAB, add the folder where files FAST_SFunc.dll FAST S-Function
and Read_FAST_Input.m are stored to the MATLAB
path by choosing “Set Path…” from the File menu, As implied above, FAST input files must be
clicking “Add Folder…”, selecting the folder, and created in order to use the FAST S-Function. Some
pressing Save and Close. Next, change the current input parameters directly control the execution of the
working directory in MATLAB to the directory in Simulink model; others cause the FAST S-Function to
which the FAST model files (including files abort; most behave exactly as they do in the executable
Simsetup.m and OpenLoop.mdl) are stored. Type version of FAST.
“Simsetup” into the MATLAB command prompt. The FAST input variables TMax and DT may be used
script file will prompt you for the name of the primary to control the Simulink simulation by entering them in
input file of FAST; type in the root name with the Stop time and Fixed step size boxes, which are
extension. Next, open the example Simulink model, contained in the “Simulation parameters…” window
OpenLoop.mdl, by choosing Open… from the File available from the Simulation menu of the Simulink
menu. The Simulink model should appear as in Figure model. These are only available if a fixed step solver
27 below (the green block in Figure 27 contains the is selected. The fixed step solver, ode4, most closely
FAST wind turbine block shown in Figure 26). emulates the solver used by FAST. These settings
Finally, click on the Play (►) button in the Simulink have already been specified in OpenLoop.mdl but may
window to run the simulation. be changed by you depending on your preference.
Under the Turbine Control section of FAST’s
primary input file, you have the option of determining
whether blade pitch, nacelle yaw, and/or variable-speed
torque is controlled by the Simulink model or by using
one of FAST’s intrinsic controllers. To control blade
pitch commands from Simulink, set PCMode to 2. In
this case, TPCOn must be set to zero since the
authority to start and stop the controller is reserved for
the Simulink model. If PCMode is either 0 or 1, the
model will behave exactly as a standalone FAST model
and the pitch commands from Simulink will be
ignored. Similarly, to control nacelle yaw angle and
rate commands from Simulink, set YCMode to 2. In
Figure 27. Simulink Model OpenLoop.mdl. this case, TYCOn must be set to zero since the
authority to start and stop the controller is reserved for
The FAST S-Function will generate the same the Simulink model. To model a variable-speed torque
ASCII output files as would be generated during a controller in Simulink, VSContrl must be set to 3. In
normal FAST simulation. These output files use the this case, the authority to start and stop the generator is
root name of the primary input file and append _SFunc reserved for the Simulink model. Thus, GenTiStr and
to the name. For example, if the primary input file GenTiStp must be set to True, TimGenOn must be set
were named fast.fst, the main output file from the to zero, and TimGenOf must be set greater than
FAST S-Function will be named fast_SFunc.out TMax.
whereas the the FAST executable would generate The override pitch and yaw maneuvers specified in
fast.out. The output for the certification test files FAST’s primary input file will supercede any pitch and
should agree quite well with the corresponding output yaw commands that originate in Simulink regardless of
from FAST. There will be slight differences due to the the setting of PCMode and YCMode. You may use
different solvers and precisions employed by each these to force faults in you pitch and yaw controllers.
program. Some features of FAST are not available within
If for any reason an error occurs during a Simulink. Thus, when running FAST within Simulink,
simulation, the FAST S-Function will display the error the FAST S-Function will abort if any of these features
message in a Simulation Diagnostics pop-up box and are selected. The ADAMS preprocessor and the
abort. Warning messages routinely written to the linearization capability are not available in the FAST
command-line window by the FAST executable are not S-Function; thus, ADAMSPrep and AnalMode must
echoed to a pop-up box nor are they echoed to the be set to 1. The high-speed shaft brake option is not
MATLAB workspace by the FAST S-Function. We available in the FAST S-Function so THSSBrDp must
hope to add this capability in the future. be set greater than TMax. Finally, Simulink can only
use the initial conditions for revolute DOFs, including
Azimuth, RotSpeed, TeetDefl, NacYaw, RotFurl,
and TailFurl. Specifying nonzero values for IPDefl,

FAST User's Guide 36 Last updated August 2005 for version 6.0
 Customizing the Simulink Model file. You can access specific channels from the
OutData array by using the OutList cell array. For
This section provides a few more details on the
example, to obtain the rotor speed (assuming rotor
FAST interface to Simulink. This should provide you
speed was specified in OutList) at the 3rd time step
with enough guidance so that you can modify the
(3rd row) in the MATLAB workspace, type
example Simulink models to include your own torque,
“OutData(3,strmatch(‘RotSpeed’,OutList))” in the
yaw, and/or pitch controllers.
MATLAB command prompt. Using this technique,
The wind turbine block requires three inputs and
you don’t need to remember the specific order you
has one output as shown in Figure 26 and Figure 27.
listed the output channel names in OutList.
Electrical generator torque and electric power demands
You can modify the Simsetup.m script file to
must be supplied in the first input, nacelle yaw position
initialize variables for any additions you make to the
and rate demands must be supplied in the second input,
Simulink model for torque, yaw, and pitch control. If
and blade pitch demand angles for all blades must be
you modify Simsetup.m, remember to use “clear all” or
supplied in the third input. Data must be provided for
“clear functions” to clear the MATLAB memory
all inputs in order for the Simulink model to run. For
before repeating a simulation. As provided, “clear all”
instance, if you want to use the FAST simple induction
is the first command in Simsetup.m. The character
generator model (by setting VSContrl to 0 and
array named input_fast, which is defined in
GenModel to 1) rather than developing a torque
Simsetup.m, must contain the name of the primary
controller in your Simulink model, you may use
input file. Also, the script that reads the FAST input
Constant blocks to supply dummy electrical generator
file for model initialization, Read_FAST_Input.m, must
torque and electrical power demands to the FAST wind
be called before running a simulation and after
turbine block—these will not be used by the wind
input_fast has been defined. Other than these
turbine block, but they must be present. This is
requirements, you are free to perform any controller
demonstrated in OpenLoop.mdl. Similarly, values
design or initialization steps in Simsetup.m or your own
must be supplied for the yaw position and rate, and the
script before performing a simulation in Simulink.
blade pitch angles. The blade pitch angles are stored in
As an example of a Simulink model more
a vector sized according to the number of blades.
advanced than OpenLoop.mdl, we distribute a Simulink
In addition to the data available in the primary
model named Test01_SIG.mdl in the FAST archive. In
output file generated by the S-Function, the wind
this example, the simple induction generator (available
turbine block will also output a variable array named
when VSContrl is set to 0 and GenModel is set to 1)
OutData that contains the output data selected in the
is implemented in Simulink rather than FAST for
FAST input file through input OutList. OutData
certification test #01 (by setting VSContrl to 2). To
contains output data at every model time step (whereas
run this example, follow the directions in the comments
the primary output file uses the value of DecFact) and
at the end of Simsetup.m. The output should be very
is available in the Simulink environment at runtime for
similar to that of certification test #01 run with the
feeding back control measurements. Thus, the control
FAST executable. There will be slight differences due
measurement channels your Simulink controller needs
to the different solvers and precisions employed by
must be specified in OutList. OutData will also be
each program.
available in the MATLAB workspace for
postprocessing.
The output data names listed in OutList are also
available in MATLAB and Simulink workspaces in a
variable cell array named, appropriately, OutList. This
variable is created by the Read_FAST_Input.m script

FAST User's Guide 37 Last updated August 2005 for version 6.0
FAST User's Guide 38 Last updated August 2005 for version 6.0
 LINEARIZATION

following conditions are not met. First, active yaw and

General Description pitch control must be disabled by setting YCMode and
PCMode to 0. FAST can’t be linearized during a
FAST has the capability of extracting linearized
startup or shutdown event, thus GenTiStr and
representations of the complete nonlinear aeroelastic
GenTiStp must both be set True, TimGenOn must be
wind turbine modeled in the code. This analysis
set to 0.0, and TimGenOf must be set greater than
capability is useful for developing state matrices of a
TMax during a linearization analysis. Inputs
wind turbine “plant” to aid in controls design and
THSSBrDp, TiDynBrk, TTpBrDpi, TYawManS, and
analysis. It is also useful for determining the full
TPitManSi must also be set greater than TMax, and
system modes of an operating or stationary HAWT
TBDepISpi must be set much greater than RotSpeed.
through the use of a simple eigenanalysis. A FAST
In AeroDyn, dynamic stall and dynamic inflow must be
linearization analysis is invoked by setting input
disabled if CompAero is True; thus StallMod should
parameter AnalMode in the primary input file to 2.
be set to “STEADY” and InfModel to
The linearization routines follow a procedure
“EQUIIL”ibrium. Also, you must use a hub-height
similar to that used by the Symbolic Dynamics
wind data file (input WindFile) that does not vary with
(SymDyn) code, which is a controls-oriented HAWT
time. At least one DOF must be enabled during a
analysis tool developed by researchers at NREL (11).
linearization analysis because it is useless to have a
The structural model of FAST, however, is of higher
“plant” model with zero states. Finally, CompNoise
fidelity than that of SymDyn.
must be disabled during a linearization analysis.
The linearization process consists of two steps: (1)
Inputs CalcStdy, TrimCase, DispTol, and
computing a periodic steady state operating point
VelTol, which are available in the linearization control-
condition for the DOFs and (2) numerically linearizing
input-file of FAST (identified by parameter LinFile in
the FAST model about this operating point to form
FAST’s primary input file), are the parameters used to
periodic state matrices. The output state matrices can
manage the steady state solution computation in FAST.
then be azimuth-averaged for nonperiodic, or time-
Input parameter CalcStdy is a flag used to
invariant controls development.
indicate whether a periodic steady state solution is
computed before linearizing the model. To disable the
Periodic Steady State Solution steady state solution computation, set CalcStdy to
The first step in the linearization process is False. In this case, the operating point is prescribed by
determining an operating point to linearize the model the values of the initial conditions specified in FAST’s
about. An operating point is a set of values of the primary input file. That is, when CalcStdy is False,
system DOF displacements, DOF velocities, DOF the operating point is set to the condition in which all
accelerations, control inputs, and wind inputs that displacements, velocities, and accelerations are zero,
characterize a steady condition of the wind turbine. except those specified with nonzero initial conditions
For a wind turbine operating in steady winds, this (for instance, the azimuth DOF will increment at a
operating point is periodic—that is, the operating point constant rate if and when the rotor is spinning).
values depend on the rotor azimuth orientation. This Setting CalcStdy to True causes FAST to
periodicity is driven by aerodynamic loads, which compute a steady state solution before linearizing the
depend on the rotor azimuth position in the presence of model. During a steady state solution computation,
prescribed shaft tilt, wind shear, yaw error, or tower FAST integrates the nonlinear equations of motion in
shadow. Gravitational loads also drive the periodic time until the solution “converges”. “Convergence” is
behavior when there is a prescribed shaft tilt or determined as follows. At each iteration, or one period
appreciable deflection of the tower due to thrust of revolution of the rotor, a 2-norm of the differences
loading. It is important to determine an accurate between conditions at the beginning and end of the
operating point because the linearized model is only iteration is computed. A 2-norm is computed for both
accurate for values of the DOFs and inputs that are the angular displacement vector differences and
close to the operating point values. angular velocity vector differences. Input parameters
To compute a steady state solution, the program DispTol and VelTol are used as convergence
inputs must necessarily produce a time invariant model tolerances for the displacement 2-norm and velocity 2­
(other than azimuth dependence). To insure this time norm, respectively. The smaller the values of DispTol
invariant condition, FAST performs a number of and VelTol, the tighter the tolerances. Once both of
checks on some of the input parameters before running the computed 2-norms become smaller than or equal to
a linearization analysis. FAST will abort without the input convergence tolerances, the solution is
computing the linearized state matrices if any of the considered to have “converged”. If the solution has not

FAST User's Guide 39 Last updated August 2005 for version 6.0
 converged by the time TMax is reached, the iteration VS_Rgn2K, and VS_SlPc to 9999.9E-9 (very small
stops and FAST aborts without computing the don’t cares > 0.0).
linearized state matrices. See Figure 28 for a The third trim option is for the nacelle yaw control
schematic. input. To trim nacelle yaw, set CalcStdy to True,
The calculation of an operating point depends on GenDOF to True, RotSpeed to the desired azimuth-
whether the rotor is spinning or stationary, whether the averaged rotor speed (nonzero), and TrimCase to 1.
turbine is variable or constant speed, and whether the Nacelle yaw can be trimmed with or without the yaw
operating point is in Region 2 (below rated wind speed) DOF enabled. With yaw DOF enabled (YawDOF =
or Region 3 (above rated wind speed). Again, see True), setting TrimCase to 1 causes FAST to trim the
Figure 28 for a schematic. neutral yaw angle, YawNeut, which passes through
To linearize a stationary HAWT, it is best to use FAST’s built-in, second-order actuator model, while
the static equilibrium point as the steady state operating maintaining constant rotor collective blade pitch
point. In this case, the steady state operating point is (indicated by inputs BlPitchi from the primary input
not periodic because the rotor is not spinning. To file), to reach the desired azimuth-averaged rotor speed
obtain the static equilibrium condition, set CalcStdy to condition. In this case, the yaw actuator, which is
True, GenDOF to False, and RotSpeed to zero. described in the Nacelle Yaw Control section of the
FAST will then integrate in time until the convergence Controls chapter, will be inherent in the output
tolerance conditions are met. This operation can be linearized model. With yaw DOF disabled (YawDOF
performed with or without aerodynamic thrust effects = False), setting TrimCase to 1 causes FAST to trim
as indicated by input flag CompAero. the actual nacelle yaw angle, while maintaining
For variable speed wind turbines, one often wants constant rotor collective blade pitch (indicated by
to determine the periodic operating point in inputs BlPitchi from the primary input file), to reach
conjunction with a trim analysis. A trim analysis is the the desired azimuth-averaged rotor speed condition. In
process of trimming a control input in order to reach a this case, the yaw actuator will be absent from the
desired azimuth-averaged rotor speed while holding all output linearized model.
other inputs constant. The FAST linearization For constant speed machines, set GenDOF to
functionality allows for three forms of trim as specified False. FAST will then ignore input TrimCase, and the
by input switch TrimCase. For all three cases, the trim analysis will be bypassed during the computation
desired azimuth-averaged rotor speed is prescribed by of the periodic steady state operating condition.
input parameter RotSpeed from the primary input file With or without trim, if a steady state solution has
(which is also used as the initial rotor speed). Input trouble converging, try increasing the simulation
parameter TrimCase is ignored when either CalcStdy runtime, TMax; increasing system-damping values; or
or GenDOF is False. increasing the convergence tolerances, DispTol and
For variable speed turbines in Region 2, set VelTol. Some steady state solutions may take 300
CalcStdy to True, GenDOF to True, RotSpeed to the seconds or more of simulation time to converge,
desired azimuth-averaged rotor speed (nonzero), and depending on the nature of the wind turbine model,
TrimCase to 2. Setting TrimCase to 2 causes FAST system-damping values, and convergence tolerances
to trim electrical generator torque, while maintaining used. When trimming, also make sure that the
constant rotor collective blade pitch (indicated by condition you are trying to trim to is “reasonable”. For
inputs BlPitchi from the primary input file), to reach example, during Region 3 trim (TrimCase = 3), it may
the desired azimuth-averaged rotor speed condition. be impossible to find a rotor collective blade pitch
For variable speed turbines in Region 3, set angle at a given rotor speed and wind speed if the
CalcStdy to True, GenDOF to True, RotSpeed to the constant generator torque is too large. The steady state
desired azimuth-averaged rotor speed (nonzero), and solution computation may become unstable if the initial
TrimCase to 3. Setting TrimCase to 3 causes FAST guess of nacelle yaw is too large when trimming yaw
to trim rotor collective blade pitch to reach the desired (TrimCase = 1). During Region 2 trim (TrimCase =
azimuth-averaged rotor speed condition. In this case, 2), the solution computation may also become unstable
the initial “guess” blade pitch angles are given by if your desired rotor speed is below the rotor speed that
BlPitchi, and the electrical generator torque is results in the maximum power coefficient at a given
determined by the torque-speed relationship indicated wind speed and rotor collective blade pitch angle. In
by inputs VSContrl or GenModel. For typical Region this case, the only way to obtain a successful trim
3 trim, collective pitch can be trimmed while solution is to increase your desired rotor speed
maintaining a constant generator torque by setting condition.
TrimCase to 3, VSContrl to 1, VS_RtTq to the
desired constant generator torque, and VS_RtGnSp,

FAST User's Guide 40 Last updated August 2005 for version 6.0
 Simulate through TMax Yes
Abort
1 rotor revolution reached?

No

Advance period: Trim control Is

n=n+1 input q
n +1 n
− q ≤ DispTol
Yes Yes AND
No n +1 n
q& − q& ≤ VelTol
No No GenDOF
CalcStdy? ?
enabled?

Yes
Use initial Add artificial Interpolate to selected
conditions damping azimuth steps

q op ,q& op ,q&&op ,
u op ,u dop

Figure 28. Periodic Steady State Computation.

averaging for you is described in the Post Processing

Model Linearization section below.
The FAST linearization routines can be used to
Once a periodic steady state solution has been
develop both a first- and a second-order linearized
found, FAST numerically linearizes the complete
representation of the nonlinear aeroelastic model. The
nonlinear aeroelastic model about the operating point.
order of the model is determined by input switch
Since the operating point is periodic with the rotor
MdlOrder. To understand the difference between
azimuth position, the linearized representation of the
these representations, we must examine how the
model is also periodic. Inputs NAzimStep, MdlOrder,
linearized state matrices relate to the nonlinear model.
NInputs, CntrlInpt, NDisturbs, and Disturbnc, which
The complete nonlinear aeroelastic equations of
are available in the linearization control-input-file of
motion as modeled in FAST can be written as follows:
FAST (identified by parameter LinFile in FAST’s
primary input file), are the parameters used to manage
the model linearization output. ( ) (
M q,u,t q&& + f q,q,u,u
& d ,t = 0 )
FAST will output the periodic linearized model at
where M is the mass matrix, f is the nonlinear “forcing
a number of equally spaced rotor azimuth steps as
function” vector, q is the vector of DOF displacements,
indicated by input parameter NAzimStep. The first
rotor azimuth location is always the initial azimuth (and q& and q&& are the DOF velocities and
position indicated by inputs Azimuth and AzimB1Up. accelerations), u is the vector of control inputs, ud is the
The subsequent azimuth steps increment in the vector of wind input “disturbances”, and t is time.
direction of rotation. Once a periodic steady state Note that in the steady state solution, only the DOF
solution has been found, FAST interpolates the displacement, velocity, and acceleration vectors are
solution to these azimuth locations. If RotSpeed is periodic with the rotor azimuth position. The vector of
zero, FAST will override NAzimStep and only control inputs and the vector of wind disturbances are
linearize the model about the initial azimuth position not periodic. In the above notation, capital letters
(as if NAzimStep was set to 1). If you are interested represent matrices and lower case underlined letters
in time-invariant control, you’ll obtain a more accurate represent vectors.
model if you output the linearized model at a number FAST numerically linearizes the aeroelastic
of different azimuth steps (by setting NAzimStep equations of motion by perturbing (represented by a ∆)
larger than 1) and then average the resulting matrices each of the system variables about their respective
rather than using one azimuth location by setting operating point (op) values:
NAzimStep equal to 1. A tool that will do this matrix

FAST User's Guide 41 Last updated August 2005 for version 6.0
 q = q op + ∆q , q& = q& op + ∆q& , q&& = q&&op + ∆q&& , transmission matrix, and Dd is the wind input
disturbance transmission matrix.
The DOF displacement, velocity, and acceleration
u = u op + ∆u , and u d = u dop + ∆u d . perturbation vectors ( ∆ q , ∆ q& , and ∆ q
&& ) are replaced
Substituting these expressions into the equations of with the first-order state vector x and state derivative
motion and expanding as a Taylor series approximation vector x& :
results in the second-order (MdlOrder equal 2)
linearized representation of the equations: ⎧⎪ ∆q ⎫⎪ ⎧⎪ ∆q& ⎫⎪
x = ⎨ ⎬ and x& = ⎨ ⎬
M ∆q&& + C ∆q& + K ∆q = F ∆u + Fd ∆u d
, ⎩⎪ ∆q& ⎪⎭ ⎩⎪ ∆q&&⎪⎭
in order to determine the first-order (MdlOrder equal
where 1) representation of the system:

M =M x& = Ax + B ∆u + Bd ∆u d
op

is the mass matrix, y = C x + D ∆u + Dd ∆u d

∂f In this form, the state matrix A, control input
C= matrix B, wind input disturbance matrix Bd, and output
∂q&
op state matrix C are related to their second-order
counterparts as follows:
is the damping/gyroscopic matrix,
⎡ 0 I ⎤ ⎡ 0 ⎤
⎡ ∂M ∂f ⎤ A=⎢ −1 −1 ⎥
, B = ⎢ −1 ⎥ ,
K =⎢ q&& + ⎥ ⎣ −M K −M C ⎦ ⎣M F ⎦
⎢⎣ ∂q ∂q ⎦⎥
op
⎡ 0 ⎤
is the stiffness matrix, Bd = ⎢ −1 ⎥ , and C = [ DspC VelC ]
⎣ M Fd ⎦
⎡ ∂M ∂f ⎤
F = −⎢ q&& + ⎥ where I is the identity matrix and 0 is a matrix of zeros.
⎣ ∂u ∂u ⎦ op The control input transmission matrix D and wind
input disturbance transmission matrix Dd are identical
is the control input matrix, and between the first- and second-order representations of
the linearized system.
∂ f
The sizes of the preceding matrices and vectors
Fd = −
depend on the number of DOFs enabled and the
∂u d op number of control inputs and wind input disturbances
selected. This is very convenient in controls-related
is the wind input disturbance matrix. The “ op
” work where it is useful to begin with a simple
notation is used to signify that the partial derivatives linearized “plant” model and then progressively add
are computed at the operating point. Internally within complexity in steps.
FAST, these partial derivatives are computed using the The number and type of DOFs incorporated in q,
central difference perturbation numerical technique. q& , and q&& are determined by the number of DOFs
Along with the linearized equations of motion,
FAST also develops a linearized system associated enabled. At least one DOF must be enabled during a
with output measurements y. The collection of output linearization analysis because a “plant” model with
measurements is specified in list OutList at the end of zero states is useless.
the primary input file. The second-order linearized The number and type of control inputs
representation of the output system is as follows: incorporated in u are specified through input
parameters NInputs and CntrlInpt. NInputs is the
y = VelC ∆q& + DspC ∆q + D ∆u + Dd ∆u d number of control inputs. Valid values are integers
from 0 to 4 + NumBl (inclusive). CntrlInpt is a list of
where VelC is the velocity output matrix, DspC is the numbers corresponding to different types of control
displacement output matrix, D is the control input inputs. Possible values are 1 to 7 (inclusive) (7 is only
available if NumBl = 3). The numbers correspond to

FAST User's Guide 42 Last updated August 2005 for version 6.0
 the seven control inputs described in Table 5. You
must enter at least NInputs values in this list. You can
separate the values with combinations of tabs, spaces, Table 6. Wind Input Disturbance Settings.
and commas, but you may use only one comma Disturbnc Description
between numbers. If NInputs is 0, input parameter Setting
CntrlInpt will be skipped. 1 Horizontal hub-height wind speed, V
2 Horizontal wind direction, DELTA
Table 5. Control Input Settings. 3 Vertical wind speed, VZ
CntrlInpt Description 4 Horizontal wind shear, HSHR
Setting 5 Vertical power law wind shear, VSHR
1 Nacelle yaw angle command 6 Linear vertical wind shear, VLinSHR
2 Nacelle yaw rate command 7 Horizontal hub-height wind gust, VG
3 Electrical generator torque
4 Rotor collective blade pitch As an example of the size of the output linearized
5 Individual pitch of blade 1 state matrices, consider a three-bladed turbine (NumBl
6 Individual pitch of blade 2 equal 3) modeled with only the FlpDOF1 and
7 Individual pitch of blade 3 (unavailable GenDOF DOFs set to True (enabled). Assume also
if NumBl = 2) that NInputs is set to 2 with CntrlInpt set to 3,4 and
NDisturbs is set to 1 with Disturbnc set to 1. This
If the yaw DOF is enabled (YawDOF = True), model has 4 DOFs (variable speed generator and one
then the commanded yaw angle and rate from flap mode for each of the three blades), 2 control inputs
CntrlInpt setting 1 and 2 are the neutral yaw angle, (electrical generator torque and rotor collective blade
YawNeut, and neutral yaw rate, YawRateNeut, in pitch), and 1 wind input disturbance (horizontal hub-
FAST's built-in second-order actuator model. In this height wind speed). Thus, q has size 4x1, x has size
case, the yaw actuator, which is described in the 8x1, M has size 4x4, A has size 8x8, B has size 8x2, Bd
Nacelle Yaw Control section of the Controls chapter, has size 8x1, etc. If 12 parameters were listed in
will be inherent in the output linearized model. If the OutList in this example, then C would have size 12x8,
yaw DOF is disabled (YawDOF = False), then the D would have size 12x2, etc. Each of these matrices
commanded yaw angle and rate from CntrlInpt setting would be output at each of the NAzimStep number of
1 and 2 are the actual yaw angle and yaw rate. In this equally spaced azimuth steps. Note also that if
case, the yaw actuator will be absent from the output NInputs were set to 0 instead of 2 in this example, then
linearized model. the B and D matrices would be absent from the
The number and type of wind input disturbances linearized system.
incorporated in ud are specified through input The name of the primary output file during a
parameters NDisturbs and Disturbnc. NDisturbs is linearization analysis uses the path and root name of
the number of wind input disturbances. Valid values the primary input file and appends .lin for an extension.
are integers from 0 to 7 (inclusive). Disturbnc is a list For example, if the input file were named fast.fst, the
of numbers corresponding to different types of wind main output file will be named fast.lin. This output file
input disturbances. Possible values are 1 to 7 contains the periodic state matrices of the linearized
(inclusive). The numbers correspond to the seven system, the periodic operating point states and state
inputs available in the hub-height wind data files of derivatives, the periodic operating point output
AeroDyn as described in Table 6. You must enter at measurements, the constant operating point values of
least NDisturbs values in this list. You can separate the control inputs and wind inputs, and other
the values with combinations of tabs, spaces, and information that is useful for post processing and for
commas, but you may use only one comma between making use of the linearized model. An example
numbers. If NDisturbs is 0, input parameter linearized model file is shown in Figure 31.
Disturbnc will be skipped.
Post Processing
The numerous output vectors and matrices may be
overwhelming, and one may wonder how to analyze
and make use of them. To aid in this effort, we have
developed a post processing script file in MATLAB
entitled Eigenanalysis.m. This script file is included in
the FAST archive in the CertTest folder. This script
file can be used as a basis for more advanced
utilization of the FAST linearization output. It is

FAST User's Guide 43 Last updated August 2005 for version 6.0
 written in MATLAB because MATLAB is the tool knowledge of the mass matrix). The form of the state
most commonly used in controls-related design work. matrix A without damping (as if C were zero) is
To run the Eigenanalysis.m script file, open a computed next. Azimuth-averaged matrices are then
MATLAB command window, change the current computed for all available periodic matrices. Finally,
working directory to the directory in which the script the script will perform an eigenanalysis on the periodic
file is stored, and type “Eigenanalysis” into the and azimuth-averaged state matrix A, with and without
MATLAB command prompt. The script file will damping. The resulting eigenvalues and eigenvectors
prompt you for the name of the FAST linearization are the full system natural frequencies and mode
output file to process. Type only the root name of this shapes. The natural frequencies are available in both
file—omit the .lin extension. The name may optionally rad/sec and Hz.
include an absolute or relative path if Eigenanalysis.m After the script file has completed execution, type
is not stored in the same directory as the FAST “who” into the MATLAB command prompt. This will
linearization output file. Naturally, a FAST cause MATLAB to list the available variables. The
linearization analysis must be run before the variable names are descriptive enough that they can be
Eigenanalysis.m script file is run, and a linearization discerned without further documentation. For
output file must be available for processing. example, the azimuth-averaged full system natural
Running Eigenanalysis.m will cause MATLAB to frequencies of the nondamped system, in Hz, are
read in the periodic state matrices from the FAST available in variable FrequenciesAvgHzNoDamp.
linearization output file. If the linearized model is Their associated mode shapes are available in variable
second-order, the script will then compute the first- ModeShapesAvgNoDamp.
order state matrices from the second-order matrices Please refer to MATLAB documentation for
using the equations documented above. If the additional help on running MATLAB and learning
linearized model is first-order, the script will not commands that are useful for controls-related design
compute the second-order matrices from the first-order work, which make use of linearized models output
state matrices because the process cannot be reversed from FAST.
(there is no unique solution available without

FAST User's Guide 44 Last updated August 2005 for version 6.0
 ADAMS PREPROCESSOR

that post-processing techniques are compatible for the

General Description codes.
Additionally, the ADAMS interface is developed
FAST has the capability of extracting “equivalent”
in such a way that all user-defined routines developed
ADAMS (Automatic Dynamic Analysis of Mechanical
for FAST, including UserGen(), UserVSCont(),
Systems) wind turbine datasets from the turbine
UserHSSBr(), UserPtfmLd(), UserTeet(),
properties specified in the FAST input file(s). That is,
UserRFrl(), UserTFrl(), UserTFin(),
FAST has the functionality of acting like an ADAMS
UserYawCont(), and PitchCntrl(), can be linked with
preprocessor capable of creating ADAMS datasets of
ADAMS just as easily as they can be with FAST. The
wind turbine models through FAST’s simple property-
routines do not need to be modified in any way for
input-style interface. Thus, FAST can be used as an
compatibility with ADAMS. The AeroDyn input files
alternative to the ADAMS/WT toolkit (12) or other are also fully compatible between the FAST models
preprocessors used to create ADAMS datasets of wind and the associated, extracted ADAMS datasets.
turbine models. The FAST to ADAMS preprocessor One of FAST’s limitations that is bypassed by
feature is enabled through the switch ADAMSPrep in ADAMS is the assumed-mode approximation of the
the primary input file. This chapter, which describes blades and tower. The blades and tower of the
the FAST to ADAMS preprocessor, assumes the reader extracted ADAMS model are developed from FAST’s
is familiar with the basics of ADAMS. distributed mass and stiffness inputs using ADAMS’
The main advantages for using FAST to create conventional approach of modeling flexible members
ADAMS datasets are to ensure consistency between through a series of lumped masses connected by
FAST and ADAMS models and to facilitate quick and stiffness and damping FIELDs. Nevertheless, FAST’s
easy creation of ADAMS datasets. The FAST-to- valuable DOF-switching functionality is still available
ADAMS preprocessor supports the mentality that all in the ADAMS model, so these flexibilities can be
pertinent configuration information be stored in a eliminated through a simple flag, just as they can be in
single location (i.e., the FAST input files). The FAST FAST (in ADAMS the flexibilities are eliminated
to ADAMS preprocessor provides a natural collectively, not one mode at a time).
progression from the medium-complexity FAST wind Moreover, several characteristics not implemented
turbine models to the highly complex models possible in the FAST model are incorporated into the extracted
using ADAMS. Once a working FAST model has ADAMS model. These include torsional and
been developed, little additional effort is required to extensional DOFs for the blades and tower, flap/twist
create the more advanced ADAMS model. This is a coupling in the blades, precurved and preswept blades,
useful way to impress your boss—you can accomplish mass and elastic offsets for the blades, mass offsets for
a lot of work with very little effort. ☺ the tower, actuator dynamics for the blade pitch
The ADAMS datasets extracted from FAST controls, graphical output capabilities, and others
contain all the functionality and usability associated documented below.
with the FAST model, while bypassing some of
FAST’s limitations. All the turbine control paradigms
available in FAST, as discussed in the Controls Compiling and Linking ADAMS
chapter, are incorporated into the ADAMS model. Using the extracted ADAMS datasets requires the
These include the functionality of yawing the nacelle, purchase and installation of the commercial ADAMS
pitching the blades, controlling the generator and HSS multibody-dynamics and analysis package. ADAMS is
brake torque, and deploying the tip brakes. The available from MSC.Software Corporation of Santa
ADAMS datasets incorporate the same generator, Ana, California. You will also need a compiler. For
drivetrain compliance, nacelle yaw, rotor-furl, tail-furl, the PC, you should use the Compaq Visual Fortran
rotor teeter, and support platform models and DOFs compiler, but its predecessor, Digital Fortran, will also
used by FAST. Also, all of the output parameters work.
specified at the end of FAST’s primary input file are Additionally, one must download the ADAMS to
passed into the ADAMS datasets. This eliminates the AeroDyn (A2AD) source files (13) and compile and
need to develop a REQSUB() user-written subroutine link the ADAMS user-created dynamic-link-library
for request output every time an ADAMS dataset is (DLL). This is where the compiler is required.
generated. Once an ADAMS analysis is run, the The A2AD archive is available for download from
format of the ADAMS output file containing time- our Web page
series data is identical to that of the FAST format so http://wind.nrel.gov/designcodes/simulators/adams2ad/
. The source files included in the archive, which are

FAST User's Guide 45 Last updated August 2005 for version 6.0
 pertinent to creating the DLL needed to run the same user-defined routines developed for FAST,
ADAMS datasets extracted from FAST, are as follows: including UserGen(), UserVSCont(), UserHSSBr(),
UserPtfmLd(), UserTeet(), UserRFrl(), UserTFrl(),
GFOSUB.f90 Contains an interface for the
UserTFin(), UserYawCont(), and PitchCntrl(), can
user-defined support platform
be used with the ADAMS model as well.
loading model and routines
Once these path variables are set, save the updated
that interface ADAMS to
script and run it from a command prompt by simply
AeroDyn so that AeroDyn can
typing its name. You can also run the script by double-
provide ADAMS with
clicking on it from Windows Explorer. Note that a file
aerodynamic forces on each
named newline.txt, which is also contained in the
blade element.
A2AD archive, must be located in the same directory
SENSUB.f90 Contains a routine to detect
as the CompileLinkA2AD.bat script in order for the
the occurrence of a successful
script to work. The script will create ADAMS.dll. This
forward time step. This is
one DLL can be used to run any ADAMS datasets
needed for the AeroDyn
extracted from FAST. In other words, you will only
interface.
need to create ADAMS.dll once (unless you change any
REQSUB_FAST.f90 Contains routines for
of the user-defined routines or source files).
calculating the desired
ADAMS output as specified in
FAST’s primary input file. Guidelines for Creating ADAMS
SFOSUB_FAST.f90 Contains a routine for Datasets
implementing the generator Here is the recommended procedure for creating
and variable-speed control ADAMS datasets using the FAST to ADAMS
models and an interface for the preprocessor:
user-defined rotor teeter, Step 1. Create a working FAST model. Do this
rotor-furl, and tail-furl spring by specifying the desired settings in the FAST input
and damper models. files, running a simulation, and then verifying that the
VARSUB_FAST.f90 Contains a routine for response predictions are reasonable.
computing the demand blade One drawback to creating both FAST and
pitch angles and demand ADAMS models from the same input file(s) is that if
nacelle yaw angle and rate. an error is made when inputting properties for a FAST
VFOSUB_FAST.f90 Contains routines for model, the extracted ADAMS model will contain the
computing the tip-brake drag same error. All redundancy checks available when
and tail fin aerodynamic FAST and ADAMS models are created independently
forces. are eliminated. To minimize resulting repercussions,
Source files GFOSUB.f90 and SENSUB.f90 are the make sure that the FAST model is in working order
generic routines provided in the A2AD archive for and is outputting reasonable response predictions
interfacing any ADAMS model to AeroDyn. A good before creating the ADAMS datasets.
description of these files is provided in (13). Step 2. Update the FAST input files to include the
additional input specifications required for creating
GFOSUB.f90 had to be modified slightly in order to
ADAMS datasets.
incorporate an interface to the user-defined support
The creation of ADAMS datasets using the FAST
platform loading model contained in routine
to ADAMS preprocessor requires the specification of
UserPtfmLd(). Source files REQSUB_FAST.f90,
additional parameters in the input file ADAMSFile.
SFOSUB_FAST.f90, VARSUB_FAST.f90, and
This file contains ADAMS-specific inputs related to
VFOSUB_FAST.f90 were written explicitly for running
the blade pitch actuators, graphical output capabilities,
ADAMS datasets extracted from FAST.
and other ADAMS-specific functionalities.
Also included in the A2AD archive is a file named
Furthermore, additional distributed blade and
CompileLinkA2AD.bat. This is a DOS command script
tower stiffness and inertial properties must be specified
useful for compiling and linking the ADAMS DLL,
in the blade and tower input files. These are input by
named appropriately, ADAMS.dll. You will need to
including additional columns of distributed data. For
modify this script before you can run it on your PC.
the blades, the additional columns are for distributed
Open the script with your favorite editor. You will
torsional and extensional stiffnesses, the distributed
need to change the variables DF_LOC, A2AD_LOC,
flap/twist coupling coefficient, distributed inertias,
AD_LOC, and FAST_LOC. Set them so they point to
distributed offsets for identifying the reference axis for
the locations of Digital Fortran and the A2AD,
precurved and preswept blades, and distributed mass
AeroDyn, and FAST source files respectively. The
and elastic offsets. For the tower, the additional
location of the FAST source files is needed so that the
columns are for distributed torsional and extensional

FAST User's Guide 46 Last updated August 2005 for version 6.0
 stiffnesses, distributed inertias, and distributed mass DOFs can’t be enabled without also enabling side-to-
offsets. See the sample input files and the Input Files side DOFs or vice-versa. The setting of input flags
chapter for more details on these additional inputs. TwFADOF2 and TwSSDOF2 do not effect the
Several input parameters are handled differently creation of the ADAMS datasets.
between FAST and the ADAMS preprocessor. Users Support platform rotational DOFs are controlled
of the FAST to ADAMS preprocessor should be aware by enabling input flags PtfmRDOF, PtfmPDOF, and
of these differences. PtfmYDOF. Due to the method ADAMS uses to
The time step size input parameter, DT, is used by implement the rotational DOFs, FAST cannot build an
the ADAMS preprocessor to specify the maximum step ADAMS dataset if one of the platform rotational DOFs
size the integrator is allowed to take in the variable- is set differently than the other two. Thus, you must set
step-size numerical-integration scheme that is used by PtfmRDOF, PtfmPDOF, and PtfmYDOF to the same
ADAMS. Users of the ADAMS preprocessor should value (i.e., all .True or all False). There is no
be aware that this is in slight contradiction to how DT restriction on which combination of support platform
is used to specify the constant time step size for the translational DOFs are enabled.
numerical integration scheme that is used by FAST. Additionally, some features available in FAST
Since the blade and tower models incorporated can’t be modeled in ADAMS. Thus, when creating
into the ADAMS datasets do not operate on the modal ADAMS datasets using the FAST to ADAMS
principle, input parameters associated with modal preprocessor, FAST will abort and not create the
properties are naturally handled differently in the ADAMS dataset if any of these features are selected.
ADAMS preprocessor. Mechanical gearbox efficiency losses are very
Blade flap and edge damping ratios incorporated difficult to model in ADAMS. Thus, GboxEff must be
in ADAMS FIELD statements are set equal to the same set to 100% when creating ADAMS datasets.
ratios used for the first flap and edge modes in FAST. Similarly, the physics of a gearbox whose LSS and
These ratios are determined by inputs BldFlDmp(1) HSS rotate in opposite directions are difficult to model
and BldEdDmp(1) in the blade input file(s). The value in ADAMS. Thus, GBRevers must be set to False
of input BldFlDmp(2) does not affect the creation of when creating ADAMS datasets.
ADAMS datasets. The initial displacements of the blades and tower,
Likewise, tower fore-aft and side-to-side damping specified using inputs OoPDefl, IPDefl, TTDspFA,
ratios incorporated in ADAMS FIELD statements are and TTDspSS, must all be zero when creating
set equal to the same ratios used for the first fore-aft ADAMS datasets. This is due to the difficulty
and side-to-side modes in FAST. These ratios are involved in assembling an ADAMS dataset that
determined by inputs TwrFADmp(1) and contains deflected flexible members at the model
TwrSSDmp(1) in the tower input files. The values of definition phase. The initial teeter angle, TeetDefl,
inputs TwrFADmp(2) and TwrSSDmp(2) do not must also be set to zero when creating ADAMS
affect the creation of ADAMS datasets. datasets. This restriction exists since the generic
Moreover, the modal stiffness tuner input GFOSUB.f90 routines provided in the A2AD archive
parameters contained in the blade and tower data files will not read initial blade element data properly if
are completely ignored by the ADAMS preprocessor. TeetDefl is nonzero. All of the other initial conditions
Blade flexibility is controlled by enabling the first specified in FAST’s primary input file, including the
blade modes through input flags FlapDOF1 and initial rotor speed (RotSpeed), are incorporated in the
EdgeDOF. Enabling blade flexibility enables all ADAMS datasets.
blade flap, edge, torsional, and extensional DOFs; Due to a restriction in ADAMS, the title line of
conversely, disabling the flexibility removes all of FAST’s primary input file, the third line in the file,
these DOFs. When using the ADAMS extractor, the must not contain any of the characters “,”, “;”, “&”, or
setting of FlapDOF1 must be identical to that of “!” when creating ADAMS datasets. This is because
EdgeDOF to emphasize that the flap DOFs can’t be the title is stored as a STRING statement in the
enabled without also enabling edge DOFs or vice- ADAMS datasets (used for providing header
versa. The setting of input flag FlapDOF2 does not information in the primary output file) and because
affect the creation of the ADAMS datasets. ADAMS STRING statements prohibit the use of these
Tower flexibility is controlled by enabling the first characters.
tower modes through input flags TwFADOF1 and Sensible limits are placed on the number of blade
TwSSDOF1. Enabling tower flexibility enables all and tower elements so that a reasonable numbering
tower fore-aft, side-to-side, torsional, and extensional scheme could be implemented for ADAMS blade and
DOFs, and conversely, disabling the flexibility tower PART, MARKER, and FIELD statements. This
removes all of these DOFs. When using the ADAMS restriction is that neither TwrNodes nor BldNodes
extractor, the setting of TwFADOF1 must be identical can be greater than 99.
to that of TwSSDOF1 to emphasize that the fore-aft

FAST User's Guide 47 Last updated August 2005 for version 6.0
 Finally, users of the FAST-to-ADAMS their associated FAST models. Please refer to Step 1
preprocessor should also be aware that the linearization of the previous section for additional information
control features and Simulink interface available in regarding this issue.
FAST are not available in the extracted ADAMS Using ADAMS.dll, an extracted ADAMS dataset,
models. and an extracted ADAMS control/command file, you
Step 3. Run FAST with ADAMSPrep set at 2 or can run the ADAMS simulation using the Run Custom
3. When ADAMSPrep is set to 2, FAST creates the Solver prompts available in ADAMS Solver.
ADAMS datasets and stops. When ADAMSPrep is set Experienced users of ADAMS can also develop script
to 3, FAST creates the ADAMS datasets and then files so that ADAMS.dll can be run from any directory
proceeds to run the FAST simulation as well. without manually going through the ADAMS Solver
When running, FAST generates two ADAMS files prompts.
as follows: In order to get the ADAMS simulation to run
<RootName>_ADAMS.adm smoothly and converge accurately, you will most likely
The ADAMS dataset containing need to play around with the time step size and
statements that characterize the integrator error. The FAST-to-ADAMS preprocessor
model configuration, analysis automatically enters a default integrator error of 0.001
settings, and output. for all of the ADAMS INTEGRATOR/GSTIFF
<RootName>_ADAMS.acf statements. This should be suitable for most
The ADAMS command file simulations, but may require adjustment for some.
containing statements that enable Please refer to (13) for additional hints on running
DOFs and drive the time- ADAMS simulations.
marching simulation. ADAMS generates several output files. The
where RootName is the name of the primary input file. primary output file of interest has a .plt (for plot)
For example, if the input file were named fast.fst, the extension. This file contains the columns of time-
extracted ADAMS files will be named series data with one column for each parameter that is
fast_ADAMS.adm and fast_ADAMS.acf. requested in the primary input file. The format of this
An additional file named output file is identical to that of FAST’s .out file so that
<RootName>_ADAMS_LIN.acf is generated when flag post-processing techniques are compatible for the
MakeLINacf is enabled in the ADAMSFile. This third codes. The other files created are generic ADAMS
file contains statements that drive an output files generated by ADAMS Solver. It may or
ADAMS/LINEAR eigenanalysis of the model. The may not be useful to review these.
eigenanalysis is performed with no gravity, rotor speed, When examining time-series data output from
damping, or aerodynamics, no matter how the ADAMS, please be aware of the following limitations.
associated inputs are otherwise specified in FAST’s If the tower is rigid, the tower base loads will all be
input files. output with zeros—unfortunately, we haven’t found a
If you want to change the ADAMS model reason or workaround for this anomoly. Also, when
properties or analysis settings, simply change the outputting local span blade loads, be aware that the
desired input properties in the FAST input file(s) and loads at the outboard strain gages will be under-
run FAST again to create the updated ADAMS predicted by ADAMS. This is because ADAMS lumps
datasets. This causes FAST to overwrite your old all of its mass at the center of each segment (rigid body
datasets if they are located in the same directory in inertia effects are also included), and thus, the mass of
which FAST is called. You never need to manually the segment in which the local span loads are output
change the ADAMS datasets unless you want to does not contribute to the local load. This difference
change the ADAMS model to include features not shows most in the outboard part of the blade and
supported by the FAST-to-ADAMS preprocessor. becomes insignificant toward the root. Finally, be
Before running the ADAMS simulation, it may be aware that the gyroscopic pitching moments induced
beneficial to examine the model in ADAMS View to when the nacelle yaws while the drivetrain is spinning
make sure the configuration is as expected. To do this, will be over-predicted by ADAMS. This results from
open ADAMS View, choose Import… from the File the fact that the HSS is lumped to the LSS in the
menu, and then browse and select the ADAMS dataset ADAMS model. A detailed explanation of why this
(.adm) of interest. lumping affects the gyroscopic pitch moments is
provided in (6).
If SaveGrphcs is enabled in ADAMSFile,
Running ADAMS ADAMS will also generate a graphics output file with
Before running ADAMS, it is imperative that you a .gra extension. The graphics output file may be used
first understand FAST. This is because the extracted to view an animation of the ADAMS simulation. To
ADAMS datasets are considerably more complex than view the simulation, open ADAMS View, choose

FAST User's Guide 48 Last updated August 2005 for version 6.0
 Import… from the File menu, and then browse and statements in the ADAMS datasets are in kilonewtons,
select the ADAMS graphics file of interest. Once kilograms, meters, and seconds for the force, mass,
loaded, the animation can be played by choosing length, and time units respectively. These are specified
Animation Controls… from the Review menu. When through the use of a UNITS statement. The
running many ADAMS simulations, it is beneficial to acceleration of gravity, specified using an ACCGRAV
disable SaveGrphcs—ADAMS will not then generate statement, acts in the negative z-direction of the
the graphics output file and will run faster as a result. GROUND reference frame and MARKER.
When ADAMS is run using the control/command The ADAMS dataset containing statements that
file for the ADAMS/LINEAR analysis, a results output characterize the model configuration, analysis settings,
file with a .res extension is generated. To animate the and output (the .adm file) is organized into five main
system mode shapes, open ADAMS View, choose sections. The first is a header section containing
Import… from the File menu, and then browse and comments that identify the model name and describe
select the ADAMS results file of interest. Once how and when the dataset was created. The second
loaded, the system modes can be animated by choosing section contains definitions of all model PART,
Linear Modes Controls… from the Review menu. If MARKER, and GRAPHICS statements. The third
you only require a listing of the system natural section contains the constraint JOINT and MOTION
frequencies, these results are written and can be statements, and the fourth section contains force
retrieved from either the .out (output) or .msg definitions, including FIELD, FRICTION, SFORCE,
(message) files automatically generated by ADAMS VFORCE, GFORCE, and SPRINGDAMPER
Solver. statements. The last section contains definitions of
analysis settings and output.
Description of the Extracted ADAMS In each section that contains model definition
Datasets statements, the model is assembled from the ground up,
from the support platform through the blade tip. So,
This section documents qualitatively how the for example, if you want to examine the tower FIELD
extracted ADAMS datasets are organized and how the statements, look at the statements near the beginning of
various components of the wind turbine model are the force definition section.
implemented in ADAMS. If you are interested in The ADAMS command (.acf) files contain
manually modifying the ADAMS datasets in order to commands used to drive the simulation. These include
incorporate features not available by the FAST-to- an INTEGRATOR statement and either
ADAMS preprocessor, it is important to review this SIMULATE/DYNAMICS or LINEAR/EIGENSOL
section first. If you have no interest in learning how commands, depending on the type of analysis
the various components of the wind turbine model are performed. The .acf files also contain DEACTIVATE
implemented in ADAMS, you may skip this section commands used to remove superfluous constraints and
entirely. enable DOFs based on the specifications in the feature
To learn the exact details on how the ADAMS flags section of FAST’s primary input file. Additional
datasets are generated, see the file FAST2ADAMS.f90 details on this last point are provided at the end of this
in the FAST source code. This Fortran file contains section.
three subroutines, one for generating each of the Each tower and blade element is characterized by
ADAMS files output by FAST. its own PART statement. The center of mass
In general, the ADAMS datasets generated by MARKERs of these PARTs are located at the same
FAST employ the same PART and MARKER vertical (for tower) and radial (for blade) locations as
numbering conventions as recommended in (13). For the analysis nodes used in FAST. For blades, the
reference, a complete listing of all of the ADAMS transverse locations of the center of mass MARKERs
statements (PARTs, MARKERs, FIELDs, JOINTs, are positioned with the reference axis and center-of-
etc.) contained in the extracted ADAMS datasets is gravity offsets specified in the blade input file. For
provided on our Web page http://wind.nrel.gov/ towers, the transverse locations of the center of mass
designcodes/simulators/adams2ad/FAST2ADAMSStat MARKERs are identified using the distributed center-
ements.xls. However, not all extracted ADAMS of-gravity offsets specified in the tower input file.
datasets contain every statement documented in this Interconnecting each PART is a stiffness and damping
spreadsheet—only statements that are needed are FIELD statement. The FIELDs attach to the PARTs at
included. For example, if the generator speed is held elastic axis MARKERs, which are located in the same
fixed by disabling the generator DOF, then the transverse planes as the center of mass MARKERs of
SFORCE statement used to model an unneeded each PART. For the blades, the transverse locations of
generator torque will not be included in the ADAMS the elastic axis MARKERs are identified using the
dataset. distributed reference axis and elastic-axis offsets
When examining the ADAMS datasets, it is
important to note that the system units for all
FAST User's Guide 49 Last updated August 2005 for version 6.0
 specified in the blade file, and for the tower, are Likewise, the tail-furl bearing is modeled with a
assumed coincident with the tower axis. revolute JOINT. When TFrlMod is set to 1, the
Users should note that a flexible member (a blade standard, linear compliance is modeled with a
or tower) with “N” analysis nodes, will be assembled rotational SPRINGDAMPER statement and the
using “N+1” FIELD statements. “N-1” of the FIELD nonlinear up- and down- spring and damper stops are
statements are used to interconnect the “N” analysis modeled with explicit function-based SFORCE
nodes to each another. The “Nth” FIELD statement is statements. When TFrlMod is set to 2, the user-
used to cantilever node 1 to the flexible member’s rigid defined tail-furl spring and damper model provided in
base. The “(N+1)th” FIELD statement is used to routine UserTFrl() is interfaced to ADAMS from
connect node “N” to the tip of the flexible member, routine SFOSUB(), which in turn, is called from an
which for the tower is the tower-top and for a blade is SFORCE statement. If the tail-furl DOF is disabled,
the tip brake. For blades with no tip brakes (indicated the tail-furl JOINT is “locked” using a steady
by setting the associated TipMass to zero), the DOFs MOTION statement.
associated with the outermost FIELD statement are Low-speed shaft compliance is modeled with a
never enabled. revolute JOINT and a rotational SPRINGDAMPER
The support platform, tower top, bed plate, statement. When drivetrain rotational flexibility is
nacelle, generator, HSS, LSS, teeter pin, hub, pitch disabled, the JOINT is “locked” with a zero-valued
plates, tip brakes, tail, and structure furling with the MOTION statement.
rotor are all modeled using rigid PART statements. Drivetrain torque models, including both the
Each of these PARTs contains several MARKERs for generator models and variable-speed control models,
identifying various locations and directions. are implemented with an SFORCE (1-component
In the graphics output, each tower and blade scalar force) statement that calls routine SFOSUB().
element is identified with its own unique GRAPHICS This routine implements each type of drivetrain torque
statement. Additional GRAPHICS statements are used model available in FAST. The parameters passed to
to illustrate the rigid tower base, nacelle, gearbox, SFOSUB() depend on the type of drivetrain torque
HSS, LSS, generator, hub, tail boom, and tail fin. model selected in FAST’s primary input file. If
The yaw bearing is modeled with a revolute necessary, SFOSUB() will call the user-defined
JOINT. Nacelle yaw demand angles and rates, arising UserGen() or UserVSCont() routines.
from both advanced yaw control algorithms and The high-speed shaft brake is implemented with a
override yaw maneuver specifications, are computed in preload-only FRICTION statement. Since the
VARSUB() and stored in VARIABLE statements. If magnitude of the preload torque cannot be time-
necessary, based on settings in FAST’s primary input varying, the full friction torque is applied throughout
file, routine VARSUB() calls FAST’s user-defined the entire simulation. To negate the unwanted
yaw control routine, UserYawCont(). The difference resistance before THSSBrDp, an SFORCE statement
between the yaw demand angle and actual yaw angle that calls SFOSUB() is used to apply a cancellation
(yaw error) and the yaw demand rate and actual yaw torque between the shaft and the nacelle. The
rate (yaw rate error) is passed through a yaw actuator SFOSUB() also applies the linear ramping component
model that is implemented with an explicit function- of the braking torque (over time increment HSSBrDT)
based SFORCE statement. This is the same yaw when HSSBrMode is set to 1. Alternatively,
actuator inherent in equivalent FAST models. If the SFOSUB() calls FAST’s user-defined HSS brake
yaw DOF is disabled, the yaw JOINT is “locked” using routine UserHSSBr() to determine the fraction of
a steady MOTION statement, but the yaw DOF cannot torque to cancel out when HSSBrMode is set to 2.
be disabled if yaw control is enabled (the ADAMS The teeter bearing is modeled with a revolute
preprocessor will abort if you try). JOINT. When TeetMod is set to 1, the standard,
The rotor-furl bearing is modeled with a revolute nonlinear teeter spring and damper models are
JOINT. When RFrlMod is set to 1, the standard, linear implemented with explicit function-based SFORCE
compliance is modeled with a rotational statements. When TeetMod is set to 2, the user-
SPRINGDAMPER statement and the nonlinear up- and defined teeter spring and damper model provided in
down- spring and damper stops are modeled with routine UserTeet() is interfaced to ADAMS from
explicit function-based SFORCE statements. When routine SFOSUB(), which in turn, is called from an
RFrlMod is set to 2, the user-defined rotor-furl spring SFORCE statement. If the teeter DOF is disabled, the
and damper model provided in routine UserRFrl() is teeter JOINT is “locked” using a steady MOTION
interfaced to ADAMS from routine SFOSUB(), which statement.
in turn, is called from an SFORCE statement. If the The blade pitch bearing is modeled with a revolute
rotor-furl DOF is disabled, the rotor-furl JOINT is JOINT. Blade pitch demand angles, arising from both
“locked” using a steady MOTION statement. advanced pitch control algorithms and override pitch
maneuver specifications, are computed in VARSUB()

FAST User's Guide 50 Last updated August 2005 for version 6.0
 and stored in VARIABLE statements. If necessary, the A2AD source file REQSUB_FAST.f90 to determine
based on settings in FAST’s primary input file, routine which channels to output. Additional parameters
VARSUB() calls FAST’s user-defined pitch control needed for computing output data are passed to
routine, PitchCntrl(). The difference between the pitch REQSUB(), a routine contained in file
demand angle and actual pitch angle (pitch error) is REQSUB_FAST.f90. Routine REQSUB() is, in turn,
passed through a pitch actuator model that is called using a REQUEST statement in the ADAMS
implemented with an explicit function-based SFORCE dataset.
statement. If pitch is not actively controlled during the No matter which DOFs are enabled when
simulation, the pitch JOINT is “locked” using a steady generating the ADAMS datasets with the FAST-to-
MOTION statement. ADAMS preprocessor, the ADAMS dataset is always
As described in the A2AD User’s Guide (13), assembled so that it possesses no DOFs upon initiation.
ADAMS is interfaced to AeroDyn through routines That is, all DOFs are essentially “locked” during the
provided in the A2AD source file GFOSUB.f90. model-loading phase. This is achieved by placing
Routine GFOSUB(), which is contained in fixed JOINTs between each PART of the flexible
GFOSUB.f90, is called from GFORCE (6-component blades and tower, placing an ORIENTATION JOINT
general force) statements placed in the ADAMS between the GROUND and support platform PART,
dataset. There is one GFORCE statement for each and by specifying steady MOTION statements at all
blade analysis node (or element) in every blade. revolute JOINTs for the yaw, rotor-furl, tail-furl, teeter,
A GFORCE statement, together with the shaft, and blade pitch bearings.
GFOSUB(), is also used to interface ADAMS with the Once the simulation begins using the ADAMS
user-defined support platform loading model, routine command (.acf) file, the first time step is processed
UserPtfmLd(). with all the DOFs “locked”. After the first time step,
Tip-brake drag forces are modeled using VFORCE the selected DOFs are enabled by removing the
(3-component vector force) statements that call routine superfluous MOTION, JPRIM, and fixed JOINT
VFOSUB(). The VFOSUB() routine employs the statements through the use of DEACTIVATE
same simple logic FAST uses for computing tip brake commands. Processing the first time step with zero
drag forces. DOFs ensures that the initial condition solution, which
Tail fin aerodynamic loads are also modeled using always precedes the first SIMULATE/DYNAMICS
a VFORCE (3-component vector force) statement that event, does not “kick” the system when the rotor is
calls routine VFOSUB(). When TFinMod is set to 1, initially spinning at some nonzero-valued rate. Instead,
the VFOSUB() routine employs the same simple logic the simulation always begins with no initial
FAST uses for computing the tail fin aerodynamic deflections. This technique essentially bypasses the
loads. When TFinMod is set to 2, the user-defined tail startup problems pertaining to most ADAMS datasets
fin aerodynamic model provided in routine UserTFin() as discussed in (13). The transient behavior associated
is interfaced to ADAMS from routine VFOSUB(). with the startup of an ADAMS analysis, should be
All output parameter names, units, and identifiers nearly identical to that associated with the startup of a
are stored in the ADAMS dataset using ARRAY and corresponding FAST analysis.
STRING statements. These are read in by routines in

FAST User's Guide 51 Last updated August 2005 for version 6.0
FAST User's Guide 52 Last updated August 2005 for version 6.0
 INPUT FILES
pertinent input files. Table 7 provides a general
Sample Input Files description of these sample models. The sample input
files associated with these models are available in the
The sections that follow describe the format of the
CertTest folder and should be used as templates for
various program input files. In the FAST archive, we
creating your own models.
provide a sample set of 17 models, including all

Table 7. Sample Models Provided with the FAST Archive.

No. Rotor Rated
Test Blades Diameter Power
Name Turbine Name (-) (m) (kW) Test Description
Test01 AWT-27CR2 2 27 175 Flexible, fixed yaw error, steady wind
Test02 AWT-27CR2 2 27 175 Flexible, start-up, HSS brake shut-down, steady wind
Test03 AWT-27CR2 2 27 175 Flexible, free yaw, steady wind
Test04 AWT-27CR2 2 27 175 Flexible, free yaw, turbulence
Test05 AWT-27CR2 2 27 175 Flexible, generator start-up, tip-brake shutdown, steady wind
Test06 AOC-15/50 3 15 50 Flexible, generator start-up, tip-brake shutdown, steady wind
Test07 AOC-15/50 3 15 50 Flexible, free yaw, turbulence
Test08 AOC-15/50 3 15 50 Flexible, fixed yaw error, steady wind
Test09 UAE VI downwind 2 10 20 Flexible, yaw ramp, steady wind
Test10 UAE VI upwind 2 10 20 Rigid, power curve, ramp wind
Test11 WP 1.5 MW 3 70 1500 Flexible, variable speed & pitch control, pitch failure, turbulence
Test12 WP 1.5 MW 3 70 1500 Flexible, variable speed & pitch control, ECD event
Test13 WP 1.5 MW 3 70 1500 Flexible, variable speed & pitch control, turbulence
Test14 WP 1.5 MW 3 70 1500 Flexible, stationary linearization, vacuum
Test15 SWRT 3 5.8 10 Flexible, variable speed control, free yaw, tail-furl, EOG01 event
Test16 SWRT 3 5.8 10 Flexible, variable speed control, free yaw, tail-furl, EDC01 event
Test17 SWRT 3 5.8 10 Flexible, variable speed control, free yaw, tail-furl, turbulence

on the line is treated as a comment. Values for string

Primary Input File variables must be delimited with a pair of apostrophes
or double quotes. Logical flags must be unquoted
FAST uses a primary input file to describe the
strings that start with t or T for True and f or F for
wind turbine operating parameters and basic geometry.
False. The Variable-Name section contains the
However, the blade, tower, furling, and aerodynamic
variable name used internally by the program and in
parameters and wind-time histories are read from
references for other parameters. The Description
separate files. Additionally, input parameters related to
section of the line contains a brief description of the
FAST linearization and parameters only necessary for
parameter as a reminder to the user of its purpose. This
creation of ADAMS datasets are read in from separate
section also contains the physical units of the
files (see Figure 29). Descriptions of the individual
numerical value, where appropriate. A sample line
inputs in the various files are provided below. Output
from the input file for input parameter TMax, divided
files are discussed in the Output Files chapter.
into its sections is shown below:
The primary input file has a default name of
primary.fst, which FAST will try to open if you do not 20.0 TMax - Total run time (s)

specify a file name on the command line. If you want

Note that there are no blank lines in the input files.
to use different names for different cases, you can
The program reads each line in sequential order. You
specify a file with a different name on the command
should never add or delete any lines except in the
line. Files with spaces in their paths or names must be
various sections that allow it. The tower, blade and
delimited by quotes. File names are limited to 99
AeroDyn input files have sections where you enter one
characters and may include absolute or relative paths.
line per input station analysis node. The primary
The parameter input files have a simple text format
FAST input file has a list of output parameters at the
that can be read and modified by any text editor. Most
end of the file that can be as long as you like.
lines in the input file are divided into three sections:
Note also that lines containing section or file titles
value(s), variable name(s), and description. For lines
may be altered to suit the user. FAST ignores these
that require more than one value, separate them with
lines when it reads the input, but these lines should not
spaces, tabs, or commas. Anything past the last value
FAST User's Guide 53 Last updated August 2005 for version 6.0
 be deleted for the same reasons mentioned in the name three times. One more parameter in the primary
previous paragraph. input file identifies the name of the file containing
Some parameters do not apply to two-bladed additional model properties for a furling turbine
turbines, and others do not apply to three-bladed (FurlFile), another specifies the aerodynamic noise
turbines. FAST treats these as comments. Any text input file (NoiseFile), and another identifies the name
may be used on the unused lines, or they may even be of the AeroDyn input file (ADFile). Additionally, the
blank, but they must exist. The sample input files name of the input file relating to a FAST linearization
identify such parameters. analysis (LinFile) and the name of the file containing
Several other files are read for additional ADAMS-specific data input (ADAMSFile) are further
parameters. One of the parameters in the primary file parameters in the primary input file.
is the name of the tower file (TwrFile). There are three Table 8 lists the input parameters for the primary
other parameters designating the names of the three input file.
blade input files (BldFilei). If you want to use identical
properties for all blades, you may specify the same

Primary Furling Blade(s) ADAMS- Primary Wind

Platform Specific Airfoil(s)
Tower Linear

FAST AeroDyn
FAST-to-ADAMS
Simulation Preprocessor
Linearization

Summary Time Periodic ADAMS Summary Element

Series Matrices Dataset/
Control(s)

Figure 29. Input and Output Files.

linearly interpolate these data to the centers of the

Tower Input File equally spaced segments, which are the analysis nodes.
There are TwrNodes segments or analysis nodes. To
In the tower file, there is a table for the tower
get the most accurate results from these properties,
characteristics, which requires several columns of data
include data points for the analysis nodes in the input
in the Distributed Tower Properties section of the input
table.
file. Only the first four columns are used to
Table 9 lists the input parameters for the tower
characterize the FAST model. The last six columns are
input file. FAST reads this file even if you requested
used only for creating ADAMS datasets using the
no tower DOFs.
FAST-to-ADAMS preprocessor feature of FAST. You
need to enter only one line if the tower is uniform.
You must specify a zero for the location of this single Blade Input Files
station. If you model a non-uniform tower, you must In the blade input files, there are tables of blade
specify at least two stations; the first must be at the 0 characteristics. There are several columns of data in
location and the last must have a location of 1 (for the Distributed Blade Properties section, each of which
100% of the flexible height of the tower). FAST will must be separated from the other by a space, tab, or

FAST User's Guide 54 Last updated August 2005 for version 6.0
 comma. Only the first six columns are used to Furling Input File
characterize the FAST model. The last 11 columns are
Table 13 lists the input parameters for the furling
used only for creating ADAMS datasets using the
input file. The inputs pertain to the lateral offset and
FAST-to-ADAMS preprocessor feature of FAST. You
skew angle of the rotor shaft, rotor-furling, tail-furling,
need to enter only one line if the blade is uniform. You
and tail inertia and aerodynamics. If the turbine you
must specify a zero for the location of this single
want to model contains any of these characteristics,
station. If you model non-uniform blades, you must
you must assemble the furling input file even if your
specify at least two stations; the first must be at the
turbine does not “furl” in the common sense of the
zero location and the last must have a location of 1 (for
word. For example, if the turbine you want to model
100% span). FAST will linearly interpolate these data
contains a tail, you must assemble the furling input file
to the analysis nodes specified in the AeroDyn input
regardless of whether or not your tail, or rotor, actively
file. There are BldNodes segments or analysis nodes.
furls about the yawing-portion of the structure atop the
To get the most accurate results from these properties,
tower.
include data points for the analysis nodes in the input
It is clear that the inputs available in the furling
table.
input file define the core configuration of the turbine,
Table 10 lists the input parameters for the blade
just like those available in the primary input file. The
input files. FAST reads this file even if you requested
reason we separated the parameters between the two
no blade DOFs.
input files is that the parameters available in the furling
input are unique to small wind turbines. The challenge
AeroDyn Input Files in defining the unique configurations of small wind
Table 11 lists the input parameters for the primary turbines relative to the configurations of conventional
AeroDyn input file. AeroDyn also uses other input machines is clearly demonstrated by the contents of the
files. The current version of AeroDyn accommodates furling input file. Who said small wind turbines are
two types of wind files. One type specifies hub-height easier to design than large wind turbines?
wind data that also includes wind shears and gusts. The furling input file is organized into sections
You can use IECWind (14) or WindMaker (15) to similar to those available in the primary input file.
generate these files for standard IEC wind conditions. This supports the notion that the furling file is simply a
You can also fabricate these simple text files from continuation and expansion of the core configuration-
scratch or even use field-test data. The other type of definition designations available in the primary file.
wind file contains full-field wind data in a binary form. FAST only reads the furling input file if the model
TurbSim (16), SNwind (17), or SNLWIND-3D (18) is designated as a furling machine (when Furling is set
can generate these files. They contain two-dimensional to True).
grids of three-component winds that march past the
turbine at a mean wind speed. AeroDyn also reads one ADAMS-Specific Input File
or more files containing airfoil data.
Table 14 lists the input parameters for the
Please see the AeroDyn user’s guide (1) for
ADAMS-specific-input data file. This file contains
additional details on these files. FAST reads these files
inputs related to the blade pitch actuators, graphical
even if you disabled aerodynamic calculations.
output capabilities, and other ADAMS-specific
functionalities.
Platform Input File FAST does not read this file if ADAMS datasets
Table 12 lists the input parameters for the platform are not generated (when ADAMSPrep is set to 1).
input data file. This file contains inputs related to the
support platform configuration, motions, and loading. Linearization Control-Input File
FAST only reads the platform input file if
Table 15 lists the input parameters relating to a
PtfmModel from the primary input file is nonzero. In
FAST linearization analysis. FAST only reads in this
FAST v6.0, all nonzero PtfmModel options will work
file when performing a linearization analysis (when
the same way by reading in the PtfmFile described in
AnalMode is set to 2).
Table 12. In future versions, the format of this file will
depend on which PtfmModel option is selected.

FAST User's Guide 55 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters.

Simulation Control
Echo Setting this flag to True will cause FAST to echo input values as it reads them. It writes the data
to the file echo.out. This is a useful tool to debug problems with your input files. For normal
operation, set this parameter to False. (flag)
ADAMSPrep This switch determines whether or not the ADAMS preprocessor is enabled. Setting
ADAMSPrep to 1 disables the ADAMS preprocessor and causes FAST to run its simulation as
normal. A setting of 2 turns on the ADAMS preprocessor and turns off FAST; when FAST is run,
the ADAMS datasets are created and FAST stops without performing a simulation. A setting of 3
enables both FAST and the ADAMS preprocessor; when FAST is run, the ADAMS datasets are
created and FAST proceeds to run its simulation. Using values other than 1, 2, or 3 will cause
FAST to abort. ADAMSPrep must be 1 when FAST is interfaced with Simulink. (switch)
AnalMode This switch determines whether to perform a time-marching analysis (simulation) or a
linearization analysis (i.e., AnalMode stands for the analysis mode). A setting of 1 indicates a
time-marching analysis. To perform a linearization analysis, set AnalMode to 2. Using values
other than 1 or 2 will cause FAST to abort. AnalMode must be 1 when FAST is interfaced with
Simulink. This input is not used in the FAST-to-ADAMS preprocessor. (switch)
NumBl This is the number of blades on the rotor. Valid values are 2 and 3. (-)
TMax The overall simulation runtime. For time-marching simulations, the simulation stops when TMax
is reached. When computing a steady state solution during a linearization analysis, the iteration
stops and FAST aborts if the solution has not converged by the time TMax is reached. (sec)
DT This is the time step for the constant-step-size numerical-integration scheme that is used by FAST.
For ADAMS datasets extracted from FAST, DT is used to specify the maximum step size the
integrator is allowed to take in the variable-step-size numerical-integration scheme that is used by
ADAMS. You should be careful to choose an appropriate value for DT because if DT is too small
or too large, the numerical solution will become unstable. Whenever you make changes to the
configuration of your model, you should experiment with different values for DT and choose
the largest value that does not affect your results. (sec)

Turbine Control
YCMode This is the yaw-control-mode switch for user-defined nacelle yaw control. Setting it to 0 disables
user-defined yaw control. Setting it to 1 causes FAST to call a user-written routine called
UserYawCont() at every time step past TYCOn. We supply a dummy routine in the software
folder to help you write your own. Setting YCMode to 2 causes FAST to accept yaw position and
rate demands externally from Simulink. The simple yaw maneuvers described below override the
control setting determined by the user-supplied yaw controllers. Please see the Controls chapter
for further details. YCMode must be 0 during a linearization analysis and must not be 2 unless
FAST is interfaced with Simulink. Using values other than 0, 1, or 2 will cause FAST to abort.
(switch)
TYCOn The time to enable active nacelle yaw control. This parameter is used only if YCMode is set to a
non-zero value. TYCOn must not be negative and must equal zero when YCMode is 2. Please
see the Controls chapter for further details. (sec)
PCMode This is the pitch-control-mode switch for user-defined pitch control. Setting it to 0 disables user-
defined pitch control. Setting it to 1 causes FAST to call a user-written routine called PitchCntrl()
at every time step past TPCOn. A real pitch-control routine created by Craig Hansen is linked
with FAST, but we supply a dummy routine in the software folder to help you write your own.
Setting PCMode to 2 causes FAST to accept pitch demands externally from Simulink. The
simple pitch maneuvers described below override the control setting determined by the user-
supplied pitch controllers. Please see the Controls chapter for further details. PCMode must be 0
during a linearization analysis and must not be 2 unless FAST is interfaced with Simulink. Using
values other than 0, 1, or 2 will cause FAST to abort. (switch)

FAST User's Guide 56 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Turbine Control (continued)

TPCOn The time to enable active pitch control. This parameter is used only if PCMode is set to a non­
zero value. TPCOn must not be negative and must equal zero when PCMode is 2. Please see the
Controls chapter for further details. (sec)
VSContrl This switch determines whether the generator torque is actively controlled for variable speed
machines. The generator DOF flag, GenDOF, must be enabled to use this feature. Setting
VSContrl to 0 will cause FAST to use one of the generator models defined by GenModel below
to determine the generator torque. A setting of 1 for VSContrl will invoke a simple variable-
speed model that uses the next four input parameters to determine the generator torque. Setting
VSContrl to 2 will enable a user-written routine, UserVSCont(), to determine the generator
torque. A sample routine is included in the file UserVSCont_KP.f90 and a dummy placeholder is
available (and commented out) in UserSubs.f90. Setting VSContrl to 3 causes FAST to accept
generator torque and electrical power demands externally from Simulink. VSContrl must not be 3
unless FAST is interfaced with Simulink. Using values other than 0, 1, 2, or 3 will cause FAST to
abort. Please see the Controls chapter for further details. (switch)
VS_RtGnSp The simple variable-speed control changes from the Region 2½ (linear torque versus speed
transition) to Region 3 (constant-torque control) at this generator speed (HSS speed). See Figure
22 for details. This value must not be less than zero, but it is ignored if VSContrl is not equal to
1. (rpm)
VS_RtTq This is the constant (or rated) torque applied to the HSS by the generator in Region 3 for the
simple variable-speed controller. See Figure 22 for details. This value must not be less than zero,
but it is ignored if VSContrl is not equal to 1. (N·m)
VS_Rgn2K When in Region 2 for the simple variable-speed controller, the generator speed is squared and
multiplied by VS_Rgn2K to compute the generator torque to apply to the HSS. See Figure 22 for
details. This value must not be less than zero, but it is ignored if VSContrl is not equal to 1.
(N·m/rpm2)
VS_SlPc This is the rated generator slip percentage in the linear torque versus speed transition Region 2½
for the simple variable-speed controller. Similar to the simple induction generator input parameter
SIG_SlPc, input VS_SlPc should be computed as the difference between the rated and the
equivalent synchronous generator speed, divided by the equivalent synchronous speed, and then
converted to percent. See Figure 22 for details. This value must greater than zero, but it is
ignored if VSContrl is not equal to 1. (%)
GenModel This switch determines which generator model is used when GenDOF is enabled and VSContrl is
set to 0. Setting it to 1 enables the simple-induction-generator model, whose parameters are
defined in the Simple-Induction-Generator section below. Setting it to 2 enables the Thevenin­
equivalent induction-generator model, whose parameters are defined in the Thevenin-Equivalent-
Induction-Generator section below. Setting it to 3 will cause FAST to call the user-written
subroutine, UserGen(). A UserGen() routine (found in UserVSCont_KP.f90), which calls
routine UserVSCont() as if VSContrl was set to 2, is normally linked with the program and a
dummy placeholder of UserGen() is also available (and commented out) in source file
UserSubs.f90. In order to define your own generator model, you will need to write your own
routine to replace it to use this option. Using values other than 1, 2, or 3 will cause FAST to abort.
(switch)
GenTiStr This flag determines whether the generator is brought online at a specific time (TimGenOn) or a
specific generator speed (SpGenOn). To use this feature, you must enable the generator DOF
(GenDOF). GenTiStr must be True and TimGenOn must be 0.0 during a linearization analysis
(AnalMod = 2) or when VSContrl is set to 3. (flag)
GenTiStp This flag determines whether the generator is taken offline at a specific time (TimGenOf) or when
power falls to zero. To use this feature, you must enable the generator DOF (GenDOF).
GenTiStp must be True and TimGenOf must greater than TMax during a linearization analysis
(AnalMod = 2) or when VSContrl is set to 3. (flag)
SpGenOn If GenTiStr is False, the generator will switch on and stay on once the HSS speed reaches
SpGenOn. This is used to do a speed startup of the generator. It applies to all generator models,
including user-defined and variable-speed control. (rpm)

FAST User's Guide 57 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Turbine Control (continued)

TimGenOn If GenTiStr is True, the generator will switch on at time TimGenOn. This is used to do a timed
startup of the generator. It applies to all generator models, including user-defined and variable-
speed control. GenTiStr must be True and TimGenOn must be 0.0 during a linearization
analysis (AnalMod = 2) or when VSContrl is set to 3. (sec)
TimGenOf This parameter determines the time to turn off the generator when GenTiStp is True. This can be
used for a shutdown maneuver or to simulate a loss of grid. This value is used whether you do a
timed or speed-based startup. Normally, you won’t simulate a startup and a loss of grid (or shut­
down) in the same run. In those cases, you will probably want to set this value to be greater than
TMax. When you do want to use TimGenOf, you will probably want to enable GenTiStr and set
TimGenOn to 0. TimGenOf must be greater than or equal to TimGenOn if GenTiStr is
enabled. This parameter is not used if GenTiStp is False. In that case, the generator is taken
offline when the power drops to zero. GenTiStp must be True and TimGenOf must greater than
TMax during a linearization analysis (AnalMod = 2) or when VSContrl is set to 3. (sec)
HSSBrMode This switch determines which HSS brake model is used when GenDOF is enabled. Setting it to 1
enables the simple built-in HSS brake torque with a linear ramp-up from zero to HSSBrTqF over
time HSSBrDT. Setting it to 2 causes FAST to call a user-written routine called UserHSSBr() at
every time step past THSSBrDp. We supply a dummy routine in the software folder to help you
write your own. Please see the Controls chapter for further details. Using values other than 1 or 2
will cause FAST to abort. (switch)
THSSBrDp At this time, the HSS brake will be deployed. In the simple model (HSSBrMode = 1), the
braking torque will start its linear ramp to full torque, which happens after HSSBrDT seconds. In
the user-defined model (HSSBrMode = 2), routine UserHSSBr() determines the fraction of full
braking torque after deployment. You will probably want to turn the generator off a short time
before this with TimGenOf before starting the HSS brake maneuver. THSSBrDp must greater
than TMax during a linearization analysis (AnalMod = 2) or when FAST is interfaced with
Simulink. (sec)
TiDynBrk The dynamic generator brake engages at this time. This input is CURRENTLY IGNORED since
logic for the dynamic generator brake is not currently coded in FAST. TiDynBrk must greater
than TMax during a linearization analysis. (sec)
TTpBrDpi The ith tip brake will start to deploy at this time. The drag constant for this brake will start to ramp
up from TBDrConN to TBDrConD. You can specify different times for different brakes to
simulate such conditions as one brake accidentally deploying or a situation in which one brake
fails to deploy when commanded to do so. TTpBrDpi must greater than TMax during a
linearization analysis. Only the first two values are used for two-bladed turbines. (sec)
TBDepISpi The ith tip brake will start to deploy when the rotor speed reaches TBDepISpi. Up to this point,
the drag constant for the tip brake is TBDrConN. Once the tip brake starts to deploy, it will take
TpBrDT seconds to reach full deployment, where it will remain deployed and using TBDrConD
for the drag constant. During the TpBrDT second deployment, the drag constant for the tip brake
will have an s-shaped ramp from TBDrConN to TBDrConD. TBDeplSpi must much greater than
RotSpeed during a linearization analysis. Only the first two values are used for two-bladed
turbines. (rpm)
TYawManS With or without yaw control or the yaw DOF enabled, after time TYawManS, the nacelle will
yaw to NacYawF using a linear ramp from its current value at TYawManS until TYawManE. If
yaw control is enabled when YCMode is not 0, the yaw commands determined from inputs
TYawManS, TYawManE, and NacYawF override whatever commands come from the yaw
controller. Also, the yaw commands determined from inputs TYawManS, TYawManE, and
NacYawF pass through FAST’s built-in second-order actuator model if the yaw DOF is enabled
when YawDOF is set to True. You can use TYawManS and TYawManE to simulate a yaw for
startup, shutdown, or runaway yaw event. For a fixed-yaw simulation, set YawDOF to False,
YCMode to 0, TYawManS greater than TMax, and NacYaw to the fixed nacelle yaw angle.
TYawManS must greater than TMax during a linearization analysis. (sec)
TYawManE The nacelle yaw command will hold at a constant a setting of NacYawF from this time until the
end of the run. TYawManE must be set larger or equal to TYawManS. (sec)
FAST User's Guide 58 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Turbine Control (concluded)

NacYawF The nacelle yaw command will hold at a constant setting of NacYawF from TYawManE until the
end of the run. (degrees)
TPitManSi With or without pitch control enabled, after time TPitManSi, the ith blade will pitch to BlPitchFi
using a linear ramp from its current value at TPitManSi until TPitManEi. If pitch control is
enabled when PCMode is not 0, the pitch commands determined from inputs TPitManSi,
TPitManEi, and BlPitchFi override whatever commands come from the pitch controller. You can
use TPitManSi and TPitManEi to simulate a pitch for startup, shutdown, or runaway pitch event.
By setting one blade different from the other(s), you can simulate a fault condition in which one
blade unexpectedly pitches or fails to pitch. For a constant-pitch simulation, set PCMode to 0,
TPitManSi greater than TMax, and BlPitchi to the fixed blade pitch angles. TPitManSi must
greater than TMax during a linearization analysis. Only the first two values are used for two-
bladed turbines. (sec)
TPitManEi The ith blade will hold at a constant a setting of BlPitchFi from this time until the end of the run.
TPitManEi must be set larger or equal to TPitManSi. Only the first two values are used for two-
bladed turbines. (sec)
BlPitchi When PCMode is 0 during a time-marching analysis or if TrimCase is 1 or 2 while computing a
steady state solution during a linearization analysis, the ith blade will hold at a constant setting of
BlPitchi until TPitManSi. If PCMode is not 0 or if TrimCase is 3 while computing a steady
state solution during a linearization analysis, BlPitchi is the initial pitch. The pitch angle is
relative to the chord line at the point of zero aerodynamic twist and is positive towards feather
(leading edge upwind). These values must be greater than –180 and less than or equal to 180
degrees. Only the first two values are used for two-bladed turbines. (deg)
BlPitchFi The ith blade will hold at a constant setting of BlPitchFi from TPitManEi until the end of the run.
This is relative to the chord line at the point of zero aerodynamic twist and is positive towards
feather (leading edge upwind). Only the first two values are used for two-bladed turbines. (deg)

Environmental Conditions
Gravity The gravitational acceleration constant. (m/sec2)

Feature Flags*
FlapDOF1 The first flapwise blade-bending mode will be enabled when this flag is True. When enabled, you
should ensure that the corresponding mode shape specified in the blade input files is accurate. For
ADAMS datasets extracted from FAST, this flag is used to enable or disable blade flexibility and
its value must be identical to that of EdgeDOF. (flag)
FlapDOF2 The second flapwise blade-bending mode will be enabled when this flag is True. It is possible to
enable the second mode without enabling the first mode, but it should be done only for research
purposes. When enabled, you should ensure that the corresponding mode shape specified in the
blade input files is accurate. The value of this input does not effect the creation of ADAMS
datasets. (flag)
EdgeDOF The first edgewise blade-bending mode will be enabled when this flag is True. When enabled,
you should ensure that the corresponding mode shape specified in the blade input files is accurate.
For ADAMS datasets extracted from FAST, this flag is used to enable or disable blade flexibility
and its value must be identical to that of FlapDOF1. (flag)
TeetDOF This flag enables rotor teetering when set to True. If this option is disabled, teeter can be set to
any fixed angle. This flag is ignored for three-bladed turbines. (flag)

*
You must enable at least one DOF during a linearization analysis (AnalMode set to 2). During a time-marching analysis
(AnalMode set to 1), there is no restriction.

FAST User's Guide 59 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Feature Flags (concluded)

DrTrDOF If set to True, this flag will enable torsional flexibility of the drivetrain. This models the drivetrain
between the generator and rotor as a lumped torsion spring and damper. (flag)
GenDOF The generator DOF will be enabled when this flag is True. This allows use of the various
variable-speed control schemes and generator models during a time-marching analysis or a trim
solution during a linearization analysis. If False, the HSS will rotate at a fixed rate
(RotSpeed*GBRatio). (flag)
YawDOF When set to True, this flag enables the nacelle yaw DOF. The initial nacelle yaw angle is
specified with NacYaw. If YawDOF is disabled, the yaw angle will be fixed at NacYaw. (flag)
TwFADOF1 The first tower fore-aft bending mode will be enabled when this variable is set to True. When en­
abled, you should ensure that the corresponding mode shape specified in the tower input file is ac­
curate. For ADAMS datasets extracted from FAST, this flag is used to enable or disable tower
flexibility and its value must be identical to that of TwSSDOF1. (flag)
TwFADOF2 The second tower fore-aft bending mode will be enabled when this variable is set to True. Except
for research purposes, this flag should be set to True only if TwFADOF1is True. When enabled,
you should ensure that the corresponding mode shape specified in the tower input file is accurate.
The value of this input does not effect the creation of ADAMS datasets. (flag)
TwSSDOF1 The first tower side-to-side bending mode will be enabled when this variable is set to True. When
enabled, you should ensure that the corresponding mode shape specified in the tower input file is
accurate. For ADAMS datasets extracted from FAST, this flag is used to enable or disable tower
flexibility and its value must be identical to that of TwFADOF1. (flag)
TwSSDOF2 The second tower side-to-side bending mode will be enabled when this variable is set to True. Ex­
cept for research purposes, this flag should be set to True only if TwSSDOF1 is True. When en­
abled, you should ensure that the corresponding mode shape specified in the tower input file is ac­
curate. The value of this input does not effect the creation of ADAMS datasets. (flag)
CompAero This flag determines whether aerodynamic loads will be computed using the AeroDyn
aerodynamic modules. If False, the simulation will occur in a vacuum (no airloads). AeroDyn
input properties are specified in the ADFile (see input below). The ADFile must exist even if
CompAero is False, since RNodes determines the location of the structural analysis points.
(flag)
CompNoise A series of semi-empirical aeroacoustic noise prediction algorithms has been incorporated into
FAST by Pat Moriarty of NREL/NWTC. The algorithms predict six different forms of
aerodynamically produced noise including turbulent inflow, turbulent boundary layer trailing
edge, separating flow, laminar boundary layer vortex shedding, trailing edge bluntness vortex
shedding, and tip vortex formation. These noise sources are then superimposed to calculate and
output the total aerocoustic signature of an operating wind turbine. CURRENTLY, THE NOISE
PREDICTION INTERFACE IS NOT DOCUMENTED IN THIS GUIDE (except as it effects the
primary-input-file). Details on the contents and validation of the aeroacoustic noise prediction
models are provided in [19]. Questions related to the noise prediction models and interface should
be directed to Pat Moriarty, preferably at night or on weekends ☺.
The CompNoise flag determines whether aerodynamic noise will be computed. Aerodynamic
noise input properties are specified in the NoiseFile (see input below). CompAero must be
enabled if CompNoise is enabled. CompNoise must be False during a linearization analysis.
The value of this input does not effect the creation of ADAMS datasets since the aeroacoustic
algorithms are not linked to ADAMS. (flag)

FAST User's Guide 60 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Initial Conditions
OoPDefl This is the initial, out-of-plane, blade-tip displacement. The same value is used for all blades.
Note that by specifying values for initial conditions close to the steady-state conditions, the
numerical solution technique will reach trimmed conditions faster. It is positive downwind. It is
possible to specify combinations of tip displacements that are not meaningful for the blade
structural pretwist distribution and the DOFs that are enabled. If so, FAST will issue a warning
message and choose meaningful values for you. The value of OoPDefl must be zero (no initial
deflection) when creating ADAMS datasets or when FAST is interfaced with Simulink. (m)
IPDefl This is the initial, in-plane, blade-tip displacement. The same value is used for all blades. Note
that by specifying values for initial conditions close to the steady-state conditions, the numerical
solution technique will reach “trimmed conditions” faster. It is positive clockwise when looking
upwind. It is possible to specify combinations of tip displacements that are not meaningful for the
blade structural pretwist distribution and the DOFs that are enabled. If so, FAST will issue a
warning message and choose meaningful values for you. The value of IPDefl must be zero (no
initial deflection) when creating ADAMS datasets or when FAST is interfaced with Simulink. (m)
TeetDefl This is the initial or fixed teeter angle. It is positive when Blade 1 is deflected downwind of the
rotor. This value must be greater than –180 and less than or equal to 180 degrees and must be zero
when creating ADAMS datasets. This parameter is ignored for three-bladed turbines. (deg)
Azimuth This is the initial azimuth angle for Blade 1. Please note that for three-bladed rotors, blade 3 is
ahead of blade 2, which is ahead of blade 1, so that the order of blades passing through a given
azimuth is 3-2-1-repeat. Azimuth works in conjunction with AzimB1Up, which is input in the
Turbine Configuration section that follows. This value must be greater or equal to 0 and less than
360 degrees. (deg)
RotSpeed This is the initial angular speed of the rotor. During a linearization analysis, this is also the
desired azimuth-average rotor speed for a trim solution. This value must not be negative. The
turbine rotates clockwise when looking downwind. (rpm)
NacYaw This is the initial or fixed nacelle yaw angle. It is positive counterclockwise when looking down
on the turbine. This value must be greater than –180 and less than or equal to 180 degrees. (deg)
TTDspFA This is the initial fore-aft tower-top displacement. It is positive downwind. The value of
TTDspFA must be zero (no initial deflection) when creating ADAMS datasets or when FAST is
interfaced with Simulink. (m)
TTDspSS This is the initial side-to-side tower-top displacement. It is positive to the right when looking
upwind. The value of TTDspSS must be zero (no initial deflection) when creating ADAMS
datasets or when FAST is interfaced with Simulink. (m)

Turbine Configuration
TipRad The blade-tip radius is the distance from the apex of the cone of rotation to the blade tip along the
pitch axis instead of the perpendicular distance from the axis of rotation. See Figure 14(a) and
Figure 16. This value must be greater than zero. (m)
HubRad The hub radius is the distance from the apex of the cone of rotation to the blade root along the
pitch axis instead of the perpendicular distance from the axis of rotation. The blade root loads are
defined at this radial span location. See Figure 14(b) and Figure 16. This value must be greater
than or equal to zero and less than TipRad. (m)
PSpnElN This is the blade element number corresponding to the innermost blade element that is part of the
pitchable portion of the blade for partial-span pitch control. The pitch of all the blade elements
from PSpnElN to BldNodes are controlled by the BlPitch(:) array, whereas all the blade
elements from 1 to (PSpnElN - 1) are not pitchable. Note that PSpnElN is CURRENTLY
IGNORED by FAST; that is, the logic for partial-span pitch control has not yet been codified in
FAST. This value must be an integer between 1 and BldNodes (inclusive). (-)
UndSling The undersling is the distance from the teeter pin to the apex of the cone of rotation. It is positive
upwind. This parameter is ignored for three-bladed turbines. See Figure 14(b). (m)
HubCM This is the distance from the rotor apex to the hub mass center. It is positive downwind. See
Figure 14(b) and Figure 16. (m)
FAST User's Guide 61 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Turbine Configuration (concluded)

OverHang This is the distance along the rotor shaft from the yn-/zn-plane to the teeter pin for two-bladed
turbines or from the yn-/ zn-plane to the rotor apex for three-bladed turbines. It is positive
downwind, so use a negative number for upwind turbines. For turbines with rotor-furl, this
distance defines the configuration at a furl angle of zero. See Figure 14(a), Figure 16, and Figure
18. (m)
NacCMxn This is the downwind distance to the nacelle mass center (reference input NacMass) from the top
of the tower, measured parallel to the xn-axis. It is positive downwind. See Figure 14(a) and
Figure 16. (m)
NacCMyn This is the lateral distance to the nacelle mass center (reference input NacMass) from the top of
the tower, measured parallel to the yn-axis. It is positive to the left when looking downwind or
positive into the page of Figure 14(a) and Figure 16. (m)
NacCMzn This is the vertical distance to the nacelle mass center (reference input NacMass) from the top of
the tower, measured parallel to the zn-axis. It is positive upward when looking downwind. See
Figure 14(a) and Figure 16. (m)
TowerHt The tower height is the distance from ground level [onshore] or mean sea level [offshore] to the
top of the tower and yaw bearing. This value must be greater than zero. See Figure 14(a), Figure
16, and Figure 20. (m)
Twr2Shft This is the vertical distance from the top of the tower and yaw bearing to the intersection of the
rotor shaft axis and the yn-/zn-plane. The distance is measured parallel to the zn-axis. See Figure
14(a), Figure 16, and Figure 18. The combination of TipRad, TowerHt, Twr2Shft, OverHang,
and ShftTilt must ensure that the blade tip does not hit the ground. This value also cannot be
negative. For turbines with rotor-furl, this distance defines the configuration at a furl angle of
zero. (m)
TwrRBHt The tower rigid base height is the distance from tower base to the beginning of the flexible portion
of the tower. The tower base loads are defined at this elevation. This value must be greater or
equal to zero and less than TowerHt + TwrDraft. (m)
ShftTilt This is the tilt angle of the rotor shaft from the nominally horizontal plane. Positive tilt means that
the downwind end of the shaft is the highest. This value must be between –90 and 90 degrees.
Upwind turbines have negative tilt for improved tower clearance. For turbines with rotor-furl, this
angle defines the configuration at a furl angle of zero. See Figure 14(a), Figure 16, and Figure 18.
(deg)
Delta3 This teeter pin orientation angle allows coupling between the flapping due to teeter and blade
pitch. A positive value means that the blade that teeters downwind has a positive change in pitch
(leading edge upwind). See Figure 15 and the discussion on page 13 for details. This value must
be between –90 and 90 degrees (exclusive) and is ignored for three-bladed turbines. (deg)
PreConei The coning angle for the ith blade is positive downwind for upwind and downwind rotors. See
Figure 14(a) and Figure 16. These values must be greater than –180 and less than or equal to 180
degrees. Only the first two values are used for two-bladed turbines. (deg)
AzimB1Up All input and output azimuth values are measured with respect to this number. If this value is 0
and the rotor azimuth is 180 degrees, then Blade 1 is pointing down. Please keep in mind that the
rotor rotates clockwise when looking downwind, so an azimuth value of 90 degrees means Blade 1
is pointing to the right (looking downwind) when AzimB1Up is 0. If AzimB1Up is 270 degrees
and Azimuth is 0 degrees, then Blade 1 is to the right when looking downwind. Also note that for
three-bladed rotors, blade 3 is ahead of blade 2, which is ahead of blade 1, so that the order of
blades passing through a given azimuth is 3-2-1-repeat. (deg)

FAST User's Guide 62 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Mass and Inertia

YawBrMass This is the mass of the yaw bearing. Its center is located at the top of the tower, at the origin of the
tower-top/base-plate and nacelle/yaw coordinate systems. This value must not be negative. (kg)
NacMass This is the mass of the nacelle. The center of the nacelle mass is located at the point specified by
inputs NacCMxn, NacCMyn, and NacCMzn relative to the tower-top. It includes everything
atop the tower excluding the rotor (blades, hub, and tip brakes), yaw bearing, and systems that furl
(tail boom, tail fin, and structure furling with the rotor). This value must not be negative. (kg)
HubMass This is the mass of the hub. Its center is located a distance of HubCM from the rotor apex. This
value must not be negative. (kg)
TipMassi This is the tip-brake mass for the ith blade. This value must not be negative. Only the first two
values are used for two-bladed turbines. (kg)
NacYIner This is the nacelle moment of inertia about the yaw axis. It includes all mass contained in
NacMass. This value must be greater than NacMass•( NacCMxn2 + NacCMyn2 ). (kg·m2)
GenIner This is the moment of inertia of the high-speed portion of the drivetrain including the gearbox,
HSS, and generator. If torsional flexibility of the drivetrain is enabled, the compliance is between
the inertia of the rotor and this inertia. This value will be multiplied by the square of the gear ratio
to map it to the low-speed reference frame. This value must not be negative. (kg·m2)
HubIner The hub moment of inertia is measured about the teeter axis for two-bladed turbines or about the
rotor shaft axis for three-bladed turbines. For two-bladed turbines, it includes those parts that
teeter, except for the blades and tip brakes, and must be greater than HubMass•( UndSling –
HubCM )2. For three-bladed turbines, it excludes the blades and tip brakes and must not be
negative. (kg·m2)

Drivetrain
GBoxEff The gearbox efficiency is the ratio of the output shaft power to the input shaft power. Enter it as a
percentage from 0 to 100. The value of GboxEff must be 100 (no mechanical losses) when
creating ADAMS datasets. (%)
GenEff The generator efficiency is the ratio of its output power to its input power. It is used by the
simple-induction-generator model (GenModel = 1) to obtain the electrical power from the
mechanical power, which is a product of the generator torque and HSS speed. It is also used by
the simple variable-speed, generator torque controller (VSContrl = 1) in the same manor. Enter it
as a percentage from 0 to 100. GenEff is ignored by the Thevenin-equivalent induction-generator
model (GenModel = 2), which incorporates a more complex expression for the electrical power
based on the input circuit resistances. The value of GenEff is passed to UserGen() and
UserVSCont() for the user-defined generator model (GenModel = 3) and user-defined variable-
speed, generator torque controller (VSContrl = 2) respectively, but the user-defined models allow
for the flexibility of implementing any relationship between input and output power. (%)
GBRatio This is the ratio of the HSS speed to the LSS speed. This value must be greater than zero and
should be 1.0 for a direct-drive turbine. (-)
GBRevers Set this value to True if the direction of rotation of the LSS is opposite that of the HSS.
GBRevers must be set to False when creating ADAMS datasets. (flag)
HSSBrTqF This maximum mechanical brake torque value is applied to the HSS end of the drivetrain
compliance. This value must not be negative. It is used by both the simple (HSSBrMode = 1)
and user-defined (HSSBrMode = 2) HSS brake models. (N·m)
HSSBrDT For the simple HSS brake model (HSSBrMode = 1), this is the amount of time it takes the HSS
brake to reach full torque once it is applied. The ramp from off to full torque is linear. This value
must not be negative and is unused when HSSBrMode is set to 2. (sec)
DynBrkFi This is name of a file containing a curve of mechanical generator torques versus HSS speeds
defining the dynamic generator brake characteristics. The name may optionally include an
absolute or relative path. This file name must contain fewer than 100 characters and must be
enclosed in apostrophes or double quotes. This input is CURRENTLY IGNORED since logic for
the dynamic generator brake is not currently coded in FAST. (-)

FAST User's Guide 63 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Drivetrain (concluded)
DTTorSpr The equivalent drive-train torsional spring constant includes compliance in the LSS, the gearbox,
and the HSS. This value must not be negative. (N·m/rad)
DTTorDmp The equivalent drive-train torsional damping constant includes compliance in the LSS, the
gearbox, and the HSS. This value must not be negative. (N·m/sec)

Simple Induction Generator

SIG_SlPc The rated generator slip percentage is the difference between the rated and the synchronous
generator speed divided by the synchronous generator speed, and then converted to percent. See
Figure 12 for details. This value must be greater than zero, but it is ignored if GenModel is not
equal to 1 or VSContrl is not equal to 0. (%)
SIG_SySp This is the synchronous or zero-torque generator speed. See Figure 12 for details. This value
must be greater than zero, but it is ignored if GenModel is not equal to 1 or VSContrl is not equal
to 0. (rpm)
SIG_RtTq This is the torque supplied by the generator when running at rated speed. See Figure 12 for de­
tails. This value must be greater than zero, but it is ignored if GenModel is not equal to 1 or
VSContrl is not equal to 0. (N·m)
SIG_PORt The pullout ratio is the ratio of the pullout torque and the rated torque. The negative of this value
is also used for the startup torque. See Figure 12 for details. This value must be greater than or
equal to one, but it is ignored if GenModel is not equal to 1 or VSContrl is not equal to 0. (-)

Thevenin-Equivalent, 3-Phase, Induction Generator

TEC_Freq This is the line frequency of the electrical grid. This value must be greater than zero and should be
50 (Europe) or 60 (U.S.), but it is ignored if GenModel is not equal to 2 or VSContrl is not equal
to 0. (Hz)
TEC_NPol This is the number of poles in the generator. This value must be an even integer greater than zero,
but it is ignored if GenModel is not equal to 2 or VSContrl is not equal to 0. (-)
TEC_SRes This is the resistance of the generator stator in the complete circuit. This value must be greater
than zero, but it is ignored if GenModel is not equal to 2 or VSContrl is not equal to 0. (ohms)
TEC_RRes This is the resistance of the generator rotor in the complete circuit. This value must be greater
than zero, but it is ignored if GenModel is not equal to 2 or VSContrl is not equal to 0. (ohms)
TEC_VLL This is the line-to-line voltage of the generator. This value must be greater than zero and is often
690 in Europe or 480 or 575 in the U.S., but it is ignored if GenModel is not equal to 2 or
VSContrl is not equal to 0. (volts)
TEC_SLR This is the leakage reactance of the generator stator in the complete circuit. This value must be
greater than zero, but it is ignored if GenModel is not equal to 2 or VSContrl is not equal to 0. It
is usually a small number and is close in value to the stator resistance. (ohms)
TEC_RLR This is the leakage reactance of the generator rotor in the complete circuit. This value must be
greater than zero, but it is ignored if GenModel is not equal to 2 or VSContrl is not equal to 0. It
is usually a small number and is close in value to the rotor resistance. (ohms)
TEC_MR This is the magnetizing reactance of the complete generator circuit. This value must be greater
than zero, but it is ignored if GenModel is not equal to 2 or VSContrl is not equal to 0. It is
usually about 10-50 times greater than the leakage reactances. (ohms)

FAST User's Guide 64 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Platform Model
PtfmModel This is a switch used to indicate the type of support platform and to tell FAST whether or not to
read in an additional file of inputs for defining the model properties of the support platform (see
next input PtfmFile). The additional inputs in PtfmFile pertain to the to the support platform
configuration, motions, and loading. Setting PtfmModel to 1 specifies an onshore foundation.
Setting it to 2 specifies a fixed bottom offshore foundation. Setting it to 3 specifies a floating
offshore configuration. Setting PtfmModel to 0 disables the platform models—in this case, FAST
will rigidly attach the tower to the inertia frame (ground) through a cantilever connection. Using
values other than 0, 1, 2, or 3 will cause FAST to abort.
PtfmFile This is the name of the file that contains additional model properties for the support platform. The
name may optionally include an absolute or relative path. This file name must contain fewer than
100 characters and must be enclosed in apostrophes or double quotes. FAST will only read this
file if PtfmModel is nonzero. See Table 12 for a listing of input parameters contained in this file.
In FAST v6.0, all nonzero PtfmModel options will work the same way by reading in PtfmFile. In
future versions, the format of this file will depend on which PtfmModel option is selected.
(quoted string)

Tower
TwrNodes The tower is divided into TwrNodes equal-length segments. The nodes at the centers of these
segments are used for the integration of elastic forces. The more segments you use, the more
accurate the integral will be, but the greater the computational time will be. A good compromise
for this parameter is 20. This value must be an integer greater than 0. When creating ADAMS
datasets, this value must be no more than 99. (-)
TwrFile This is the name of the file that contains the tower properties. The name may optionally include
an absolute or relative path. This file name must contain fewer than 100 characters and must be
enclosed in apostrophes or double quotes. FAST will read this file even when there are no tower
DOFs. See Table 9 for a listing of input parameters contained in this file. (quoted string)

Nacelle Yaw
YawSpr This is the torsional spring stiffness in FAST’s built-in, second-order, nacelle yaw actuator model.
The linear nacelle-yaw spring moment is proportional to the nacelle-yaw error through this
constant. If a yaw actuator natural frequency is known in place of an actuator spring stiffness,
compute the spring stiffness as follows: YawSpr = YawIner•ωn2, where ωn is the natural
frequency in rad/sec and YawIner is the nominal inertia of the nacelle, rotor, and tail about the
yaw axis in kg·m2. This value must not be negative. (N·m/rad)
YawDamp This is the torsional damping constant in FAST’s built-in, second-order, nacelle yaw actuator
model. The linear nacelle-yaw damping moment is proportional to the nacelle-yaw rate error
through this constant. If a yaw actuator natural frequency and damping ratio are known in place of
an actuator damping constant, compute the damping constant as follows: YawDamp =
2•ζ•YawIner•ωn, where ωn is the natural frequency in rad/sec, ζ is the damping ratio in fraction of
critical, and YawIner is the nominal inertia of the nacelle, rotor, and tail about the yaw axis in
kg·m2. This value must not be negative. (N·m/(rad/sec))
YawNeut When YCMode is 0, this is the neutral nacelle yaw position (constant yaw command) as described
on page 12. When YCMode is not zero, this is the initial, constant yaw command before active
yaw control is enabled at time TYCOn. This value must be greater than –180 and less than or
equal to 180 degrees. (deg)

FAST User's Guide 65 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Furling
Furling This flag is used to tell FAST whether or not to read in an additional file of inputs for defining the
model configuration of a furling turbine (see next input FurlFile). The additional inputs in
FurlFile pertain to the lateral offset and skew angle of the rotor shaft, rotor-furling, tail-furling,
and tail inertia and aerodynamics. If the turbine you want to model contains any of these
characteristics, you must assemble the furling input file even if your turbine does not “furl” in the
common sense of the word. For example, if the turbine you want to model contains a tail, you
must assemble the furling input file regardless of whether or not your tail, or rotor, actively furls
about the yawing-portion of the structure atop the tower. (flag)
FurlFile This is the name of the file that contains additional model properties for a furling turbine. The
name may optionally include an absolute or relative path. This file name must contain fewer than
100 characters and must be enclosed in apostrophes or double quotes. FAST will only read this
file if the model is designated as a furling machine (when Furling is set to True). See Table 13 for
a listing of input parameters contained in this file. (quoted string)

Rotor Teeter
TeetMod The teeter springs and dampers can be modeled three ways. For a value of 0 for TeetMod, there
will be no teeter spring nor damper and the moment normally produced will be set to zero. A
TeetMod of 1 will invoke simple spring and damper models using the inputs provided below as
appropriate coefficients. If you set TeetMod to 2, FAST will call the routine UserTeet() to
compute the teeter spring and damper moments. You should replace the dummy routine supplied
with the code with your own, which will need to be linked with the rest of FAST. Using values
other than 0, 1, or 2 will cause FAST to abort. This parameter is ignored for three-bladed turbines.
(switch)
TeetDmpP The teeter damper is effective when the teeter deflection exceeds this value. This value must be
between 0 and 180 degrees (inclusive). This parameter is ignored for three-bladed turbines and
when TeetMod is not set to 1. (deg)
TeetDmp The linear teeter damping moment is proportional to the teeter rate through this constant and is
effective when the teeter deflection exceeds TeetDmpP. This value must not be negative. This
parameter is ignored for three-bladed turbines and when TeetMod is not set to 1. (N·m/(rad/sec))
TeetCDmp The Coulomb-friction damping moment resists teeter motion, but it is a constant that is not
proportional to the teeter rate. However, if the teeter rate is zero, the damping is zero. This value
must not be negative. This parameter is ignored for three-bladed turbines and when TeetMod is
not set to 1. (N·m)
TeetSStP The teeter soft-stop spring is effective when the teeter deflection exceeds this value. This value
must be between 0 and 180 degrees (inclusive). This parameter is ignored for three-bladed
turbines and when TeetMod is not set to 1. (deg)
TeetHStP The teeter hard-stop spring is effective when the teeter deflection exceeds this value. This value
must be between TeetSStP and 180 degrees (inclusive). This parameter is ignored for three-
bladed turbines and when TeetMod is not set to 1. (deg)
TeetSSSp The linear teeter soft-stop spring restoring moment is proportional to the teeter soft-stop deflection
by this constant and is effective when the teeter deflection exceeds TeetSStP. This value must
not be negative. This parameter is ignored for three-bladed turbines and when TeetMod is not set
to 1. (N·m/rad)
TeetHSSp The linear teeter hard-stop spring restoring moment is proportional to the teeter hard-stop
deflection by this constant and is effective when the teeter deflection exceeds TeetHStP. This
value must not be negative. This parameter is ignored for three-bladed turbines and when
TeetMod is not set to 1. (N·m/rad)

FAST User's Guide 66 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Tip Brakes
TBDrConN When tip brakes are not deployed (normal operation), this value is multiplied by the dynamic
pressure to produce the drag of the brakes at the tip of every blade. This value is Cd times the flat
plate drag area. This value must not be negative. (m2)
TBDrConD When tip brakes are deployed (braking operation), the tip drag follows an S curve from
TBDrConN to this fully deployed value. The resulting value is multiplied by the dynamic
pressure to produce the drag of the brakes at the tip of every blade. This value is Cd times the flat
plate drag area. This value must not be negative. (m2)
TpBrDT When tip brakes are deployed it takes TpBrDT seconds to fully deploy them. This value must not
be negative. (m2)

Blades
BldFilei This is the name of the file that contains the properties for the ith blade. The names may optionally
include an absolute or relative path. These file names must contain fewer than 100 characters and
must be enclosed in apostrophes or double quotes. Only the first two names are used for two-
bladed turbines. FAST will read this file even when there are no blade DOFs. See Table 10 for a
listing of input parameters contained in this file. Please note that for three-bladed rotors, blade 3 is
ahead of blade 2, which is ahead of blade 1, so that the order of blades passing through a given
azimuth is 3-2-1-repeat. (quoted string)

AeroDyn
ADFile This is the name of the file that contains the AeroDyn aerodynamics parameters. The name may
optionally include an absolute or relative path. This file name must contain fewer than 100
characters and must be enclosed in apostrophes or double quotes. FAST will read this file even
when aerodynamic calculations are disabled. See Table 11 for a listing of input parameters
contained in this file. (quoted string)

Noise
NoiseFile This is the name of the file that contains input parameters needed for aeroacoustic noise
predictions. The name may optionally include an absolute or relative path. This file name must
contain fewer than 100 characters and must be enclosed in apostrophes or double quotes. FAST
will not read in this file if aerodynamic noise is not computed (when CompNoise is False) and
during linearization analyses (when AnalMode is set to 2). Also, the inputs in this file do not
effect the creation of ADAMS datasets. (quoted string)

ADAMS
ADAMSFile This is the name of the file that contains input parameters needed only for creation of ADAMS
datasets. The name may optionally include an absolute or relative path. This file name must
contain fewer than 100 characters and must be enclosed in apostrophes or double quotes. FAST
will not read in this file if ADAMS datasets are not generated (when ADAMSPrep is set to 1).
See Table 14 for a listing of input parameters contained in this file. (quoted string)

FAST User's Guide 67 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (continued).

Linearization Control
LinFile This is the name of the file that contains FAST linearization input parameters. The name may
optionally include an absolute or relative path. This file name must contain fewer than 100
characters and must be enclosed in apostrophes or double quotes. FAST will not read in this file
for time-marching analyses, when a linearization analysis is not performed (when AnalMode is
set to 1) and the inputs in this file do not effect the creation of ADAMS datasets. See Table 15 for
a listing of input parameters contained in this file. (quoted string)

Output
SumPrint Set this value to True if you want FAST to generate the summary file (see Figure 32). (flag)
TabDelim Set this value to True if you want FAST to delimit the tabular output data with tabs instead of
using fixed-width columns. Tab-delimited files are easier to import into spreadsheets, and fixed-
column files are better for viewing with a text editor or for printing. (flag)
OutFmt FAST will use this string as the numerical format specifier for output of floating-point values.
The length of this string must not exceed 20 characters and must be enclosed in apostrophes or
double quotes. You may not specify an empty string. To ensure that fixed-width column data
align properly with the column titles, you should ensure that the width of the field is 10 characters.
Using an E, EN, or ES specifier will guarantee that you will never overflow the field because the
number is too big, but such numbers are harder to read. Using an F specifier will give you
numbers that are easier to read, but you may overflow the field. Please refer to any Fortran
manual for details for format specifiers. (quoted string)
TStart This tells the program how much simulation time should pass before outputting data to the tabular
output file. A delay of at least five seconds is advised to allow the transient effects associated with
starting from rest to damp out. This value must not be negative or greater than TMax. This
parameter is ignored during a linearization analysis. (sec)
DecFact This parameter sets the decimation factor for output. FAST will output data only once each
DecFact integration time steps. For instance, a value of 5 will cause FAST to generate output
only every fifth time step. This value must be an integer greater than zero. (-)
SttsTime This parameter represents the amount of simulation time between the status messages that are
displayed to the screen during the simulation. The messages show how much simulation time has
elapsed and estimate when the job will complete. This value must be greater than zero. The value
of this input does not effect the creation of ADAMS datasets. (sec)
NcIMUxn This is the downwind distance to the virtual nacelle inertial measurement unit (IMU) from the top
of the tower, measured parallel to the xn-axis. It is positive downwind. See Figure 14(a) and
Figure 16. (m)
NcIMUyn This is the lateral distance to the virtual nacelle inertial measurement unit (IMU) from the top of
the tower, measured parallel to the yn-axis. It is positive to the left when looking downwind or
positive into the page of Figure 14(a) and Figure 16. (m)
NcIMUzn This is the vertical distance to the virtual nacelle inertial measurement unit (IMU) from the top of
the tower, measured parallel to the zn-axis. It is positive upward when looking downwind. See
Figure 14(a) and Figure 16. (m)
ShftGagL The distance from the teeter pin (two blades) or rotor apex (three blades) to the shaft-moment
output station along the positive xs axis allows you to put a virtual strain gage anywhere you like
along the shaft. It is positive for upwind rotors. (m)

FAST User's Guide 68 Last updated August 2005 for version 6.0
 Table 8. Primary-Input-File Parameters (concluded).

Output (concluded)
NTwGages The number of strain-gage locations along the tower indicates the number of input values on the
next line. Valid values are integers from 0 to 5 (inclusive). (-)
TwrGagNd The virtual strain-gage locations along the tower are assigned to the tower analysis nodes specified
on this line. Possible values are 1 to TwrNodes (inclusive), where 1 corresponds to the node
closest to the tower base (but not at the base) and a value of TwrNodes corresponds to the node
closest to the tower top. The exact elevations of each analysis node in the undeflected tower,
relative to the base of the tower, are determined as follows:
Elev. of node J = TwrRBHt + ( J – ½ ) • [ ( TowerHt + TwrDraft – TwrRBHt ) / TwrNodes ]
(for J = 1,2,…,TwrNodes)
You must enter at least NTwGages values on this line. If NTwGages is 0, this line will be
skipped, but you must have a line taking up space in the input file. You can separate the values
with combinations of tabs, spaces, and commas, but you may use only one comma between
numbers. (-)
NBlGages The number of strain-gage locations along the blade indicates the number of input values on the
next line. Valid values are integers from 0 to 5 (inclusive). (-)
BldGagNd The virtual strain-gage locations along the blade are assigned to the blade analysis nodes specified
on this line. Possible values are 1 to BldNodes (inclusive), where 1 corresponds to the node
closest to the blade root (but not at the root) and a value of BldNodes corresponds to the node
closest to the blade tip. The radial span locations of the analysis nodes are determined by
AeroDyn input RNodes. You must enter at least NBlGages values on this line. If NBlGages is
0, this line will be skipped, but you must have a line taking up space in the input file. You can
separate the values with combinations of tabs, spaces, and commas, but you may use only one
comma between numbers. (-)
OutList For a time-marching analysis, this list of parameters determines what you want printed in the
output file. For a linearization analysis, this provides the list of output measurements. The line
containing the array name OutList is a comment line, which must then be followed by one or
more lines containing quoted strings that in turn contain one or more parameter names. Separate
the parameter names by any combination of commas, semicolons, spaces, and/or tabs. If you
prefix a parameter name with a minus sign, “-”, underscore, “_”, or the characters “m” or “M”,
FAST will multiply the value for that channel by –1 before writing the data. The parameters are
written in the order they are listed in the input file. You may include any parameter as many times
as you like. FAST allows you to use multiple lines so that you can break your list into meaningful
groups and so the lines can be shorter. However, you cannot have the strings within the quotes
longer than 1000 characters, so you are effectively limited to 100 channels per line in the input
file. The limit on the total number of output channels in all lines is 200. During time-marching
analyses, the simulation time will always be the first column in the output file and is not explicitly
entered in this list. You may enter comments after the closing quote on any of the lines. For
instance, you may want to group all of your blade loads together on one line and then comment the
line to document the fact that they are blade loads. Entering a line with the string “END” at the
beginning of the line or at the beginning of a quoted string found at the beginning of the line will
cause FAST to quit scanning for more lines of channel names. See Table 16 through Table 44 for
the list of possible parameters. (-)

FAST User's Guide 69 Last updated August 2005 for version 6.0
 Table 9. Tower-Input-File Parameters.
The following input parameters are contained in the file indicated by input TwrFile from the primary input file.
FAST will read this file even when there are no tower DOFs.

Tower
NTwInpSt The table of tower sectional data that follows below has NtwInpSt rows of data. The values in
this table will be interpolated to the TwrNodes analysis nodes. For uniform towers, you may set
this value to 1 and include just one row in the table with the fractional height set to 0. This value
must be an integer greater than 0. There is no upper limit on the number of input stations. (-)
CalcTMode When set to True, this flag tells FAST to calculate the tower mode shapes internally instead of
using the input mode shapes. This feature is NOT CURRENTLY ENABLED, so set the value to
False. (flag)
TwrFADmpi This is the tower’s fore-aft structural damping in percent of critical for the ith bending mode.
Typical values are 0.5%–1.5% and must be between 0 and 100 (inclusive). The damping ratio for
the second mode should usually be greater than the damping ratio for the first mode. The value
for the first mode is used to determine the tower fore-aft damping ratio for the FIELD statements
of extracted ADAMS datasets. The value for the second mode does not effect the creation of
ADAMS datasets. (%)
TwrSSDmpi This is the tower’s side-to-side structural damping in percent of critical for the ith bending mode.
Typical values are 0.5%–1.5% and must be between 0 and 100 (inclusive). The damping ratio for
the second mode should usually be greater than the damping ratio for the first mode. The value
for the first mode is used to determine the tower side-to-side damping ratio for the FIELD
statements of extracted ADAMS datasets. The value for the second mode does not effect the
creation of ADAMS datasets. (%)

Tower Adjustment Factors

FAStTunri These tower stiffness tuners allow you to adjust the stiffness only during the calculation of the ith
tower fore-aft bending mode. Set them to 1.0 to leave the stiffness unchanged. The values of
these inputs do not effect the creation of ADAMS datasets. (-)
SSStTunri These tower stiffness tuners allow you to adjust the stiffness only during the calculation of the ith
tower side-to-side bending mode. Set them to 1.0 to leave the stiffness unchanged. The values of
these inputs do not effect the creation of ADAMS datasets. (-)
AdjTwMa This factor adjusts the tower mass as it is input. This effects all calculations using the tower mass.
Set it to 1.0 to leave the mass unchanged. (-)
AdjFASt This factor adjusts the tower fore-aft stiffness as it is input. This effects all calculations using the
tower fore-aft stiffness. Set it to 1.0 to leave the fore-aft stiffness unchanged. (-)
AdjSSSt This factor adjusts the tower side-to-side stiffness, as it is input. This effects all calculations using
the tower side-to-side stiffness. Set it to 1.0 to leave the side-to-side stiffness unchanged. (-)

Distributed Tower Properties

HtFract This is the fractional height along tower for the other parameters in this table. Values must vary
from 0 to 1. If you are modeling a uniform tower, set NTwInpSt to 1 and set HtFract to 0 for the
single row of distributed tower properties. (-)
TMassDen This is the tower section mass per unit length. It should be computed as the integral of the mass
density over the cross-sectional area of the section. That is, TMassDen = ∫∫ ρ ( x, y ) dxdy ,
where ρ ( x, y ) is the mass density in kg/m3 and x and y are the fore-aft and side-to-side
distances in meters from the tower section mass center to the differential area element,
respectively. These values must be greater than zero. (kg/m)

FAST User's Guide 70 Last updated August 2005 for version 6.0
 Table 9. Tower-Input-File Parameters (continued).

Distributed Tower Properties (continued)

TwFAStif This is the tower section fore-aft stiffness. It should be computed as the integral of the modulus of
elasticity times the square of the fore-aft distance from the tower centerline to the differential area

∫∫ E ( x, y ) x dxdy ,
2
element over the cross-sectional area of the section. That is, TwFAStif =

where E ( x, y ) is the modulus of elasticity in N/m2 and x and y are the fore-aft and side-to-side
distances in meters from the tower centerline to the differential area element, respectively. These
values must be greater than zero. (N·m2)
TwSSStif This is the tower section side-to-side stiffness. It should be computed as the integral of the
modulus of elasticity times the square of the side-to-side distance from the tower centerline to the
differential area element over the cross-sectional area of the section. That is, TwSSStif =

∫∫ E ( x, y ) y dxdy , where E ( x, y ) is the modulus of elasticity in N/m

2 2
and x and y are the
fore-aft and side-to-side distances in meters from the tower centerline to the differential area
element, respectively. These values must be greater than zero. (N·m2)
TwGJStif This is the tower section torsion stiffness used for creation of ADAMS datasets. The FAST model
does not use it. It should be computed as the integral of the modulus of rigidity times the square
of the radial distance from the tower centerline to the differential area element over the cross­

∫∫ G ( x, y ) ( x + y 2 ) dxdy , where
2
sectional area of the section. That is, TwGJStif =

G ( x, y ) is the modulus of rigidity in N/m2 and x and y are the fore-aft and side-to-side distances
in meters from the tower centerline to the differential area element, respectively. When creating
ADAMS datasets, these values must be greater than zero. If the ADAMS preprocessor is
disabled, this input can be left blank. (N m2)
TwEAStif This is the tower section extensional stiffness used for creation of ADAMS datasets. The FAST
model does not use it. It should be computed as the integral of the modulus of elasticity over the
cross-sectional area of the section. That is, TwEAStif =
2
∫∫ E ( x, y ) dxdy , where E ( x, y ) is
the modulus of elasticity in N/m and x and y are the fore-aft and side-to-side distances in meters
from the tower centerline to the differential area element, respectively. When creating ADAMS
datasets, these values must be greater than zero. If the ADAMS preprocessor is disabled, this
input can be left blank. (N)
TwFAIner This is the tower section fore-aft mass inertia per unit length used for creation of ADAMS
datasets. The FAST model does not use it. It should be computed as the integral of the mass
density times the square of the fore-aft distance from the tower section mass center to the
differential area element over the cross-sectional area of the section. That is, TFAIner =

∫∫ ρ ( x, y ) x dxdy , where ρ ( x, y ) is the mass density in kg/m

2 3
and x and y are the fore-aft
and side-to-side distances in meters from the tower section mass center to the differential area
element, respectively. When creating ADAMS datasets, these values must not be less than zero.
If the ADAMS preprocessor is disabled, this input can be left blank. (kg m)
TwSSIner This is the tower section side-to-side mass inertia per unit length used for creation of the ADAMS
dataset. The FAST model does not use it. It should be computed as the integral of the mass
density times the square of the side-to-side distance from the tower section mass center to the
differential area element over the cross-sectional area of the section. That is, TSSIner =

∫∫ ρ ( x, y ) y dxdy , where ρ ( x, y ) is the mass density in kg/m

2 3
and x and y are the fore-aft
and side-to-side distances in meters from the tower section mass center to the differential area
element, respectively. When creating ADAMS datasets, these values must not be less than zero.
If the ADAMS preprocessor is disabled, this input can be left blank. (kg m)

FAST User's Guide 71 Last updated August 2005 for version 6.0
 Table 9. Tower-Input-File Parameters (concluded).

Distributed Tower Properties (concluded)

TwFAcgOf This is the tower section mass offset measured from the tower centerline in the fore-aft direction,
positive downwind. It is used for creation of the ADAMS dataset. The FAST model does not use
it. If the ADAMS preprocessor is disabled, this input can be left blank. (m)
TwSScgOf This is the tower section mass offset measured from the tower centerline in the side-to-side
direction, positive toward the left when looking downwind. It is used for creation of the ADAMS
dataset. The FAST model does not use it. If the ADAMS preprocessor is disabled, this input can
be left blank. (m)

Tower Fore-Aft Mode Shapes

TwFAM1Shi These are the coefficients of the polynomial equation used to model the first fore-aft mode shape
of the tower. The five coefficients (second through sixth) of the polynomial equation define the
mode shape, where the variable in the polynomial varies from 0 to 1. The zeroth and first terms
are not included in the list because they must always be 0 for cantilevered beams. The polynomial
should describe a curve that has a value of 1 at the free end. That is, the five numbers must add up
to 1. (-)
TwFAM2Shi These are the coefficients of the polynomial equation used to model the second fore-aft mode
shape of the tower. The five coefficients (second through sixth) of the polynomial equation define
the mode shape, where the variable in the polynomial varies from 0 to 1. The zeroth and first
terms are not included in the list because they must always be 0 for cantilevered beams. The
polynomial should describe a curve that has a value of 1 at the free end. That is, the five numbers
must add up to 1. (-)

Tower Side-to-Side Mode Shapes

TwSSM1Shi These are the coefficients of the polynomial equation used to model the first side-to-side mode
shape of the tower. The five coefficients (second through sixth) of the polynomial equation define
the mode shape, where the variable in the polynomial varies from 0 to 1. The zeroth and first
terms are not included in the list because they must always be 0 for cantilevered beams. The
polynomial should describe a curve that has a value of 1 at the free end. That is, the five numbers
must add up to 1. (-)
TwSSM2Shi These are the coefficients of the polynomial equation used to model the second side-to-side mode
shape of the tower. The five coefficients (second through sixth) of the polynomial equation define
the mode shape, where the variable in the polynomial varies from 0 to 1. The zeroth and first
terms are not included in the list because they must always be 0 for cantilevered beams. The
polynomial should describe a curve that has a value of 1 at the free end. That is, the five numbers
must add up to 1. (-)

FAST User's Guide 72 Last updated August 2005 for version 6.0
 Table 10. Blade-Input-File Parameters.
The following input parameters are contained in the file indicated by input BldFile from the primary input file.
FAST will read this file even when there are no blade DOFs.

Blade Parameters
NBlInpSt The table of blade sectional data that follows below has NBlInpSt rows of data for each blade.
The values in this table will be interpolated to the BldNodes analysis nodes. For uniform
(untwisted and untapered) blades, set this value to 1 and enter only one row in the distributed-
properties table. For that row, set BlFract equal to zero. This value must be an integer greater
than zero. There is no upper limit on the number of input stations. (-)
CalcBMode When set to True, this flag tells FAST to calculate the blade mode shapes internally instead of
using the input mode shapes. This feature is NOT CURRENTLY ENABLED, so set the value to
False. (flag)
BldFlDmpi The structural damping for the ith flapwise blade-bending mode is entered in percent of critical
damping. Typical values are 0.5%–1.5% and must be between 0 and 100 (inclusive). The second
mode should usually have a higher damping ratio than the first. The value for the first mode is
used to determine the blade-flap damping ratio for the FIELD statements of extracted ADAMS
datasets. The value for the second mode does not effect the creation of ADAMS datasets. (%)
BldEdDmp The structural damping for the edgewise blade bending mode is entered in percent of critical
damping. Typical values are 0.5%–1.5% and must be between 0 and 100 (inclusive). The value is
used to determine the blade-edge damping ratio for the FIELD statements of extracted ADAMS
datasets. (%)

Blade Adjustment Factors

FlStTunri These flapwise stiffness tuners allow you to adjust the flapwise stiffness only during the
calculation of the ith flap-bending mode. Set them to 1 to leave the stiffnesses unchanged. The
values of these inputs do not effect the creation of ADAMS datasets. (-)
AdjBlMs This factor allows you to adjust equally all the blade mass densities in the Distributed Blade
Properties section. The adjustment is made as the data are read in, so this adjustment will affect
all calculations that depend on the blade mass properties. This value must be greater than 0. (-)
AdjFlSt This factor allows you to adjust equally all the flap stiffnesses in the Distributed Blade Properties
section. The adjustment is made as the data are read in, so this adjustment will affect all
calculations that depend on the blade stiffness. This value must be greater than 0. (-)
AdjEdSt This factor allows you to adjust equally all the edge stiffnesses in the Distributed Blade Properties
section. The adjustment is made as the data are read in, so this adjustment will affect all
calculations that depend on the blade stiffness. This value must be greater than 0. (-)

Distributed Blade Properties

BlFract This is the fractional distance of the blade along the blade pitch axis. Values must vary from 0 to
1. The first row, which corresponds to the root of the blade, must have a value of 0. The last row,
which corresponds to the tip of the blade, must have a value of 1. FAST will interpolate this data
table to produce values at the locations specified in the AeroDyn input file. If you don’t want
FAST to use linear interpolation for this, you should specify data at the same analysis nodes
specified in the AeroDyn input file in addition to the root and tip points. (-)

FAST User's Guide 73 Last updated August 2005 for version 6.0
 Table 10. Blade-Input-File Parameters (continued).

Distributed Blade Properties (continued)

AeroCent This input is used to locate the aerodynamic center of the corresponding airfoil section.
AeroCent represents the fractional distance along the chordline from the leading to the trailing
edge, where it is assumed that pitch axis passes through the airfoil section at 25% chord so that
the leading edge is 25% ahead of the pitch axis along the chordline and the trailing edge is 75% aft
of the pitch axis along the chordline. AeroCent is limited to values between 0.0 and 1.0; a value
of 0.0 corresponds to the leading edge, a value of 0.25 corresponds to the blade pitch axis, and a
value of 1.0 corresponds to the trailing edge in FAST models. If the pitch axis in the turbine blade
you are trying to model does not actually pass through the airfoil section at 25% chord (at a
cylindrical root, for example, where it passes at 50% chord), then the AeroCent input may cause
confusion and may not correspond to notation you are used to. The following equation will
convert from your notation to FAST's notation:
AeroCent = 0.25 - [ (fraction of chord from leading edge to actual pitch axis)
- (fraction of chord from leading edge to actual aerodynamic center) ]
For example, in a cylindrical root, where both the actual pitch axis and aerodynamic center lie at
50% chord, you must set AeroCent as follows:
AeroCent = 0.25 - [ (0.5) - (0.5) ] = 0.25 (corresponding to the fact that the aerodynamic center
lies on the pitch axis)
Also as an example, if your pitch axis lies at 30% chord and the aerodynamic center lies at 25%
chord, you must set AeroCent as follows:
AeroCent = 0.25 - [ (0.3) - (0.25) ] = 0.20 (corresponding to the fact that the aerodynamic
center lies 5% ahead of the pitch axis)
ADAMS models generated using the FAST-to-ADAMS preprocessor assume that the reference
axis, indicated by inputs PrecrvRef and PreswpRef, passes through each airfoil section at 25%
chord; thus, a value of 0.25 for AeroCent corresponds to the blade reference axis in ADAMS
models. In this case, the equation above can be adopted if the reference axis in the turbine blade
you are trying to model does not pass through 25% chord by substituting “pitch” with “reference”.
(-)
StrcTwst This is the structural twist angle. It indicates the orientation of the principal axis. A positive
structural twist is one that points the leading edge more upwind. These values must be greater
than –180 and less than or equal to 180 degrees. (deg)
BMassDen This is the blade section mass per unit length. It should be computed as the integral of the mass
density over the cross-sectional area of the section. That is, BMassDen = ∫∫ ρ ( x, y ) dxdy ,
where ρ ( x, y ) is the mass density in kg/m3 and x and y are the flapwise and edgewise distances
in meters from the blade section mass center to the differential area element, respectively. These
values must be greater than zero. (kg/m)
FlpStff This is the blade section flapwise stiffness, not the out-of-plane stiffness. It should be computed
as the integral of the modulus of elasticity times the square of the flapwise distance from the blade
section elastic center to the differential area element over the cross-sectional area of the section.

∫∫ E ( x, y ) x dxdy , where E ( x, y ) is the modulus of elasticity in N/m

2 2
That is, FlpStff = and
x and y are the flapwise and edgewise distances in meters from the blade section elastic center to
the differential area element, respectively. These values must be greater than zero. (N·m2)

FAST User's Guide 74 Last updated August 2005 for version 6.0
 Table 10. Blade-Input-File Parameters (continued).

Distributed Blade Properties (continued)

EdgStff This is the blade section edgewise stiffness, not the in-plane stiffness. It should be computed as
the integral of the modulus of elasticity times the square of the edgewise distance from the blade
section elastic center to the differential area element over the cross-sectional area of the section.

∫∫ E ( x, y ) x dxdy , where E ( x, y )
2
That is, EdgStff = is the modulus of elasticity in N/m2
and x and y are the flapwise and edgewise distances in meters from the blade section elastic center
to the differential area element, respectively. These values must be greater than zero. (N·m2)
GJStff This is the blade section torsion stiffness used for creation of the ADAMS dataset. The FAST
model does not use it. It should be computed as the integral of the modulus of rigidity times the
square of the radial distance from the blade section elastic center to the differential area element

∫∫ G ( x, y ) ( x + y 2 ) dxdy ,
2
over the cross-sectional area of the section. That is, GJStff =

where G ( x, y ) is the modulus of rigidity in N/m2 and x and y are the flapwise and edgewise
distances in meters from the blade section elastic center to the differential area element,
respectively. When creating ADAMS datasets, these values must be greater than zero. If the
ADAMS preprocessor is disabled, this input can be left blank. (N m2)
EAStff This is the blade section extensional stiffness used for creation of the ADAMS dataset. The FAST
model does not use it. It should be computed as the integral of the modulus of elasticity over the
cross-sectional area of the section. That is, EAStff =
2
∫∫ E ( x, y ) dxdy , where E ( x, y ) is the
modulus of elasticity in N/m and x and y are the flapwise and edgewise distances in meters from
the blade section elastic center to the differential area element, respectively. When creating
ADAMS datasets, these values must be greater than zero. If the ADAMS preprocessor is
disabled, this input can be left blank. (N)
Alpha This is the blade section flap/twist coupling coefficient. Valued values are between –1 and 1
(exclusive). Positive values correspond to the blade twisting towards feather as the blade bends
downwind due to thrust loading. Likewise, the blade will twist toward stall as it flaps downwind
due to thrust loading if Alpha is negative. Set Alpha to zero to eliminate the coupling between
flap bending and torsion. The FAST model does not use it. If the ADAMS preprocessor is
disabled, this input can be left blank. (-)
FlpIner This is the blade section flapwise mass inertia per unit length used for creation of the ADAMS
dataset. The FAST model does not use it. It should be computed as the integral of the mass
density times the square of the flapwise distance from the blade section mass center to the
differential area element over the cross-sectional area of the section. That is, FlpIner =

∫∫ ρ ( x, y ) x dxdy , where ρ ( x, y ) is the mass density in kg/m

2 3
and x and y are the flapwise
and edgewise distances in meters from the blade section mass center to the differential area
element, respectively. When creating ADAMS datasets, these values must not be less than zero.
If the ADAMS preprocessor is disabled, this input can be left blank. (kg m)
EdgIner This is the blade section edgewise mass inertia per unit length used for creation of the ADAMS
dataset. The FAST model does not use it. It should be computed as the integral of the mass
density times the square of the edgewise distance from the blade section mass center to the
differential area element over the cross-sectional area of the section. That is, EdgIner =

∫∫ ρ ( x, y ) y dxdy , where ρ ( x, y ) is the mass density in kg/m

2 3
and x and y are the flapwise
and edgewise distances in meters from the blade section mass center to the differential area
element, respectively. When creating ADAMS datasets, these values must not be less than zero.
If the ADAMS preprocessor is disabled, this input can be left blank. (kg m)

FAST User's Guide 75 Last updated August 2005 for version 6.0
 Table 10. Blade-Input-File Parameters (concluded).

Distributed Blade Properties (concluded)

PrecrvRef This is the sectional offset that defines the reference axis for precurved blades. This value is
directed from the blade pitch axis along the xb,i-axis, positive nominally downwind. For upwind
turbines, this value should be negative in order to increase tower clearance. PrecrvRef must be
set to zero for blades without precurve. If PrecrvRef and PreswpRef (next input) are both zero,
the reference axis and pitch axis are coincident. The FAST model does not use it. If the ADAMS
preprocessor is disabled, this input can be left blank. (m)
PreswpRef This is the sectional offset that defines the reference axis for preswept blades. This value is
directed from the blade pitch axis along the yb,i-axis, negative in the direction of rotation.
PreswpRef must be set to zero for blades without presweep. If PrecrvRef (previous input) and
PreswpRef are both zero, the reference axis and pitch axis are coincident. The FAST model does
not use it. If the ADAMS preprocessor is disabled, this input can be left blank. (m)
FlpcgOf This is the blade section mass offset measured from the reference axis in the flapwise direction,
positive toward the suction surface. It is used for creation of the ADAMS dataset. The FAST
model does not use it. If the ADAMS preprocessor is disabled, this input can be left blank. (m)
EdgcgOf This is the blade section mass offset measured from the reference axis in the edgewise direction,
positive toward the trailing edge. It is used for creation of the ADAMS dataset. The FAST model
does not use it. If the ADAMS preprocessor is disabled, this input can be left blank. (m)
FlpEAOf This is the blade section elastic offset measured from the reference axis in the flapwise direction,
positive toward the suction surface. It is used for creation of the ADAMS dataset. The FAST
model does not use it. If the ADAMS preprocessor is disabled, this input can be left blank. (m)
EdgEAOf This is the blade section elastic offset measured from the reference axis in the edgewise direction,
positive toward the trailing edge. It is used for creation of the ADAMS dataset. The FAST model
does not use it. If the ADAMS preprocessor is disabled, this input can be left blank. (m)

Blade Mode Shapes

BldFl1Shi These are the coefficients of the polynomial equation used to model the first flapwise mode shape
of the blade. The five coefficients (second through sixth) of the polynomial equation define the
mode shape, where the variable in the polynomial varies from 0 to 1. The zeroth and first terms
are not included in the list because they must always be 0 for cantilevered beams. The polynomial
should describe a curve that has a value of 1 at the free end. That is, the five numbers must add up
to 1. (-)
BldFl2Shi These are the coefficients of the polynomial equation used to model the second flapwise mode
shape of the blade. The five coefficients (second through sixth) of the polynomial equation define
the mode shape, where the variable in the polynomial varies from 0 to 1. The zeroth and first
terms are not included in the list because they must always be 0 for cantilevered beams. The
polynomial should describe a curve that has a value of 1 at the free end. That is, the five numbers
must add up to 1. (-)
BldEdgShi These are the coefficients of the polynomial equation used to model the edgewise mode shape of
the blade. The five coefficients (second through sixth) of the polynomial equation define the
mode shape, where the variable in the polynomial varies from 0 to 1. The zeroth and first terms
are not included in the list because they must always be 0 for cantilevered beams. The polynomial
should describe a curve that has a value of 1 at the free end. That is, the five numbers must add up
to 1. (-)

FAST User's Guide 76 Last updated August 2005 for version 6.0
 Table 11. AeroDyn-Input-File Parameters.
The following input parameters are contained in the file indicated by input ADFile from the primary input file.
AeroDyn will read this file even when CompAero from the primary input file is disabled. For more details about
the parameters documented here, please see the latest AeroDyn User’s Guide [1].

Aerodynamics
ADTitle This is an input-file descriptor that is displayed on the screen during execution. AeroDyn will
read the first 96 characters of the text. There is no need to put it in quotes. You may enter
anything you like—even an empty string, but you must consume exactly one line in the input file.
(nonquoted string)
SysUnits This string controls the setting of the SIUnits flag, which tells AeroDyn whether to assume input
and output parameters are given in metric or English units. This string must be “SI” for FAST.
Some other codes that use AeroDyn allow input and output parameters in English units, but FAST
does not. (nonquoted string)
StallMod This string controls the setting of the DynStall flag, which tells AeroDyn whether or not to use the
Leishman-Beddoes dynamic stall in AeroDyn. The only permissible values are “BEDDOES” and
“STEADY”. This string is normally set to “BEDDOES” to use dynamic stall for production
simulations. During a linearization analysis, dynamic stall must be disabled by specifying
StallMod to “STEADY”. (nonquoted string)
UseCm This string controls the setting of the PitchMom flag, which tells AeroDyn whether to compute
pitching moments in AeroDyn. The only permissible values are “USE_CM” and “NO_CM”.
Although pitching moments will have an effect on the loads and motion of the turbine in FAST,
there is no twist degree of freedom. (nonquoted string)
InfModel This string controls the setting of the DynInflo flag, which tells AeroDyn whether to use the
generalized-dynamic-wake model or the equilibrium-inflow model. The two possible string
values are “DYNIN” and “EQUIL”. For production runs, this string should be set to “DYNIN”.
During a linearization analysis, the “EQUIL”ibrium-inflow model must be engaged. (nonquoted
string)
IndModel When using the equilibrium-inflow model, this quoted string controls the setting of the AxialInd
and TangInd flags. The three possible values are “NONE”, “WAKE”, and “SWIRL”. A setting
of “NONE” disables both flags, a setting of “WAKE” enables just the AxialInd flag, and a setting
of “SWIRL” enables both flags. If you are doing production runs using equilibrium inflow, you
should set this value to “SWIRL”. (nonquoted string)
AToler When using the equilibrium inflow model, an iterative solution is used to calculate the induction
factors. AeroDyn uses the value of AToler as the convergence criterion. A good default value to
use is 0.005. This value may be reduced to increase accuracy or increased to speed the calcula­
tions. This value must be greater than zero. (-)
TLModel When using the equilibrium inflow model, you can select from two tip-loss models or disable tip-
loss calculations. {“NONE”: no tip-loss calculations, “PRAND”: standard Prandtl tip-loss model,
“GTECH”: Georgia Tech’s modified Prandtl model} (nonquoted string)
HLModel When using the equilibrium inflow model, you can include or disable hub-loss calculations.
{“NONE”: no hub-loss calculations, “PRAND”: standard Prandtl hub-loss model} (nonquoted
string)
WindFile This quoted string holds the name or root name of the wind input file. AeroDyn will check the file
system to determine whether the file contains hub-height wind data or full-field (FF) wind data.
For FF winds, omit the file extension. The file name may optionally include an absolute or rela­
tive path. This file name must contain fewer than 100 characters and must be enclosed in apostro­
phes or double quotes. During a linearization analysis, you must use a hub-height wind data file
that does not vary with time. (quoted string)
HH This is the height above the ground [onshore] or height above the mean sea level [offshore] that
AeroDyn will use as a hub height for the winds. You should set this to TowerHt + Twr2Shft +
OverHang•SIN( ShftTilt ). This value must not be negative. (m)

FAST User's Guide 77 Last updated August 2005 for version 6.0
 Table 11. AeroDyn-Input-File Parameters (continued).

Aerodynamics (concluded)
TwrShad The tower-shadow maximum velocity deficit is the fractional amount the horizontal wind speed is
reduced at the middle of the shadow a distance T_Shad_RefPt downstream from the center of the
tower. This varies from 0 to 1. A value of 0 means there is no shadow. A value of 1 means the
wind is completely stopped. A typical number might be something like 0.3. This value must not
be negative. (-)
ShadHWid The tower-shadow half-width tells AeroDyn how wide the tower shadow is at a distance
T_Shad_RefPt downstream from the center of the tower. This number should normally be
slightly larger than the half-width of the tower, as the tower shadow usually widens as it goes
downstream. This value must not be negative. (m)
T_Shad_RefPt This distance downstream of the tower specifies the point where the input values of the velocity
deficit and shadow width are defined. An appropriate value would be the horizontal distance from
the tower centerline to the hub, which would be OverHang•COS( ShftTilt ). This value must not
be negative. (m)
Rho This is the ambient air density at the altitude of the hub. The standard density at sea level is 1.225.
By setting this value to 0 you will effectively eliminate all aerodynamic forces on the turbine, but
you will save a lot of calculations by instead disabling the CompAero flag. This value must not
be negative. (kg/m3)
KinVisc This is the ambient relative viscosity at the altitude of the hub. This value is not currently used in
AeroDyn or FAST, but it will eventually be used to compute the Reynolds number. The standard
relative viscosity at sea level is 1.46e-5. This value must not be negative. (kg/m·sec)
DTAero This is the time step size that tells AeroDyn how often to compute aerodynamic forces. This value
must be greater than zero. It does not need to be specified as an integral multiplier of FAST’s time
step, DT, but if it is not, FAST will still only call AeroDyn at the least greatest integer multiple of
DT that is larger or equal to DTAero. (sec)
NumFoil This parameter determines how many airfoil tables will be available for assignment to the various
blade stations and tail fin airfoil. Any non-zero number of airfoil files can be specified. Any or all
of them may be used by more than one blade station or never used at all. (-)
FoilNmi The next NumFoil lines are a list of airfoil-table file names entered in quoted strings. The file
names are limited to 80 characters and may contain absolute or relative paths. Leading and
trailing spaces are trimmed, but imbedded spaces are kept. Leading spaces count against the 80­
character limit. Only one filename is entered on each line. (quoted strings)
BldNodes The blades will have BldNodes analysis nodes in FAST, which are used for the integration of
aerodynamic and elastic forces. The more segments you use, the more accurate the integral will
be, but the greater the computational time will be. A good compromise for this parameter is 20.
This integer number must be greater than 1. When creating ADAMS datasets, this value must be
no more than 99. (-)

FAST User's Guide 78 Last updated August 2005 for version 6.0
 Table 11. AeroDyn-Input-File Parameters (concluded).

Distributed Blade Information

RNodes These values are the distances of the analysis nodes from the rotor apex along the pitch axis,
which is not consistent with the YawDyn convention for 3-bladed turbines (but is consistent
for 2-bladed turbines). The analysis nodes are located at the centers of the blade segments. You
do not have complete freedom to put the RNodes anywhere you like. For instance, they cannot
be at the blade root or tip. It is also not always possible to alternate between nodes that are close
together and far apart. An easy way to ensure consistency is to divide the blade segments and then
compute the centers of the segments to use for the RNodes. The nodes do not need to be equally
spaced. These values must fall between the HubRad and TipRad (exclusive). (m)
AeroTwst This is the aerodynamic twist angle. It indicates the orientation of the chord of the local airfoil. A
positive aerodynamic twist is one that points the leading edge more upwind. These values must be
greater than –180 and less than or equal to 180 degrees. (deg)
DRNodes These values represent the portion of the blade span that is assigned to an analysis node. This
length times the local chord defines the area used in the aerodynamics calculations. The sum of all
the DRNodes should add up to the blade length. FAST checks the values of the RNodes and
DRNodes to make sure they are consistent and meaningful. These values must be greater than
zero. (m)
Chord These values are the local chords of the analysis nodes. This local chord times DRNodes defines
the area used in the aerodynamics calculations. These values must be greater than zero. (m)
NFoil These integers tell AeroDyn which of the input airfoil files (FoilNm) are assigned to the various
analysis nodes. For instance, a value of 2 means that node 2 will use FoilNm2 for the local airfoil.
Airfoils may be assigned to more than one blade station. These values must be between 1 and
NumFoil. (-)
PrnElm If the whole word “PRINT” is found anywhere on a line, the element data for that element will be
printed to the element output file (element.plt). You can also enter the nonquoted string
“NOPRINT”, or leave this field blank to skip printing element data for this element. (nonquoted,
case-insensitive string)

FAST User's Guide 79 Last updated August 2005 for version 6.0
 Table 12. Platform-Input-File Parameters.
The following input parameters are contained in the file indicated by input PtfmFile from the primary input file.
FAST will only read this file if PtfmModel from the primary input file is nonzero. In FAST v6.0, all nonzero
PtfmModel options will work the same way by reading in the PtfmFile described here. In future versions, the
format of this file will depend on which PtfmModel option is selected.

Feature Flags
PtfmSgDOF The support platform surge DOF will be enabled when this is True. The surge DOF allows the
platform to translate horizontally relative to the inertia frame as shown in Figure 20. The platform
reference point (located by input PtfmRef) translates with the platform during this motion. The
initial surge displacement is specified with PtfmSurge. If PtfmSgDOF is disabled, the surge
displacement will be fixed at PtfmSurge. (flag)
PtfmSwDOF The support platform sway DOF will be enabled when this is True. The sway DOF allows the
platform to translate horizontally relative to the inertia frame as shown in Figure 20. The platform
reference point (located by input PtfmRef) translates with the platform during this motion. The
initial sway displacement is specified with PtfmSway. If PtfmSwDOF is disabled, the sway
displacement will be fixed at PtfmSway. (flag)
PtfmHvDOF The support platform heave DOF will be enabled when this is True. The heave DOF allows the
platform to translate vertically relative to the inertia frame as shown in Figure 20. The platform
reference point (located by input PtfmRef) translates with the platform during this motion. The
initial heave displacement is specified with PtfmHeave. If PtfmHvDOF is disabled, the heave
displacement will be fixed at PtfmHeave. (flag)
PtfmRDOF The support platform roll DOF will be enabled when this is True. The roll DOF allows the
platform to tilt (rotate) about its reference point (located by input PtfmRef) relative to the inertia
frame as shown in Figure 20. The initial roll displacement is specified with PtfmRoll. If
PtfmRDOF is disabled, the roll displacement will be fixed at PtfmRoll. (flag)
PtfmPDOF The support platform pitch DOF will be enabled when this is True. The pitch DOF allows the
platform to tilt (rotate) about its reference point (located by input PtfmRef) relative to the inertia
frame as shown in Figure 20. The initial pitch displacement is specified with PtfmPitch. If
PtfmPDOF is disabled, the pitch displacement will be fixed at PtfmPitch. (flag)
PtfmYDOF The support platform yaw DOF will be enabled when this is True. The yaw DOF allows the
platform to yaw (rotate) about its reference point (located by input PtfmRef) relative to the inertia
frame as shown in Figure 20. The initial yaw displacement is specified with PtfmYaw. If
PtfmYDOF is disabled, the yaw displacement will be fixed at PtfmYaw. (flag)

Initial Conditions
PtfmSurge This is the fixed or initial support platform surge displacement. The surge displacement indicates
a horizontal translation of the platform relative to the inertia frame as shown in Figure 20. (m)
PtfmSway This is the fixed or initial support platform sway displacement. The sway displacement indicates a
horizontal translation of the platform relative to the inertia frame as shown in Figure 20. (m)
PtfmHeave This is the fixed or initial support platform heave displacement. The heave displacement indicates
a vertical translation of the platform relative to the inertia frame as shown in Figure 20. (m)
PtfmRoll This is the fixed or initial support platform roll displacement. The roll displacement indicates a tilt
rotation of the platform about its reference point (located by input PtfmRef) relative to the inertia
frame as shown in Figure 20. This value must be between –15 and 15 degrees (inclusive). (deg)
PtfmPitch This is the fixed or initial support platform pitch displacement. The pitch displacement indicates a
tilt rotation of the platform about its reference point (located by input PtfmRef) relative to the
inertia frame as shown in Figure 20. This value must be between –15 and 15 degrees (inclusive).
(deg)
PtfmYaw This is the fixed or initial support platform yaw displacement. The yaw displacement indicates a
yaw rotation of the platform about its reference point (located by input PtfmRef) relative to the
inertia frame as shown in Figure 20. This value must be between –15 and 15 degrees (inclusive).
(deg)

FAST User's Guide 80 Last updated August 2005 for version 6.0
 Table 12. Platform-Input-File Parameters (concluded).

Turbine Configuration
TwrDraft The tower draft is the downward distance from ground level [onshore] or mean sea level
[offshore] to the tower base platform connection. This value must be greater than –TowerHt. See
Figure 20. (m)
PtfmCM This is the downward distance from ground level [onshore] or mean sea level [offshore] to the
support platform mass center. This value must not be less than TwrDraft. See Figure 20. (m)
PtfmRef This is the downward distance from ground level [onshore] or mean sea level [offshore] to the
support platform reference point. The platform reference point is the origin in the platform about
which the translational (surge, sway, and heave) and rotational (roll, pitch, and yaw) motions of
the support platform are defined. It is also the point at which external loading is applied to the
platform—see input parameter PtfmLdMod. This value must not be less than TwrDraft. See
Figure 20. (m)

Mass and Inertia

PtfmMass This is the mass of the support platform. Its center is located a downward distance of PtfmCM
from ground level [onshore] or mean sea level [offshore]. If TwrRBHt is nonzero, the mass of the
rigid portion of the tower should be included with the support platform mass in PtfmMass. This
value must not be negative. (kg)
PtfmRIner This is the support platform moment of inertia in roll about the platform mass center. It includes
all mass contained in PtfmMass. This value must not be negative. (kg·m2)
PtfmPIner This is the support platform moment of inertia in pitch about the platform mass center. It includes
all mass contained in PtfmMass. This value must not be negative. (kg·m2)
PtfmYIner This is the support platform moment of inertia in yaw about the platform mass center. It includes
all mass contained in PtfmMass. This value must not be negative. (kg·m2)

Platform Loading
PtfmLdMod In FAST v6.0, only user-defined platform loading is available. For a value of 0 for PtfmLdMod,
there will be no platform loading and the support reactions normally produced will be set to zero
(causing the wind turbine to fall due to gravity if PtfmHvDOF is True). If you set PtfmLdMod to
1, FAST will call the routine UserPtfmLd() to compute the platform loading. You should replace
the dummy routine supplied with the code with your own, which will need to be linked with the
rest of FAST. The platform loads returned by UserPtfmLd() should contain contributions from
any external load acting on the platform other than loads transmitted from the wind turbine. For
example, these loads should contain contributions from foundation stiffness and damping [not
floating] or mooring line restoring and damping [floating], as well as hydrostatic and
hydrodynamic contributions [offshore]. The platform loads will be applied on the platform at the
instantaneous platform reference position (located by input PtfmRef). The routine assumes that
the platform loads are transmitted through a medium like soil [foundation] and/or water [offshore],
so that added mass effects are important. See the dummy UserPtfmLd() routine for more
information. Using values other than 0 or 1 for PtfmLdMod will cause FAST to abort. (switch)

FAST User's Guide 81 Last updated August 2005 for version 6.0
 Table 13. Furling-Input-File Parameters.
The following input parameters are contained in the file indicated by input FurlFile from the primary input file.
FAST will only read this file if the model is designated as a furling machine (when Furling from the primary input
file is set to True).

Feature Flags
RFrlDOF The rotor-furl DOF will be enabled when this is True. The initial rotor-furl angle is specified with
RotFurl. If RFrlDOF is disabled, the rotor-furl angle will be fixed at RotFurl. (flag)
TFrlDOF The tail-furl DOF will be enabled when this is True. The initial tail-furl angle is specified with
TailFurl. If TFrlDOF is disabled, the tail-furl angle will be fixed at TailFurl. (flag)

Initial Conditions
RotFurl This is the fixed or initial rotor-furl angle. It is positive about the rotor-furl axis as shown in
Figure 17. The rotor-furl axis is defined through inputs RFrlPntxn, RFrlPntyn, RFrlPntzn,
RFrlSkew, and RFrlTilt below. This value must be greater than –180 and less than or equal to
180 degrees. (deg)
TailFurl This is the fixed or initial tail-furl angle. It is positive about the tail-furl axis as shown in Figure
17. The tail-furl axis is defined through inputs TFrlPntxn, TFrlPntyn, TFrlPntzn, TFrlSkew,
and TFrlTilt below. This value must be greater than –180 and less than or equal to 180 degrees.
(deg)

Turbine Configuration
Yaw2Shft This is the lateral offset distance from the yaw axis to the intersection of the rotor shaft axis with
the yn-/zn-plane. The distance is measured parallel to the yn-axis. It is positive to the left when
looking downwind as shown in Figure 18. For turbines with rotor-furl, this distance defines the
configuration at a furl angle of zero. (m)
ShftSkew This is the skew angle of the rotor shaft in the nominally horizontal plane. Positive skew acts like
positive nacelle yaw as shown in Figure 18; however, ShftSkew should only be used to skew the
shaft a few degrees away from the zero-yaw position and must not be used as a replacement for
the yaw angle. This value must be between –15 and 15 degrees (inclusive). For turbines with
rotor-furl, this angle defines the configuration at a furl angle of zero. (deg)
RFrlCMxn This is the downwind distance to the center of mass of the structure that furls with the rotor (not
including the rotor—reference input RFrlMass) from the top of the tower, measured parallel to
the xn-axis. It is positive downwind. See Figure 18. For turbines with rotor-furl, this distance
defines the configuration at a furl angle of zero. (m)
RFrlCMyn This is the lateral distance to the center of mass of the structure that furls with the rotor (not
including the rotor—reference input RFrlMass) from the top of the tower, measured parallel to
the yn-axis. It is positive to the left when looking downwind. See Figure 18. For turbines with
rotor-furl, this distance defines the configuration at a furl angle of zero. (m)
RFrlCMzn This is the vertical distance to the center of mass of the structure that furls with the rotor (not
including the rotor—reference input RFrlMass) from the top of the tower, measured parallel to
the zn-axis. It is positive upward when looking downwind. See Figure 18. For turbines with
rotor-furl, this distance defines the configuration at a furl angle of zero. (m)
BoomCMxn This is the downwind distance to the tail boom mass center (reference input BoomMass) from the
top of the tower, measured parallel to the xn-axis. It is positive downwind. See Figure 19. For
turbines with tail-furl, this distance defines the configuration at a furl angle of zero. (m)
BoomCMyn This is the lateral distance to the tail boom mass center (reference input BoomMass) from the top
of the tower, measured parallel to the yn-axis. It is positive to the left when looking downwind.
See Figure 19. For turbines with tail-furl, this distance defines the configuration at a furl angle of
zero. (m)
BoomCMzn This is the vertical distance to the tail boom mass center (reference input BoomMass) from the
top of the tower, measured parallel to the zn-axis. It is positive upward when looking downwind.
See Figure 19. For turbines with tail-furl, this distance defines the configuration at a furl angle of
zero. (m)
FAST User's Guide 82 Last updated August 2005 for version 6.0
 Table 13. Furling-Input-File Parameters (continued).

Turbine Configuration (continued)

TFinCMxn This is the downwind distance to the tail fin mass center (reference input TFinMass) from the top
of the tower, measured parallel to the xn-axis. It is positive downwind. See Figure 19. For
turbines with tail-furl, this distance defines the configuration at a furl angle of zero. (m)
TFinCMyn This is the lateral distance to the tail fin mass center (reference input TFinMass) from the top of
the tower, measured parallel to the yn-axis. It is positive to the left when looking downwind. See
Figure 19. For turbines with tail-furl, this distance defines the configuration at a furl angle of
zero. (m)
TFinCMzn This is the vertical distance to the tail fin mass center (reference input TFinMass) from the top of
the tower, measured parallel to the zn-axis. It is positive upward when looking downwind. See
Figure 19. For turbines with tail-furl, this distance defines the configuration at a furl angle of
zero. (m)
TFinCPxn This is the downwind distance to the tail fin center-of-pressure from the top of the tower,
measured parallel to the xn-axis. It is positive downwind. See Figure 19. For turbines with tail-
furl, this distance defines the configuration at a furl angle of zero. (m)
TFinCPyn This is the lateral distance to the tail fin center-of-pressure from the top of the tower, measured
parallel to the yn-axis. It is positive to the left when looking downwind. See Figure 19. For
turbines with tail-furl, this distance defines the configuration at a furl angle of zero. (m)
TFinCPzn This is the vertical distance to the tail fin center-of-pressure from the top of the tower, measured
parallel to the zn-axis. It is positive upward when looking downwind. See Figure 19. For
turbines with tail-furl, this distance defines the configuration at a furl angle of zero. (m)
TFinSkew This is the skew angle of the tail fin chordline in the nominally horizontal plane. Positive skew
orients the nominal horizontal projection of the tail fin chordline about the zn-axis. The
aforementioned chordline is the chordline passing through the tail fin center-of-pressure. See
Figure 19. This value must be greater than –180 and less than or equal to 180 degrees. For
turbines with tail-furl, this angle defines the configuration at a furl angle of zero. (deg)
TFinTilt This is the tilt angle of the tail fin chordline from the nominally horizontal plane. The
aforementioned chordline is the chordline passing through the tail fin center-of-pressure. This
value must be between –90 and 90 degrees (inclusive). Positive tilt means that the trailing edge of
the tail fin is higher than the leading edge. See Figure 19. For turbines with tail-furl, this angle
defines the configuration at a furl angle of zero. (deg)
TFinBank This is the bank angle of the tail fin plane about the tail fin chordline. The aforementioned
chordline is the chordline passing through the tail fin center-of-pressure. This value must be
greater than –180 and less than or equal to 180 degrees. See Figure 19. For turbines with tail-furl,
this angle defines the configuration at a furl angle of zero. (deg)
RFrlPntxn This is the downwind distance to an arbitrary point on the rotor-furl axis from the top of the tower,
measured parallel to the xn-axis. It is positive downwind. The arbitrary point referred to in this
input must be the same point identified by inputs RFrlPntyn and RFrlPntzn. Inputs RFrlPntxn,
RFrlPntyn, RFrlPntzn, RFrlSkew, and RFrlTilt define the orientation of the rotor-furl axis and
associated DOF, RFrlDOF. See Figure 17. (m)
RFrlPntyn This is the lateral distance to an arbitrary point on the rotor-furl axis from the top of the tower,
measured parallel to the yn-axis. It is positive to the left when looking downwind. The arbitrary
point referred to in this input must be the same point identified by inputs RFrlPntxn and
RFrlPntzn. Inputs RFrlPntxn, RFrlPntyn, RFrlPntzn, RFrlSkew, and RFrlTilt define the
orientation of the rotor-furl axis and associated DOF, RFrlDOF. See Figure 17. (m)
RFrlPntzn This is the vertical distance to an arbitrary point on the rotor-furl axis from the top of the tower,
measured parallel to the zn-axis. It is positive upward when looking downwind. The arbitrary
point referred to in this input must be the same point identified by inputs RFrlPntxn and
RFrlPntyn. Inputs RFrlPntxn, RFrlPntyn, RFrlPntzn, RFrlSkew, and RFrlTilt define the
orientation of the rotor-furl axis and associated DOF, RFrlDOF. See Figure 17. (m)

FAST User's Guide 83 Last updated August 2005 for version 6.0
 Table 13. Furling-Input-File Parameters (continued).

Turbine Configuration (concluded)

RFrlSkew This is the skew angle of the rotor-furl axis in the nominally horizontal plane. Positive skew
orients the nominal horizontal projection of the rotor-furl axis about the zn-axis. Inputs
RFrlPntxn, RFrlPntyn, RFrlPntzn, RFrlSkew, and RFrlTilt define the orientation of the rotor-
furl axis and associated DOF, RFrlDOF. See Figure 17. This value must be greater than –180
and less than or equal to 180 degrees. (deg)
RFrlTilt This is the tilt angle of the rotor-furl axis from the nominally horizontal plane. This value must be
between –90 and 90 degrees (inclusive). Inputs RFrlPntxn, RFrlPntyn, RFrlPntzn, RFrlSkew,
and RFrlTilt define the orientation of the rotor-furl axis and associated DOF, RFrlDOF. See
Figure 17. (deg)
TFrlPntxn This is the downwind distance to an arbitrary point on the tail-furl axis from the top of the tower,
measured parallel to the xn-axis. It is positive downwind. The arbitrary point referred to in this
input must be the same point identified by inputs TFrlPntyn and TFrlPntzn. Inputs TFrlPntxn,
TFrlPntyn, TFrlPntzn, TFrlSkew, and TFrlTilt define the orientation of the tail-furl axis and
associated DOF, TFrlDOF. See Figure 17. (m)
TFrlPntyn This is the lateral distance to an arbitrary point on the tail-furl axis from the top of the tower,
measured parallel to the yn-axis. It is positive to the left when looking downwind. The arbitrary
point referred to in this input must be the same point identified by inputs TFrlPntxn and
TFrlPntzn. Inputs TFrlPntxn, TFrlPntyn, TFrlPntzn, TFrlSkew, and TFrlTilt define the
orientation of the tail-furl axis and associated DOF, TFrlDOF. See Figure 17. (m)
TFrlPntzn This is the vertical distance to an arbitrary point on the tail-furl axis from the top of the tower,
measured parallel to the zn-axis. It is positive upward when looking downwind. The arbitrary
point referred to in this input must be the same point identified by inputs TFrlPntxn and
TFrlPntyn. Inputs TFrlPntxn, TFrlPntyn, TFrlPntzn, TFrlSkew, and TFrlTilt define the
orientation of the tail-furl axis and associated DOF, RFrlDOF. See Figure 17. (m)
TFrlSkew This is the skew angle of the tail-furl axis in the nominally horizontal plane. Positive skew orients
the nominal horizontal projection of the tail-furl axis about the zn-axis. Inputs TFrlPntxn,
TFrlPntyn, TFrlPntzn, TFrlSkew, and TFrlTilt define the orientation of the tail-furl axis and
associated DOF, TFrlDOF. See Figure 17. This value must be greater than –180 and less than or
equal to 180 degrees. (deg)
TFrlTilt This is the tilt angle of the tail-furl axis from the nominally horizontal plane. This value must be
between –90 and 90 degrees (inclusive). Inputs TFrlPntxn, TFrlPntyn, TFrlPntzn, TFrlSkew,
and TFrlTilt define the orientation of the tail-furl axis and associated DOF, TFrlDOF. See Figure
17. (deg)

Mass and Inertia

RFrlMass This is the mass of the structure that furls with the rotor (not including the rotor). The center of
this mass is located at the point specified by inputs RFrlCMxn, RFrlCMyn, and RFrlCMzn
relative to the tower-top at a rotor-furl angle of zero. It includes everything that furls with the
rotor excluding the rotor (blades, hub, and tip brakes). This value must not be negative. (kg)
BoomMass This is the mass of the tail boom. The center of the tail boom mass is located at the point specified
by inputs BoomCMxn, BoomCMyn, and BoomCMzn relative to the tower-top at a tail-furl angle
of zero. It includes everything that furls with the tail except the tail fin (see next input). This
value must not be negative. (kg)
TFinMass This is the mass of the tail fin. The center of the tail fin mass is located at the point specified by
inputs TFinCMxn, TFinCMyn, and TFinCMzn relative to the tower-top at a tail-furl angle of
zero. TFinMass and BoomMass combined should include everything that furls with the tail.
This value must not be negative. (kg)
RFrlIner This is the moment of inertia of the structure that furls with the rotor (not including the rotor)
about the rotor-furl axis. It includes all mass contained in RFrlMass. This value must be greater
than RFrlMass•( perpendicular distance between rotor-furl axis and C.M. of the structure
that furls with the rotor [not including the rotor] )2. (kg·m2)

FAST User's Guide 84 Last updated August 2005 for version 6.0
 Table 13. Furling-Input-File Parameters (continued).

Mass and Inertia (concluded)

TFrlIner This is the tail boom moment of inertia about the tail-furl axis. It includes all mass contained in
BoomMass. This value must be greater than BoomMass•( perpendicular distance between
tail-furl axis and tail boom C.M. )2. (kg·m2)

Rotor-Furl
RFrlMod The rotor-furl springs and dampers can be modeled three ways. For a value of 0 for RFrlMod,
there will be no rotor-furl spring nor damper and the moment normally produced will be set to
zero. A RFrlMod of 1 will invoke simple spring and damper models using the inputs provided
below as appropriate coefficients. If you set RFrlMod to 2, FAST will call the routine UserRFrl()
to compute the rotor-furl spring and damper moments. You should replace the dummy routine
supplied with the code with your own, which will need to be linked with the rest of FAST. Using
values other than 0, 1, or 2 will cause FAST to abort. (switch)
RFrlSpr The linear rotor-furl spring restoring moment is proportional to the rotor-furl deflection through
this constant. This value must not be negative and is only used when RFrlMod is set to 1.
(N·m/rad)
RFrlDmp The linear rotor-furl damping moment is proportional to the rotor-furl rate through this constant.
This value must not be negative and is only used when RFrlMod is set to 1. (N·m/(rad/sec))
RFrlCDmp This Coulomb-friction damping moment resists rotor-furl motion, but it is a constant that is not
proportional to the rotor-furl rate. However, if the rotor-furl rate is zero, the damping is zero.
This value must not be negative and is only used when RFrlMod is set to 1. (N·m)
RFrlUSSP The rotor-furl up-stop spring is effective when the rotor-furl deflection exceeds this value. This
value must be greater than –180 and less than or equal to 180 degrees and is only used when
RFrlMod is set to 1. (deg)
RFrlDSSP The rotor-furl down-stop spring is effective when the rotor-furl deflection exceeds this value. This
value must be greater than –180 and less than or equal to RFrlUSSP degrees and is only used
when RFrlMod is set to 1. (deg)
RFrlUSSpr The linear rotor-furl up-stop spring restoring moment is proportional to the rotor-furl up-stop
deflection by this constant and is effective when the rotor-furl deflection exceeds RFrlUSSP.
This value must not be negative and is only used when RFrlMod is set to 1. (N·m/rad)
RFrlDSSpr The linear rotor-furl down-stop spring restoring moment is proportional to the rotor-furl down-
stop deflection by this constant and is effective when the rotor-furl deflection exceeds RFrlDSSP.
This value must not be negative and is only used when RFrlMod is set to 1. (N·m/rad)
RFrlUSDP The rotor-furl up-stop damper is effective when the rotor-furl deflection exceeds this value. This
value must be greater than –180 and less than or equal to 180 degrees and is only used when
RFrlMod is set to 1. (deg)
RFrlDSDP The rotor-furl down-stop damper is effective when the rotor-furl deflection exceeds this value.
This value must be greater than –180 and less than or equal to RFrlUSDP degrees and is only
used when RFrlMod is set to 1. (deg)
RFrlUSDmp The linear rotor-furl up-stop damping moment is proportional to the rotor-furl rate by this constant
and is effective when the rotor-furl deflection exceeds RFrlUSDP. This value must not be
negative and is only used when RFrlMod is set to 1. (N·m/(rad/sec))
RFrlDSDmp The linear rotor-furl down-stop damping restoring moment is proportional to the rotor-furl rate by
this constant and is effective when the rotor-furl deflection exceeds RFrlDSDP. This value must
not be negative and is only used when RFrlMod is set to 1. (N·m/(rad/sec))

FAST User's Guide 85 Last updated August 2005 for version 6.0
 Table 13. Furling-Input-File Parameters (continued).

Tail-Furl
TFrlMod The tail-furl springs and dampers can be modeled three ways. For a value of 0 for TFrlMod, there
will be no tail-furl spring nor damper and the moment normally produced will be set to zero. A
TFrlMod of 1 will invoke simple spring and damper models using the inputs provided below as
appropriate coefficients. If you set TFrlMod to 2, FAST will call the routine UserTFrl() to
compute the tail-furl spring and damper moments. You should replace the dummy routine
supplied with the code with your own, which will need to be linked with the rest of FAST. Using
values other than 0, 1, or 2 will cause FAST to abort. (switch)
TFrlSpr The linear tail-furl spring restoring moment is proportional to the tail-furl deflection through this
constant. This value must not be negative and is only used when TFrlMod is set to 1. (N·m/rad)
TFrlDmp The linear tail-furl damping moment is proportional to the tail-furl rate through this constant. This
value must not be negative and is only used when TFrlMod is set to 1. (N·m/(rad/sec))
TFrlCDmp This Coulomb-friction damping moment resists tail-furl motion, but it is a constant that is not
proportional to the tail-furl rate. However, if the tail-furl rate is zero, the damping is zero. This
value must not be negative and is only used when TFrlMod is set to 1. (N·m)
TFrlUSSP The tail-furl up-stop spring is effective when the tail-furl deflection exceeds this value. This value
must be greater than –180 and less than or equal to 180 degrees and is only used when TFrlMod is
set to 1. (deg)
TFrlDSSP The tail-furl down-stop spring is effective when the tail-furl deflection exceeds this value. This
value must be greater than –180 and less than or equal to TFrlUSSP degrees and is only used
when TFrlMod is set to 1. (deg)
TFrlUSSpr The linear tail-furl up-stop spring restoring moment is proportional to the tail-furl up-stop
deflection by this constant and is effective when the tail-furl deflection exceeds TFrlUSSP. This
value must not be negative and is only used when TFrlMod is set to 1. (N·m/rad)
TFrlDSSpr The linear tail-furl down-stop spring restoring moment is proportional to the tail-furl down-stop
deflection by this constant and is effective when the tail-furl deflection exceeds TFrlDSSP. This
value must not be negative and is only used when TFrlMod is set to 1. (N·m/rad)
TFrlUSDP The tail-furl up-stop damper is effective when the tail-furl deflection exceeds this value. This
value must be greater than –180 and less than or equal to 180 degrees and is only used when
TFrlMod is set to 1. (deg)
TFrlDSDP The tail-furl down-stop damper is effective when the tail-furl deflection exceeds this value. This
value must be greater than –180 and less than or equal to TFrlUSDP degrees and is only used
when TFrlMod is set to 1. (deg)
TFrlUSDmp The linear tail-furl up-stop damping moment is proportional to the tail-furl rate by this constant
and is effective when the tail-furl deflection exceeds TFrlUSDP. This value must not be negative
and is only used when TFrlMod is set to 1. (N·m/(rad/sec))
TFrlDSDmp The linear tail-furl down-stop damping restoring moment is proportional to the tail-furl rate by this
constant and is effective when the tail-furl deflection exceeds TFrlDSDP. This value must not be
negative and is only used when TFrlMod is set to 1. (N·m/(rad/sec))

FAST User's Guide 86 Last updated August 2005 for version 6.0
 Table 13. Furling-Input-File Parameters (concluded).

Tail Fin Aerodynamics

TFinMod The tail fin aerodynamics can be modeled three ways. For a value of 0 for TFinMod, there will be
no tail fin aerodynamics and the aerodynamic loads normally produced will be set to zero. A
TFinMod of 1 will invoke a simplified tail fin aerodynamics model using the inputs provided
below as appropriate parameters. If you set TFinMod to 2, FAST will call the routine UserTFin()
to compute the tail fin aerodynamic loads. You should replace the dummy routine supplied with
the code with your own, which will need to be linked with the rest of FAST. Using values other
than 0, 1, or 2 will cause FAST to abort. (switch)
TFinNFoil This integer tells AeroDyn which of the input airfoil files (FoilNm) is assigned to the tail fin. For
instance, a value of 2 means that the tail fin will use FoilNm2 for the local tail fin airfoil. The tail
fin airfoil may be assigned to the same airfoil as one or more blade stations. This value must be
between 1 and NumFoil and is only used when TFinMod is set to 1. (-)
TFinArea This is the plan form area of the tail fin plate used to relate the local dynamic pressure and airfoil
coefficients to aerodynamic loads. This value must not be negative and is only used when
TFinMod is set to 1. (m2)
SubAxInd Set this value to False if you want the wind velocity at the tail fin to be unobstructed by the rotor
wake. Set this value to True if you want FAST to decrease (i.e., subtract) the wind velocity at the
tail fin center-of-pressure in the rotor shaft direction by the average rotor axial induction. This
input is only used when TFinMod is set to 1. (flag)

FAST User's Guide 87 Last updated August 2005 for version 6.0
 Table 14. ADAMS-Specific-Input-File Parameters.
The following input parameters are contained in the file indicated by input ADAMSFile from the primary input file.
FAST will only read this file if the FAST-to-ADAMS preprocessor is enabled (when ADAMSPrep from the
primary input file is set to 2 or 3).

Feature Flags
SaveGrphcs If set to True, this flag tells ADAMS to generate a graphics output file for viewing an animation of
the ADAMS simulation. Set this to False if you don’t want graphics output generated; this saves a
lot of hard disk space if the simulation is long. (flag)
MakeLINacf If set to True, this flag tells FAST to generate an ADAMS control/command file used to drive an
ADAMS/LINEAR eigenanalysis of the model. The eigenanalysis is performed with no gravity,
rotor speed, damping, or aerodynamics, no matter how the associated inputs are otherwise
specified in FAST’s other input file(s). Set to False if you don’t want this additional
control/command file generated. SaveGrphcs must be True if this input is True. (flag)

Damping Parameters
CRatioTGJ This is the ratio of the tower’s torsional damping to stiffness in ADAMS. A typical value is 0.01
and it must not be negative. (-)
CRatioTEA This is the ratio of the tower’s extensional damping to stiffness in ADAMS. A typical value is
0.01 and it must not be negative. (-)
CRatioBGJ This is the ratio of a blade’s torsional damping to stiffness in ADAMS. A typical value is 0.01
and it must not be negative. The same ratio is used for all blades. (-)
CRatioBEA This is the ratio of a blade’s extensional damping to stiffness in ADAMS. A typical value is 0.01
and it must not be negative. The same ratio is used for all blades. (-)

Blade Pitch Actuator Parameters

BPActrSpr This is the torsional spring stiffness of the blade pitch actuators in ADAMS. The linear blade
pitch spring moment is proportional to the pitch error through this constant. If a pitch actuator
natural frequency is known in place of an actuator spring stiffness, compute the spring stiffness as
follows: BPActrSpr = PitchIner•ωn2, where ωn is the natural frequency in rad/sec and PitchIner
is the nominal inertia of the blade about the pitch axis in kg·m2. The same stiffness is used for all
blade pitch actuators and it must not negative. (N·m/rad)
BPActrDmp This is the torsional damping constant of the blade pitch actuators in ADAMS. The linear blade
pitch damping moment is proportional to the blade pitch rate through this constant. If a pitch
actuator natural frequency and damping ratio are known in place of an actuator damping constant,
compute the damping constant as follows: BPActrDmp = 2•ζ•PitchIner•ωn, where ωn is the
natural frequency in rad/sec, ζ is the damping ratio in fraction of critical, and PitchIner is the
nominal inertia of the blade about the pitch axis in kg·m2. The same damping is used for all blade
pitch actuators and it must not be negative. (N·m/(rad/sec))

GRAPHICS Parameters
NSides This is the number of line segments ADAMS includes when drawing GRAPHICS cylinder and
frustum statements in graphical output. This value must not be negative. (-)
TwrBaseRad This is the radius of the tower base (at elevation TwrRBHt above base of the tower). It is used to
define GRAPHICS cylinders for depicting the linearly tapered tower in ADAMS’ graphical
output. This value must not be negative. (m)
TwrTopRad This is the radius of the tower-top (at elevation TowerHt). It is used to define GRAPHICS
cylinders for depicting the linearly tapered tower in ADAMS’ graphical output. This value must
not be negative. (m)

FAST User's Guide 88 Last updated August 2005 for version 6.0
 Table 14. ADAMS-Specific-Input-File Parameters (concluded).

GRAPHICS Parameters (concluded)

NacLength This is the length of the nacelle. It is used to define the nacelle GRAPHICS frustum statement for
ADAMS’ graphical output. The nacelle GRAPHICS is centered about the point of intersection
between the rotor shaft axis and the yn-/zn-plane. This value must not be negative or larger than
twice the magnitude of OverHang. (m)
NacRadBot This is the radius of the bottom of the nacelle (opposite side of rotor). It is used to define the
nacelle GRAPHICS frustum statement for ADAMS’ graphical output. This value must not be
negative. (m)
NacRadTop This is the radius of the top of the nacelle (same side as rotor). It is used to define the nacelle
GRAPHICS frustum statement for ADAMS’ graphical output. This value must not be negative.
(m)
GBoxLength This is the length, width, and height of a cube depicting the gearbox in ADAMS’ graphical output.
It is used to define the gearbox GRAPHICS box statement. The gearbox GRAPHICS is centered
about the point of intersection between the low- and high-speed shafts. This value must not be
negative. (m)
GenLength This is the length of the generator. It is used to define the length of a GRAPHICS cylinder
depicting the generator. The generator GRAPHICS extends from the end of the HSS. This value
must not be negative. (m)
HSSLength This is the length of the HSS. It is used to define the length of a GRAPHICS cylinder depicting
the HSS. The HSS GRAPHICS extends from the end of the LSS. The generator GRAPHICS
originates at the end of the HSS opposite the rotor. This value must not be negative. (m)
LSSLength This is the length of the LSS. It is used to define the length of a GRAPHICS cylinder depicting
the LSS. The LSS GRAPHICS extends toward the tower from the teeter pin for two-bladed
turbines or from the rotor apex for three-bladed turbines. The HSS GRAPHICS originates at the
end of the LSS opposite the rotor. This value must not be negative. (m)
GenRad This is the radius of the generator. It is used to define the radius of a GRAPHICS cylinder
depicting the generator for ADAMS’ graphical output. This value must not be negative. (m)
HSSRad This is the radius of the HSS. It is used to define the radius of a GRAPHICS cylinder depicting
the HSS for ADAMS’ graphical output. This value must not be negative. (m)
LSSRad This is the radius of the LSS. It is used to define the radius of a GRAPHICS cylinder depicting
the LSS for ADAMS’ graphical output. This value must not be negative. (m)
HubCylRad This is the radius of the cylinder depicting the hub in ADAMS’ graphical output. It is used in the
hub GRAPHICS cylinder statements, which extend from the apex of the cone of rotation to the
blade roots along the pitch axes. This value must not be negative. (m)
ThkOvrChrd This is the ratio of blade thickness to blade chord for depicting the blade elements in ADAMS’
graphical output. It is used in the blade element GRAPHICS box statements. The same value is
used for each blade element of each blade and must not be negative. (m)
BoomRad This is the radius of the tail boom. It is used to define the radius of a GRAPHICS cylinder
depicting the tail boom for ADAMS’ graphical output. The tail boom GRAPHICS extends from
the specified point on the tail-furl axis (characterized by inputs TFrlPntxn, TFrlPntyn, and
TFrlPntzn) to a point just below the tail fin center-of-pressure. This value must not be negative.
(m)

FAST User's Guide 89 Last updated August 2005 for version 6.0
 Table 15. Linearization Control-Input-File Parameters.
The following input parameters are contained in the file indicated by input LinFile from the primary input file.
FAST will only read this file when a linearization is performed (when AnalMode from the primary input file is set
to 2).

Periodic Steady State Solution

CalcStdy This flag determines whether a periodic steady state solution is computed before linearizing the
model. If False, the next three inputs are ignored and the linearization occurs about the initial
conditions specified in FAST’s primary input file. That is, when CalcStdy is False, the operating
point is set to the condition in which all displacements, velocities, and accelerations are zero,
except those specified with nonzero initial conditions (for instance, the azimuth DOF will
increment at a constant rate if and when the rotor is spinning). If CalcStdy is True and
RotSpeed is nonzero, FAST integrates in time until a periodic steady state solution is reached.
The method of solution is determined by the next input, TrimCase. FAST is then linearized
about this periodic operating point. If CalcStdy is True and RotSpeed is zero, FAST will
disable GenDOF (if previously enabled) and integrate in time until a static equilibrium position is
found. FAST is then linearized about this position. The accuracy of the steady state solution is
determined through input convergence tolerances DispTol and VelTol (see below). This input is
not used in the FAST-to-ADAMS preprocessor. (flag)
TrimCase This switch determines, for a variable speed machine, which control input to trim in order to reach
the desired azimuth-averaged rotor speed indicated through input RotSpeed (which is also the
initial rotor speed). Setting it to 1 causes FAST to trim nacelle yaw command (demand) angle,
while maintaining constant rotor collective blade pitch (indicated by inputs BlPitchi), to reach the
desired azimuth-averaged rotor speed. With yaw DOF enabled (YawDOF = True), the nacelle
yaw command is the neutral yaw angle, YawNeut, which is passed through FAST’s built-in,
second-order actuator model. With yaw DOF disabled (YawDOF = False), the nacelle yaw
command is the actual nacelle yaw angle. Setting TrimCase to 2 causes FAST to trim electrical
generator torque, while maintaining constant rotor collective blade pitch (indicated by inputs
BlPitchi), to reach the desired azimuth-averaged rotor speed (i.e., Region 2 trim). Setting
TrimCase to 3 causes FAST to trim rotor collective blade pitch to reach the desired azimuth-
averaged rotor speed (i.e., Region 3 trim). In this case, the initial “guess” blade pitch angles are
given by BlPitchi and the electrical generator torque is determined by the torque-speed
relationship indicated by inputs VSContrl or GenModel. For typical Region 3 trim, collective
pitch can be trimmed while maintaining a constant generator torque by setting TrimCase to 3,
VSContrl to 1, VS_RtTq to the desired constant generator torque, and VS_RtGnSp,
VS_Rgn2K, and VS_SlPc to 9999.9E-9 (very small don’t cares > 0.0). Input parameter
TrimCase is ignored when either CalcStdy or GenDOF is False. For a constant speed machine,
GenDOF should be set to False when linearizing FAST, in which case, input TrimCase is
ignored. Using values other than 1, 2, or 3 will cause FAST to abort. This input is not used in the
FAST-to-ADAMS preprocessor. (switch)
DispTol This is the convergence tolerance for the 2-norm of angular displacements in the calculation of
periodic steady state solution. The steady state solution is found when this tolerance and VelTol
are both met. The smaller the number, the tighter the tolerance is. This input is ignored if
CalcStdy is False. This input is not used in the FAST-to-ADAMS preprocessor. (rad)
VelTol This is the convergence tolerance for the 2-norm of angular velocities in the calculation of the
periodic steady state solution. The steady state solution is found when this tolerance and DispTol
are both met. The smaller the number, the tighter the tolerance is. This input is ignored if
CalcStdy is False. This input is not used in the FAST-to-ADAMS preprocessor. (rad/s)

FAST User's Guide 90 Last updated August 2005 for version 6.0
 Table 15. Linearization Control-Input-File Parameters (concluded).

Model Linearization
NAzimStep This is the number of equally spaced rotor azimuth steps in the output periodic linearized model.
The first rotor azimuth location is always the initial azimuth position indicated by inputs Azimuth
and AzimB1Up. The subsequent azimuth steps increment in the direction of rotation. If
RotSpeed is zero, FAST will override NAzimStep and only linearize the model about the initial
azimuth position (as if NAzimStep was set to 1). This input is not used in the FAST-to-ADAMS
preprocessor. (-)
MdlOrder This is the order of the output linearized model. A setting of 1 causes FAST to output the first-
order representation of the linearized model. A setting of 2 causes FAST to output the second-
order representation of the linearized model. Using values other than 1 or 2 will cause FAST to
abort. This input is not used in the FAST-to-ADAMS preprocessor. (-)

Inputs and Disturbances

NInputs The number of control inputs indicates the number of input values on the next line. Valid values
are integers from 0 to 4 + NumBl (inclusive). This input is not used in the FAST-to-ADAMS
preprocessor. (-)
CntrlInpt This is a list of numbers corresponding to different types of control inputs. Possible values are 1
to 7 (inclusive) (7 is only available if NumBl = 3). The numbers correspond to seven control
inputs as follows: (1) nacelle yaw angle command, (2) nacelle yaw rate command, (3) electrical
generator torque, (4) rotor collective blade pitch, (5) individual pitch of blade 1, (6) individual
pitch of blade 2, and (7) individual pitch of blade 3 (unavailable if NumBl = 2). If the yaw DOF is
enabled (YawDOF = True), then the commanded yaw angle and rate from CntrlInpt setting 1 and
2 are the neutral yaw angle, YawNeut, and neutral yaw rate, YawRateNeut, in FAST's built-in
second-order actuator model. In this case, the yaw actuator, which is described in the Nacelle
Yaw Control section of the Controls chapter, will be inherent in the output linearized model. If
the yaw DOF is disabled (YawDOF = False), then the commanded yaw angle and rate from
CntrlInpt setting 1 and 2 are the actual yaw angle and yaw rate. In this case, the yaw actuator will
be absent from the output linearized model. You must enter at least Ninputs values on the line of
input CntrlInpt. If NInputs is 0, this line will be skipped, but you must have a line taking up
space in the input file. You can separate the values with combinations of tabs, spaces, and
commas, but you may use only one comma between numbers. This input is not used in the FAST-
to-ADAMS preprocessor. (-)
NDisturbs The number of wind input disturbances indicates the number of input values on the next line.
Valid values are integers from 0 to 7 (inclusive). This input is not used in the FAST-to-ADAMS
preprocessor. (-)
Disturbnc This is a list of numbers corresponding to different types of wind input disturbances. Possible
values are 1 to 7 (inclusive). The numbers correspond to the seven inputs available in the hub-
height wind data files of AeroDyn as follows: (1) horizontal hub-height wind speed, V, (2)
horizontal wind direction, DELTA, (3) vertical wind speed, VZ, (4) horizontal wind shear, HSHR,
(5) vertical power law wind shear, VSHR, (6) linear vertical wind shear, VLinSHR, and (7)
horizontal hub-height wind gust, VG. You must enter at least NDisturbs values on this line. If
NDisturbs is 0, this line will be skipped, but you must have a line taking up space in the input file.
You can separate the values with combinations of tabs, spaces, and commas, but you may use only
one comma between numbers. This input is not used in the FAST-to-ADAMS preprocessor. (-)

FAST User's Guide 91 Last updated August 2005 for version 6.0
FAST User's Guide 92 Last updated August 2005 for version 6.0
 OUTPUT FILES

The program generates one or more output files named fast.fsm. This file contains some of the basic
based on settings in the input file. input file parameters and computed inertia properties of
For time-marching analyses, the primary output the blades and tower. An example summary file is
file contains columns of time-series data with one shown in Figure 32.
column for each parameter that is requested in the If the SumPrint flag is set to True, AeroDyn also
primary input file. The name of this file uses the path generates a summary file that contains blade-element
and root name of the primary input file and appends geometry data, airfoil data files at the corresponding
.out for an extension. For example, if the input file blade element, and the summary of combined
were named fast.fst, the main output file will be named FAST/AeroDyn input parameters. In the above
fast.out. The available output parameters are shown in example, this file will be named fast.opt. An example
Table 16 through Table 44 and are also documented in of this AeroDyn output can be found in Figure 33.
the OutList.txt file of the FAST archive. An example If ADAMSPrep is set to 2 or 3, FAST generates
output file is shown in Figure 30. ADAMS dataset files corresponding to the model
In some situations, some output channels are configuration and analysis settings specified in the
meaningless. For instance, if aerodynamic calculations FAST input file(s). See the ADAMS Preprocessor
are disabled, parameters such as the wind speed are chapter for a description of these output files.
invalid. You can still leave those parameters in your A final file is generated only when the word
output list, but the data generated will be all zeros. The “PRINT” is found on one or more of the lines defining
name and units for the channel will also be replaced the blade elements in the AeroDyn input file. This file
with “INVALID” and “CHANNEL” respectively. contains a time series of aerodynamic data and has a
Output loads and motions follow the IEC system. .elm extension, as in, fast.elm. Please see the AeroDyn
Please refer to Figure 3 through Figure 9 to get a sense User’s Guide [1] for details on this file.
for which directions are positive. When running FAST within Simulink, the output
For linearization analyses, the primary output file file names use the root name of the primary input file
provides the periodic state matrices of the linearized and append _SFunc to the name. For example, if the
model. The name of this file uses the path and root primary input file were named fast.fst, the main output
name of the primary input file and appends .lin for an file from the FAST S-Function will be named
extension. For example, if the input file were named fast_SFunc.out whereas the the FAST executable
fast.fst, the main output file will be named fast.lin. An would generate fast.out. Please see the Simulink
example linearized model file is shown in Figure 31. Interface chapter for further details.
If the SumPrint flag is set to True, FAST
generates a second output file, with a .fsm for an
extension. In the above example, this file will be

FAST User's Guide 93 Last updated August 2005 for version 6.0
 Table 16. Output Parameters for Wind Motions.

Name Other Name(s) Description Convention Units

Nominally downwind component of the hub-height Directed along the
WindVxi uWind (m/sec)
wind velocity (unavailable if CompAero is False) xi-axis
Cross-wind component of the hub-height wind Directed along the
WindVyi vWind (m/sec)
velocity (unavailable if CompAero is False) yi-axis
Vertical component of the hub-height wind velocity Directed along the
WindVzi wWind (m/sec)
(unavailable if CompAero is False) zi-axis
Total hub-height wind speed magnitude
TotWindV N/A (m/sec)
(unavailable if CompAero is False)
Horizontal hub-height wind speed magnitude In the xi- and yi-
HorWindV (m/sec)
(unavailable if CompAero is False) plane
Horizontal hub-height wind direction. Please note
that FAST uses the opposite of the sign convention
that AeroDyn uses. Put a “-“, “_”, “m”, or “M”
HorWndDir character in front of this variable name in the input About the zi-axis (deg)
file to change its sign if you want to use the
AeroDyn convention. (unavailable if CompAero is
False)
About an axis
Vertical hub-height wind direction (unavailable if orthogonal to the zi-
VerWndDir (deg)
CompAero is False) axis and the
HorWindV-vector

FAST User's Guide 94 Last updated August 2005 for version 6.0
 Table 17. Output Parameters for Blade 1 Tip Motions.
Other
Name Description Convention Units
Name(s)
Blade 1 out-of-plane tip deflection (relative to the Directed along the
TipDxc1 OoPDefl1 (m)
pitch axis) xc,1-axis
Blade 1 in-plane tip deflection (relative to the pitch Directed along the
TipDyc1 IPDefl1 (m)
axis) yc,1-axis
Blade 1 axial tip deflection (relative to the pitch Directed along the
TipDzc1 TipDzb1 (m)
axis) zc,1- and zb,1-axes
Blade 1 flapwise tip deflection (relative to the pitch Directed along the
TipDxb1 (m)
axis) xb,1-axis
Blade 1 edgewise tip deflection (relative to the pitch Directed along the
TipDyb1 (m)
axis) yb,1-axis
Directed along the
TipALxb1 Blade 1 local flapwise tip acceleration (absolute) (m/sec^2)
local xb,1-axis
Directed along the
TipALyb1 Blade 1 local edgewise tip acceleration (absolute) (m/sec^2)
local yb,1-axis
Directed along the
TipALzb1 Blade 1 local axial tip acceleration (absolute) (m/sec^2)
local zb,1-axis
Blade 1 roll (angular/rotational) tip deflection
(relative to the undeflected position). In ADAMS,
it is output as an Euler angle computed as the 3rd
TipRDxb1 RollDefl1 rotation in the yaw-pitch-roll rotation sequence. It About the xb,1-axis (deg)
is not output as an Euler angle in FAST, which
assumes small blade deflections, so that the rotation
sequence does not matter.
Blade 1 pitch (angular/rotational) tip deflection
(relative to the undeflected position). In ADAMS,
it is output as an Euler angle computed as the 2nd
TipRDyb1 PtchDefl1 rotation in the yaw-pitch-roll rotation sequence. It About the yb,1-axis (deg)
is not output as an Euler angle in FAST, which
assumes small blade deflections, so that the rotation
sequence does not matter.
Blade 1 torsional tip deflection (relative to the
undeflected position). This output will always be
zero for FAST simulation results. Use it for
examining blade torsional deflections of ADAMS
simulations run using ADAMS datasets created
TipRDzb1 About the zc,1- and
TipRDzc1 using the FAST-to-ADAMS preprocessor. In (deg)
TwstDefl1 zb,1-axes
ADAMS, it is output as an Euler angle computed as
the 1st rotation in the yaw-pitch-roll rotation
sequence. Please note that this output uses the
opposite of the sign convention used for blade pitch
angles.
Blade 1 tip-to-tower clearance estimate. This is
computed as the perpendicular distance from the
yaw axis to the tip of blade 1 when the blade tip is
below the yaw bearing. When the tip of blade 1 is
TwrClrnc1
TipClrnc1 above the yaw bearing, it is computed as the N/A (m)
Tip2Twr1
absolute distance from the yaw bearing to the blade
tip. Please note that you should reduce this value by
the tower radius to obtain the actual tower
clearance.

FAST User's Guide 95 Last updated August 2005 for version 6.0
 Table 18. Output Parameters for Blade 2* Tip Motions.
Other
Name Description Convention Units
Name(s)
Blade 2 out-of-plane tip deflection (relative to the Directed along the
TipDxc2 OoPDefl2 (m)
pitch axis) xc,2-axis
Blade 2 in-plane tip deflection (relative to the pitch Directed along the
TipDyc2 IPDefl2 (m)
axis) yc,2-axis
Blade 2 axial tip deflection (relative to the pitch Directed along the
TipDzc2 TipDzb2 (m)
axis) zc,2- and zb,2-axes
Blade 2 flapwise tip deflection (relative to the pitch Directed along the
TipDxb2 (m)
axis) xb,2-axis
Blade 2 edgewise tip deflection (relative to the pitch Directed along the
TipDyb2 (m)
axis) yb,2-axis
Directed along the
TipALxb2 Blade 2 local flapwise tip acceleration (absolute) (m/sec^2)
local xb,2-axis
Directed along the
TipALyb2 Blade 2 local edgewise tip acceleration (absolute) (m/sec^2)
local yb,2-axis
Directed along the
TipALzb2 Blade 2 local axial tip acceleration (absolute) (m/sec^2)
local zb,2-axis
Blade 2 roll (angular/rotational) tip deflection
(relative to the undeflected position). In ADAMS,
it is output as an Euler angle computed as the 3rd
TipRDxb2 RollDefl2 rotation in the yaw-pitch-roll rotation sequence. It About the xb,2-axis (deg)
is not output as an Euler angle in FAST, which
assumes small blade deflections, so that the rotation
sequence does not matter.
Blade 2 pitch (angular/rotational) tip deflection
(relative to the undeflected position). In ADAMS,
it is output as an Euler angle computed as the 2nd
TipRDyb2 PtchDefl2 rotation in the yaw-pitch-roll rotation sequence. It About the yb,2-axis (deg)
is not output as an Euler angle in FAST, which
assumes small blade deflections, so that the rotation
sequence does not matter.
Blade 2 torsional tip deflection (relative to the
undeflected position). This output will always be
zero for FAST simulation results. Use it for
examining blade torsional deflections of ADAMS
simulations run using ADAMS datasets created
TipRDzb2 About the zc,2- and
TipRDzc2 using the FAST-to-ADAMS preprocessor. In (deg)
TwstDefl2 zb,2-axes
ADAMS, it is output as an Euler angle computed as
the 1st rotation in the yaw-pitch-roll rotation
sequence. Please note that this output uses the
opposite of the sign convention used for blade pitch
angles.
Blade 2 tip-to-tower clearance estimate. This is
computed as the perpendicular distance from the
TwrClrnc2 yaw axis to the tip of blade 2 when the blade tip is
TipClrnc2 N/A (m)
Tip2Twr2 below the yaw bearing. When the tip of blade 2 is
above the yaw bearing, it is computed as the
absolute distance from the yaw bearing to the blade

*
For three-bladed rotors, blade 3 is ahead of blade 2, which is ahead of blade 1, so that the order of blades passing through a
given azimuth is 3-2-1-repeat.

FAST User's Guide 96 Last updated August 2005 for version 6.0
 Other
Name Description Convention Units
Name(s)
tip. Please note that you should reduce this value by
the tower radius to obtain the actual tower
clearance.

FAST User's Guide 97 Last updated August 2005 for version 6.0
 Table 19. Output Parameters for Blade 3* Tip Motions.
Other
Name Description Convention Units
Name(s)
Blade 3 out-of-plane tip deflection (relative to the Directed along the
TipDxc3 OoPDefl3 (m)
pitch axis) (unavailable for two-bladed turbines) xc,3-axis
Blade 3 in-plane tip deflection (relative to the pitch Directed along the
TipDyc3 IPDefl3 (m)
axis) (unavailable for two-bladed turbines) yc,3-axis
Blade 3 axial tip deflection (relative to the pitch Directed along the
TipDzc3 TipDzb3 (m)
axis) (unavailable for two-bladed turbines) zc,3- and zb,3-axes
Blade 3 flapwise tip deflection (relative to the pitch Directed along the
TipDxb3 (m)
axis) (unavailable for two-bladed turbines) xb,3-axis
Blade 3 edgewise tip deflection (relative to the pitch Directed along the
TipDyb3 (m)
axis) (unavailable for two-bladed turbines) yb,3-axis
Blade 3 local flapwise tip acceleration (absolute) Directed along the
TipALxb3 (m/sec^2)
(unavailable for two-bladed turbines) local xb,3-axis
Blade 3 local edgewise tip acceleration (absolute) Directed along the
TipALyb3 (m/sec^2)
(unavailable for two-bladed turbines) local yb,3-axis
Blade 3 local axial tip acceleration (absolute) Directed along the
TipALzb3 (m/sec^2)
(unavailable for two-bladed turbines) local zb,3-axis
Blade 3 roll (angular/rotational) tip deflection
(relative to the undeflected position). In ADAMS,
it is output as an Euler angle computed as the 3rd
rotation in the yaw-pitch-roll rotation sequence. It
TipRDxb3 RollDefl3 About the xb,3-axis (deg)
is not output as an Euler angle in FAST, which
assumes small blade deflections, so that the rotation
sequence does not matter. (unavailable for two-
bladed turbines)
Blade 3 pitch (angular/rotational) tip deflection
(relative to the undeflected position). In ADAMS,
it is output as an Euler angle computed as the 2nd
rotation in the yaw-pitch-roll rotation sequence. It
TipRDyb3 PtchDefl3 About the yb,3-axis (deg)
is not output as an Euler angle in FAST, which
assumes small blade deflections, so that the rotation
sequence does not matter. (unavailable for two-
bladed turbines)
Blade 3 torsional tip deflection (relative to the
undeflected position). This output will always be
zero for FAST simulation results. Use it for
examining blade torsional deflections of ADAMS
simulations run using ADAMS datasets created
TipRDzb3 About the zc,3- and
TipRDzc3 using the FAST-to-ADAMS preprocessor. In (deg)
TwstDefl3 zb,3-axes
ADAMS, it is output as an Euler angle computed as
the 1st rotation in the yaw-pitch-roll rotation
sequence. Please note that this output uses the
opposite of the sign convention used for blade pitch
angles. (unavailable for two-bladed turbines)
Blade 3 tip-to-tower clearance estimate. This is
TwrClrnc3
TipClrnc3 computed as the perpendicular distance from the N/A (m)
Tip2Twr3
yaw axis to the tip of blade 3 when the blade tip is

*
For three-bladed rotors, blade 3 is ahead of blade 2, which is ahead of blade 1, so that the order of blades passing through a
given azimuth is 3-2-1-repeat.

FAST User's Guide 98 Last updated August 2005 for version 6.0
 Other
Name Description Convention Units
Name(s)
below the yaw bearing. When the tip of blade 3 is
above the yaw bearing, it is computed as the
absolute distance from the yaw bearing to the blade
tip. Please note that you should reduce this value
by the tower radius to obtain the actual tower
clearance. (unavailable for two-bladed turbines)

FAST User's Guide 99 Last updated August 2005 for version 6.0
 Table 20. Output Parameters for Blade 1 Local Span Motions*.
Other
Name Description Convention Units
Name(s)
Blade 1 local flapwise acceleration (absolute) of Directed along the
Spn1ALxb1 (m/sec^2)
span station 1 (unavailable if NBlGages = 0) local xb,1-axis
Blade 1 local edgewise acceleration (absolute) of Directed along the
Spn1ALyb1 (m/sec^2)
span station 1 (unavailable if NBlGages = 0) local yb,1-axis
Blade 1 local axial acceleration (absolute) of span Directed along the
Spn1ALzb1 (m/sec^2)
station 1 (unavailable if NBlGages = 0) local zb,1-axis
Blade 1 local flapwise acceleration (absolute) of Directed along the
Spn2ALxb1 (m/sec^2)
span station 2 (unavailable if NBlGages < 2) local xb,1-axis
Blade 1 local edgewise acceleration (absolute) of Directed along the
Spn2ALyb1 (m/sec^2)
span station 2 (unavailable if NBlGages < 2) local yb,1-axis
Blade 1 local axial acceleration (absolute) of span Directed along the
Spn2ALzb1 (m/sec^2)
station 2 (unavailable if NBlGages < 2) local zb,1-axis
Blade 1 local flapwise acceleration (absolute) of Directed along the
Spn3ALxb1 (m/sec^2)
span station 3 (unavailable if NBlGages < 3) local xb,1-axis
Blade 1 local edgewise acceleration (absolute) of Directed along the
Spn3ALyb1 (m/sec^2)
span station 3 (unavailable if NBlGages < 3) local yb,1-axis
Blade 1 local axial acceleration (absolute) of span Directed along the
Spn3ALzb1 (m/sec^2)
station 3 (unavailable if NBlGages < 3) local zb,1-axis
Blade 1 local flapwise acceleration (absolute) of Directed along the
Spn4ALxb1 (m/sec^2)
span station 4 (unavailable if NBlGages < 4) local xb,1-axis
Blade 1 local edgewise acceleration (absolute) of Directed along the
Spn4ALyb1 (m/sec^2)
span station 4 (unavailable if NBlGages < 4) local yb,1-axis
Blade 1 local axial acceleration (absolute) of span Directed along the
Spn4ALzb1 (m/sec^2)
station 4 (unavailable if NBlGages < 4) local zb,1-axis
Blade 1 local flapwise acceleration (absolute) of Directed along the
Spn5ALxb1 (m/sec^2)
span station 5 (unavailable if NBlGages < 5) local xb,1-axis
Blade 1 local edgewise acceleration (absolute) of Directed along the
Spn5ALyb1 (m/sec^2)
span station 5 (unavailable if NBlGages < 5) local yb,1-axis
Blade 1 local axial acceleration (absolute) of span Directed along the
Spn5ALzb1 (m/sec^2)
station 5 (unavailable if NBlGages < 5) local zb,1-axis

*
These motions are for the nodes you specify with the BldGagNd input array.

FAST User's Guide 100 Last updated August 2005 for version 6.0
 Table 21. Output Parameters for Blade* Pitch Motions.
Other
Name Description Convention Units
Name(s)
Positive towards
PtchPMzb1
feather about the
PtchPMzc1 BldPitch1 Blade 1 pitch angle (position) (deg)
minus zc,1- and
BlPitch1
minus zb,1-axes
Positive towards
PtchPMzb2
feather about the
PtchPMzc2 BldPitch2 Blade 2 pitch angle (position) (deg)
minus zc,2- and
BlPitch2
minus zb,2-axes
Positive towards
PtchPMzb3
Blade 3 pitch angle (position) (unavailable for two- feather about the
PtchPMzc3 BldPitch3 (deg)
bladed turbines) minus zc,3- and
BlPitch3
minus zb,3-axes

Table 22. Output Parameters for Teeter Motions.

Other
Name Description Convention Units
Name(s)
RotTeetP Rotor teeter angle (position) (unavailable for three-
TeetPya About the ya-axis (deg)
TeetDefl bladed turbines)
Rotor teeter angular velocity (unavailable for
TeetVya RotTeetV About the ya-axis (deg/sec)
three-bladed turbines)
Rotor teeter angular acceleration (unavailable for
TeetAya RotTeetA About the ya-axis (deg/sec^2)
three-bladed turbines)

*
For three-bladed rotors, blade 3 is ahead of blade 2, which is ahead of blade 1, so that the order of blades passing through a
given azimuth is 3-2-1-repeat.

FAST User's Guide 101 Last updated August 2005 for version 6.0
 Table 23. Output Parameters for Shaft Motions.
Other
Name Description Convention Units
Name(s)
LSSTipPxs
About the xa- and
LSSTipPxa LSSTipP Rotor azimuth angle (position) (deg)
xs-axes
Azimuth
LSSTipVxs
About the xa- and
LSSTipVxa LSSTipV Rotor azimuth angular speed (rpm)
xs-axes
RotSpeed
LSSTipAxs
About the xa- and
LSSTipAxa LSSTipA Rotor azimuth angular acceleration (deg/sec^2)
xs-axes
RotAccel
LSSGagPxs LSS strain-gage azimuth angle (position) (on the About the xa- and
LSSGagPxa (deg)
LSSGagP gearbox side of the LSS) xs-axes
LSSGagVxs LSS strain-gage angular speed (on the gearbox About the xa- and
LSSGagVxa (rpm)
LSSGagV side of the LSS) xs-axes
LSSGagAxs LSS strain-gage angular acceleration (on the About the xa- and
LSSGagAxa (deg/sec^2)
LSSGagA gearbox side of the LSS) xs-axes
Same sign as
LSSGagVxa /
HSShftV GenSpeed Angular speed of the HSS and generator (rpm)
LSSGagVxs /
LSSGagV
Same sign as
LSSGagAxa /
HSShftA GenAccel Angular acceleration of the HSS and generator (deg/sec^2)
LSSGagAxs /
LSSGagA
Rotor blade tip speed ratio (unavailable if
TipSpdRat TSR N/A (-)
CompAero is False)

FAST User's Guide 102 Last updated August 2005 for version 6.0
 Table 24. Output Parameters for Nacelle Inertial Measurement Unit Motions*.
Other
Name Description Convention Units
Name(s)
Nacelle inertial measurement unit translational Directed along the
NcIMUTVxs (m/sec)
velocity (absolute) xs-axis
Nacelle inertial measurement unit translational Directed along the
NcIMUTVys (m/sec)
velocity (absolute) ys-axis
Nacelle inertial measurement unit translational Directed along the
NcIMUTVzs (m/sec)
velocity (absolute) zs-axis
Nacelle inertial measurement unit translational Directed along the
NcIMUTAxs (m/sec^2)
acceleration (absolute) xs-axis
Nacelle inertial measurement unit translational Directed along the
NcIMUTAys (m/sec^2)
acceleration (absolute) ys-axis
Nacelle inertial measurement unit translational Directed along the
NcIMUTAzs (m/sec^2)
acceleration (absolute) zs-axis
Nacelle inertial measurement unit angular
NcIMURVxs About the xs-axis (deg/sec)
(rotational) velocity (absolute)
Nacelle inertial measurement unit angular
NcIMURVys About the ys-axis (deg/sec)
(rotational) velocity (absolute)
Nacelle inertial measurement unit angular
NcIMURVzs About the zs-axis (deg/sec)
(rotational) velocity (absolute)
Nacelle inertial measurement unit angular
NcIMURAxs About the xs-axis (deg/sec^2)
(rotational) acceleration (absolute)
Nacelle inertial measurement unit angular
NcIMURAys About the ys-axis (deg/sec^2)
(rotational) acceleration (absolute)
Nacelle inertial measurement unit angular
NcIMURAzs About the zs-axis (deg/sec^2)
(rotational) acceleration (absolute)

*
The location of the nacelle inertial measurement unit is determined by inputs NcIMUxn, NcIMUyn, and NcIMUzn.

FAST User's Guide 103 Last updated August 2005 for version 6.0
 Table 25. Output Parameters for Rotor-Furl Motions.
Other
Name Description Convention Units
Name(s)
About the rotor-furl
RotFurlP RotFurl Rotor-furl angle (position) (deg)
axis (see Figure 17)
About the rotor-furl
RotFurlV Rotor-furl angular velocity (deg/sec)
axis (see Figure 17)
About the rotor-furl
RotFurlA Rotor-furl angular acceleration (deg/sec^2)
axis (see Figure 17)

Table 26. Output Parameters for Tail-Furl Motions.

Other
Name Description Convention Units
Name(s)
About the tail-furl
TailFurlP TailFurl Tail-furl angle (position) (deg)
axis (see Figure 17)
About the tail-furl
TailFurlV Tail -furl angular velocity (deg/sec)
axis (see Figure 17)
About the tail-furl
TailFurlA Tail -furl angular acceleration (deg/sec^2)
axis (see Figure 17)

Table 27. Output Parameters for Nacelle Yaw Motions.

Other
Name Description Convention Units
Name(s)
YawPzp
NacYawP About the zn- and
YawPzn Nacelle yaw angle (position) (deg)
NacYaw zp-axes
YawPos
YawVzp
About the zn- and
YawVzn NacYawV Nacelle yaw angular velocity (deg/sec)
zp-axes
YawRate
YawAzp
About the zn- and
YawAzn NacYawA Nacelle yaw angular acceleration (deg/sec^2)
zp-axes
YawAccel
Nacelle yaw error estimate. This is computed as
follows: NacYawErr = HorWndDir - YawPzn -
YawBrRDzt - PtfmRDzi. This estimate is not
accurate instantaneously in the presence of
significant tower deflection or platform angular
(rotational) displacement since the angles used in
NacYawErr the computation are not all defined about the same About the zi-axis (deg)
axis of rotation. However, the estimate should be
useful in a yaw controller if averaged over a time
scale long enough to diminish the effects of tower
and platform motions (i.e., much longer than the
period of oscillation). (unavailable if CompAero
= False)

FAST User's Guide 104 Last updated August 2005 for version 6.0
 Table 28. Output Parameters for Tower-Top, Yaw-Bearing Motions.
Other
Name Description Convention Units
Name(s)
Tower-top / yaw bearing fore-aft (translational) Directed along the
YawBrTDxp (m)
deflection (relative to the undeflected position) xp-axis
Tower-top / yaw bearing side-to-side
Directed along the
YawBrTDyp (translational) deflection (relative to the (m)
yp-axis
undeflected position)
Tower-top / yaw bearing axial (translational) Directed along the
YawBrTDzp (m)
deflection (relative to the undeflected position) zp-axis
Tower-top / yaw bearing fore-aft (translational) Directed along the
YawBrTDxt TTDspFA (m)
deflection (relative to the undeflected position) xt-axis
Tower-top / yaw bearing side-to-side
Directed along the
YawBrTDyt TTDspSS (translational) deflection (relative to the (m)
yt-axis
undeflected position)
Tower-top / yaw bearing axial (translational) Directed along the
YawBrTDzt TTDspAx (m)
deflection (relative to the undeflected position) zt-axis
Tower-top / yaw bearing fore-aft (translational) Directed along the
YawBrTAxp (m/sec^2)
acceleration (absolute) xp-axis
Tower-top / yaw bearing side-to-side Directed along the
YawBrTAyp (m/sec^2)
(translational) acceleration (absolute) yp-axis
Tower-top / yaw bearing axial (translational) Directed along the
YawBrTAzp (m/sec^2)
acceleration (absolute) zp-axis
Tower-top / yaw bearing angular (rotational) roll
deflection (relative to the undeflected position).
In ADAMS, it is output as an Euler angle
computed as the 3rd rotation in the yaw-pitch-roll
YawBrRDxt TTDspRoll About the xt-axis (deg)
rotation sequence. It is not output as an Euler
angle in FAST, which assumes small tower
deflections, so that the rotation sequence does not
matter.
Tower-top / yaw bearing angular (rotational)
pitch deflection (relative to the undeflected
position). In ADAMS, it is output as an Euler
angle computed as the 2nd rotation in the yaw-
YawBrRDyt TTDspPtch About the yt-axis (deg)
pitch-roll rotation sequence. It is not output as an
Euler angle in FAST, which assumes small tower
deflections, so that the rotation sequence does not
matter.
Tower-top / yaw bearing torsional deflection
(relative to the undeflected position). This output
will always be zero for FAST simulation results.
Use it for examining tower torsional deflections
YawBrRDzt TTDspTwst of ADAMS simulations run using ADAMS About the zt-axis (deg)
datasets created using the FAST-to-ADAMS
preprocessor. In ADAMS, it is output as an Euler
angle computed as the 1st rotation in the yaw-
pitch-roll rotation sequence.
Tower-top / yaw bearing angular (rotational) roll
YawBrRVxp About the xp-axis (deg/sec)
velocity (absolute)
Tower-top / yaw bearing angular (rotational)
YawBrRVyp About the yp-axis (deg/sec)
pitch velocity (absolute)

FAST User's Guide 105 Last updated August 2005 for version 6.0
 Other
Name Description Convention Units
Name(s)
Tower-top / yaw bearing angular (rotational)
torsion velocity. This output will always be very
close to zero for FAST simulation results. Use it
YawBrRVzp for examining tower torsional deflections of About the zp-axis (deg/sec)
ADAMS simulations run using ADAMS datasets
created using the FAST-to-ADAMS
preprocessor. (absolute)
Tower-top / yaw bearing angular (rotational) roll
YawBrRAxp About the xp-axis (deg/sec^2)
acceleration (absolute)
Tower-top / yaw bearing angular (rotational)
YawBrRAyp About the yp-axis (deg/sec^2)
pitch acceleration (absolute)
Tower-top / yaw bearing angular (rotational)
torsion acceleration. This output will always be
very close to zero for FAST simulation results.
YawBrRAzp Use it for examining tower torsional deflections About the zp-axis (deg/sec^2)
of ADAMS simulations run using ADAMS
datasets created using the FAST-to-ADAMS
preprocessor. (absolute)

FAST User's Guide 106 Last updated August 2005 for version 6.0
 Table 29. Output Parameters for Local Tower Motions*.
Other
Name Description Convention Units
Name(s)
Local tower fore-aft (translational) acceleration
Directed along the
TwHt1ALxt (absolute) of tower gage 1 (unavailable if (m/sec^2)
local xt-axis
NTwGages = 0)
Local tower side-to-side (translational) acceleration
Directed along the
TwHt1ALyt (absolute) of tower gage 1 (unavailable if (m/sec^2)
local yt-axis
NTwGages = 0)
Local tower axial (translational) acceleration
Directed along the
TwHt1ALzt (absolute) of tower gage 1 (unavailable if (m/sec^2)
local zt-axis
NTwGages = 0)
Local tower fore-aft (translational) acceleration
Directed along the
TwHt2ALxt (absolute) of tower gage 2 (unavailable if (m/sec^2)
local xt-axis
NTwGages < 2)
Local tower side-to-side (translational) acceleration
Directed along the
TwHt2ALyt (absolute) of tower gage 2 (unavailable if (m/sec^2)
local yt-axis
NTwGages < 2)
Local tower axial (translational) acceleration
Directed along the
TwHt2ALzt (absolute) of tower gage 2 (unavailable if (m/sec^2)
local zt-axis
NTwGages < 2)
Local tower fore-aft (translational) acceleration
Directed along the
TwHt3ALxt (absolute) of tower gage 3 (unavailable if (m/sec^2)
local xt-axis
NTwGages < 3)
Local tower side-to-side (translational) acceleration
Directed along the
TwHt3ALyt (absolute) of tower gage 3 (unavailable if (m/sec^2)
local yt-axis
NTwGages < 3)
Local tower axial (translational) acceleration
Directed along the
TwHt3ALzt (absolute) of tower gage 3 (unavailable if (m/sec^2)
local zt-axis
NTwGages < 3)
Local tower fore-aft (translational) acceleration
Directed along the
TwHt4ALxt (absolute) of tower gage 4 (unavailable if (m/sec^2)
local xt-axis
NTwGages < 4)
Local tower side-to-side (translational) acceleration
Directed along the
TwHt4ALyt (absolute) of tower gage 4 (unavailable if (m/sec^2)
local yt-axis
NTwGages < 4)
Local tower axial (translational) acceleration
Directed along the
TwHt4ALzt (absolute) of tower gage 4 (unavailable if (m/sec^2)
local zt-axis
NTwGages < 4)
Local tower fore-aft (translational) acceleration
Directed along the
TwHt5ALxt (absolute) of tower gage 5 (unavailable if (m/sec^2)
local xt-axis
NTwGages < 5)
Local tower side-to-side (translational) acceleration
Directed along the
TwHt5ALyt (absolute) of tower gage 5 (unavailable if (m/sec^2)
local yt-axis
NTwGages < 5)
Local tower axial (translational) acceleration
Directed along the
TwHt5ALzt (absolute) of tower gage 5 (unavailable if (m/sec^2)
local zt-axis
NTwGages < 5)

*
These motions are for the nodes you specify with the TwrGagNd input array.

FAST User's Guide 107 Last updated August 2005 for version 6.0
 Table 30. Output Parameters for Platform Motions.
Other
Name Description Convention Units
Name(s)
Platform horizontal surge (translational) Directed along the
PtfmTDxt (m)
displacement xt-axis
Platform horizontal sway (translational) Directed along the
PtfmTDyt (m)
displacement yt-axis
Platform vertical heave (translational) Directed along the
PtfmTDzt (m)
displacement zt-axis
Platform horizontal surge (translational) Directed along the
PtfmTDxi PtfmSurge (m)
displacement xi-axis
Platform horizontal sway (translational) Directed along the
PtfmTDyi PtfmSway (m)
displacement yi-axis
Platform vertical heave (translational) Directed along the
PtfmTDzi PtfmHeave (m)
displacement zi-axis
Directed along the
PtfmTVxt Platform horizontal surge (translational) velocity (m/sec)
xt-axis
Directed along the
PtfmTVyt Platform horizontal sway (translational) velocity (m/sec)
yt-axis
Directed along the
PtfmTVzt Platform vertical heave (translational) velocity (m/sec)
zt-axis
Directed along the
PtfmTVxi Platform horizontal surge (translational) velocity (m/sec)
xi-axis
Directed along the
PtfmTVyi Platform horizontal sway (translational) velocity (m/sec)
yi-axis
Directed along the
PtfmTVzi Platform vertical heave (translational) velocity (m/sec)
zi-axis
Platform horizontal surge (translational) Directed along the
PtfmTAxt (m/sec^2)
acceleration xt-axis
Platform horizontal sway (translational) Directed along the
PtfmTAyt (m/sec^2)
acceleration yt-axis
Directed along the
PtfmTAzt Platform vertical heave (translational) acceleration (m/sec^2)
zt-axis
Platform horizontal surge (translational) Directed along the
PtfmTAxi (m/sec^2)
acceleration xi-axis
Platform horizontal sway (translational) Directed along the
PtfmTAyi (m/sec^2)
acceleration yi-axis
Directed along the
PtfmTAzi Platform vertical heave (translational) acceleration (m/sec^2)
zi-axis
Platform roll tilt angular (rotational) displacement.
In ADAMS, it is output as an Euler angle
computed as the 3rd rotation in the yaw-pitch-roll
PtfmRDxi PtfmRoll rotation sequence. It is not output as an Euler About the xi-axis (deg)
angle in FAST, which assumes small rotational
platform displacements, so that the rotation
sequence does not matter.
Platform pitch tilt angular (rotational)
displacement. In ADAMS, it is output as an Euler
angle computed as the 2nd rotation in the yaw-
PtfmRDyi PtfmPitch pitch-roll rotation sequence. It is not output as an About the yi-axis (deg)
Euler angle in FAST, which assumes small
rotational platform displacements, so that the
rotation sequence does not matter.

FAST User's Guide 108 Last updated August 2005 for version 6.0
 Other
Name Description Convention Units
Name(s)
Platform yaw angular (rotational) displacement.
In ADAMS, it is output as an Euler angle
computed as the 1st rotation in the yaw-pitch-roll
PtfmRDzi PtfmYaw rotation sequence. It is not output as an Euler About the zi-axis (deg)
angle in FAST, which assumes small rotational
platform displacements, so that the rotation
sequence does not matter.
PtfmRVxt Platform roll tilt angular (rotational) velocity About the xt-axis (deg/sec)

PtfmRVyt Platform pitch tilt angular (rotational) velocity About the yt-axis (deg/sec)

PtfmRVzt Platform yaw angular (rotational) velocity About the zt-axis (deg/sec)

PtfmRVxi Platform roll tilt angular (rotational) velocity About the xi-axis (deg/sec)

PtfmRVyi Platform pitch tilt angular (rotational) velocity About the yi-axis (deg/sec)

PtfmRVzi Platform yaw angular (rotational) velocity About the zi-axis (deg/sec)

PtfmRAxt Platform roll tilt angular (rotational) acceleration About the xt-axis (deg/sec^2)

PtfmRAyt Platform pitch tilt angular (rotational) acceleration About the yt-axis (deg/sec^2)

PtfmRAzt Platform yaw angular (rotational) acceleration About the zt-axis (deg/sec^2)

PtfmRAxi Platform roll tilt angular (rotational) acceleration About the xi-axis (deg/sec^2)

PtfmRAyi Platform pitch tilt angular (rotational) acceleration About the yi-axis (deg/sec^2)

PtfmRAzi Platform yaw angular (rotational) acceleration About the zi-axis (deg/sec^2)

FAST User's Guide 109 Last updated August 2005 for version 6.0
 Table 31. Output Parameters for Blade 1 Root Loads.
Other
Name Description Convention Units
Name(s)
Directed along the
RootFxc1 Blade 1 out-of-plane shear force at the blade root (kN)
xc,1-axis
Directed along the
RootFyc1 Blade 1 in-plane shear force at the blade root (kN)
yc,1-axis
Directed along the
RootFzc1 RootFzb1 Blade 1 axial force at the blade root (kN)
zc,1- and zb,1-axes
Directed along the
RootFxb1 Blade 1 flapwise shear force at the blade root (kN)
xb,1-axis
Directed along the
RootFyb1 Blade 1 edgewise shear force at the blade root (kN)
yb,1-axis
Blade 1 in-plane moment (i.e., the moment caused
RootMxc1 RootMIP1 About the xc,1-axis (kN·m)
by in-plane forces) at the blade root
Blade 1 out-of-plane moment (i.e., the moment
RootMyc1 RootMOoP1 About the yc,1-axis (kN·m)
caused by out-of-plane forces) at the blade root
About the zc,1- and
RootMzc1 RootMzb1 Blade 1 pitching moment at the blade root (kN·m)
zb,1-axes
Blade 1 edgewise moment (i.e., the moment caused
RootMxb1 RootMEdg1 About the xb,1-axis (kN·m)
by edgewise forces) at the blade root
Blade 1 flapwise moment (i.e., the moment caused
RootMyb1 RootMFlp1 About the yb,1-axis (kN·m)
by flapwise forces) at the blade root

FAST User's Guide 110 Last updated August 2005 for version 6.0
 Table 32. Output Parameters for Blade 2* Root Loads.
Other
Name Description Convention Units
Name(s)
Directed along the
RootFxc2 Blade 2 out-of-plane shear force at the blade root (kN)
xc,2-axis
Directed along the
RootFyc2 Blade 2 in-plane shear force at the blade root (kN)
yc,2-axis
Directed along the
RootFzc2 RootFzb2 Blade 2 axial force at the blade root (kN)
zc,2- and zb,2-axes
Directed along the
RootFxb2 Blade 2 flapwise shear force at the blade root (kN)
xb,2-axis
Directed along the
RootFyb2 Blade 2 edgewise shear force at the blade root (kN)
yb,2-axis
Blade 2 in-plane moment (i.e., the moment caused
RootMxc2 RootMIP2 About the xc,2-axis (kN·m)
by in-plane forces) at the blade root
Blade 2 out-of-plane moment (i.e., the moment
RootMyc2 RootMOoP2 About the yc,2-axis (kN·m)
caused by out-of-plane forces) at the blade root
About the zc,2- and
RootMzc2 RootMzb2 Blade 2 pitching moment at the blade root (kN·m)
zb,2-axes
Blade 2 edgewise moment (i.e., the moment caused
RootMxb2 RootMEdg2 About the xb,2-axis (kN·m)
by edgewise forces) at the blade root
Blade 2 flapwise moment (i.e., the moment caused
RootMyb2 RootMFlp2 About the yb,2-axis (kN·m)
by flapwise forces) at the blade root

*
For three-bladed rotors, blade 3 is ahead of blade 2, which is ahead of blade 1, so that the order of blades passing through a
given azimuth is 3-2-1-repeat.

FAST User's Guide 111 Last updated August 2005 for version 6.0
 Table 33. Output Parameters for Blade 3* Root Loads.
Other
Name Description Convention Units
Name(s)
Blade 3 out-of-plane shear force at the blade root Directed along the
RootFxc3 (kN)
(unavailable for two-bladed turbines) xc,3-axis
Blade 3 in-plane shear force at the blade root Directed along the
RootFyc3 (kN)
(unavailable for two-bladed turbines) yc,3-axis
Blade 3 axial force at the blade root (unavailable for Directed along the
RootFzc3 RootFzb3 (kN)
two-bladed turbines) zc,3- and zb,3-axes
Blade 3 flapwise shear force at the blade root Directed along the
RootFxb3 (kN)
(unavailable for two-bladed turbines) xb,3-axis
Blade 3 edgewise shear force at the blade root Directed along the
RootFyb3 (kN)
(unavailable for two-bladed turbines) yb,3-axis
Blade 3 in-plane moment (i.e., the moment caused
RootMxc3 RootMIP3 by in-plane forces) at the blade root (unavailable for About the xc,3-axis (kN·m)
two-bladed turbines)
Blade 3 out-of-plane moment (i.e., the moment
RootMyc3 RootMOoP3 caused by out-of-plane forces) at the blade root About the yc,3-axis (kN·m)
(unavailable for two-bladed turbines)
Blade 3 pitching moment at the blade root About the zc,3- and
RootMzc3 RootMzb3 (kN·m)
(unavailable for two-bladed turbines) zb,3-axes
Blade 3 edgewise moment (i.e., the moment caused
RootMxb3 RootMEdg3 by edgewise forces) at the blade root (unavailable About the xb,3-axis (kN·m)
for two-bladed turbines)
Blade 3 flapwise moment (i.e., the moment caused
RootMyb3 RootMFlp3 by flapwise forces) at the blade root (unavailable for About the yb,3-axis (kN·m)
two-bladed turbines)

*
For three-bladed rotors, blade 3 is ahead of blade 2, which is ahead of blade 1, so that the order of blades passing through a
given azimuth is 3-2-1-repeat.

FAST User's Guide 112 Last updated August 2005 for version 6.0
 Table 34. Output Parameters for Blade 1 Local Span Loads*.
Other
Name Description Convention Units
Name(s)
Blade 1 local edgewise moment at span station 1 About the local xb,1-
Spn1MLxb1 (kN·m)
(unavailable if NBlGages = 0) axis
Blade 1 local flapwise moment at span station 1 About the local yb,1-
Spn1MLyb1 (kN·m)
(unavailable if NBlGages = 0) axis
Blade 1 local pitching moment at span station 1 About the local zb,1-
Spn1MLzb1 (kN·m)
(unavailable if NBlGages = 0) axis
Blade 1 local edgewise moment at span station 2 About the local xb,1-
Spn2MLxb1 (kN·m)
(unavailable if NBlGages < 2) axis
Blade 1 local flapwise moment at span station 2 About the local yb,1-
Spn2MLyb1 (kN·m)
(unavailable if NBlGages < 2) axis
Blade 1 local pitching moment at span station 2 About the local zb,1-
Spn2MLzb1 (kN·m)
(unavailable if NBlGages < 2) axis
Blade 1 local edgewise moment at span station 3 About the local xb,1-
Spn3MLxb1 (kN·m)
(unavailable if NBlGages < 3) axis
Blade 1 local flapwise moment at span station 3 About the local yb,1-
Spn3MLyb1 (kN·m)
(unavailable if NBlGages < 3) axis
Blade 1 local pitching moment at span station 3 About the local zb,1-
Spn3MLzb1 (kN·m)
(unavailable if NBlGages < 3) axis
Blade 1 local edgewise moment at span station 4 About the local xb,1-
Spn4MLxb1 (kN·m)
(unavailable if NBlGages < 4) axis
Blade 1 local flapwise moment at span station 4 About the local yb,1-
Spn4MLyb1 (kN·m)
(unavailable if NBlGages < 4) axis
Blade 1 local pitching moment at span station 4 About the local zb,1-
Spn4MLzb1 (kN·m)
(unavailable if NBlGages < 4) axis
Blade 1 local edgewise moment at span station 5 About the local xb,1-
Spn5MLxb1 (kN·m)
(unavailable if NBlGages < 5) axis
Blade 1 local flapwise moment at span station 5 About the local yb,1-
Spn5MLyb1 (kN·m)
(unavailable if NBlGages < 5) axis
Blade 1 local pitching moment at span station 5 About the local zb,1-
Spn5MLzb1 (kN·m)
(unavailable if NBlGages < 5) axis

*
These loads are for the nodes you specify with the BldGagNd input array.

FAST User's Guide 113 Last updated August 2005 for version 6.0
 Table 35. Output Parameters for Hub and Rotor Loads.
Other
Name Description Convention Units
Name(s)
LSShftFxs
LSSGagFxa LSS thrust force (this is constant along the shaft Directed along the
LSShftFxa (kN)
LSSGagFxs and is equivalent to the rotor thrust force) xa- and xs-axes
RotThrust
Rotating LSS shear force (this is constant along the Directed along the
LSShftFya LSSGagFya (kN)
shaft) ya-axis
Rotating LSS shear force (this is constant along the Directed along the
LSShftFza LSSGagFza (kN)
shaft) za-axis
Nonrotating LSS shear force (this is constant along Directed along the
LSShftFys LSSGagFys (kN)
the shaft) ys-axis
Nonrotating LSS shear force (this is constant along Directed along the
LSShftFzs LSSGagFzs (kN)
the shaft) zs-axis
LSShftMxs
LSSGagMxa
LSS torque (this is constant along the shaft and is About the xa- and
LSShftMxa LSSGagMxs (kN·m)
equivalent to the rotor torque) xs-axes
RotTorq
LSShftTq
Rotating LSS bending moment at the shaft tip
LSSTipMya (teeter pin for two-bladed turbines, apex of rotation About the ya-axis (kN·m)
for three-bladed turbines)
Rotating LSS bending moment at the shaft tip
LSSTipMza (teeter pin for two-bladed turbines, apex of rotation About the za-axis (kN·m)
for three-bladed turbines)
Nonrotating LSS bending moment at the shaft tip
LSSTipMys (teeter pin for two-bladed turbines, apex of rotation About the ys-axis (kN·m)
for three-bladed turbines)
Nonrotating LSS bending moment at the shaft tip
LSSTipMzs (teeter pin for two-bladed turbines, apex of rotation About the zs-axis (kN·m)
for three-bladed turbines)
Azimuth location of the center of thrust. This is
About the xa- and
CThrstAzm estimated using values of LSSTipMys, (deg)
xs-axes
LSSTipMzs, and RotThrust.
Dimensionless radial (arm) location of the center of
thrust. This is estimated using values of Always positive
LSSTipMys, LSSTipMzs, and RotThrust. (directed radially
CThrstRad CThrstArm (-)
(nondimensionalized using the undeflected tip outboard at azimuth
radius normal to the shaft and limited to values angle CThrstAzm)
between 0 and 1 (inclusive))
RotPwr LSShftPwr Rotor power (this is equivalent to the LSS power) N/A (kW)
Rotor torque coefficient (this is equivalent to the
RotCq LSShftCq LSS torque coefficient) (unavailable if CompAero N/A (-)
is False)
Rotor power coefficient (this is equivalent to the
RotCp LSShftCp LSS power coefficient) (unavailable if CompAero N/A (-)
is False)
Rotor thrust coefficient (this is equivalent to the
RotCt LSShftCt LSS thrust coefficient) (unavailable if CompAero N/A (-)
is False)

FAST User's Guide 114 Last updated August 2005 for version 6.0
 Table 36. Output Parameters for Shaft Strain-Gage Loads.
Other
Name Description Convention Units
Name(s)
Rotating LSS bending moment at the shaft's strain
LSSGagMya About the ya-axis (kN·m)
gage (shaft strain gage located by input ShftGagL)
Rotating LSS bending moment at the shaft's strain
LSSGagMza About the za-axis (kN·m)
gage (shaft strain gage located by input ShftGagL)
Nonrotating LSS bending moment at the shaft's
LSSGagMys strain gage (shaft strain gage located by input About the ys-axis (kN·m)
ShftGagL)
Nonrotating LSS bending moment at the shaft's
LSSGagMzs strain gage (shaft strain gage located by input About the zs-axis (kN·m)
ShftGagL)

Table 37. Output Parameters for Generator and HSS Loads.

Other
Name Description Convention Units
Name(s)
Same sign as
LSShftTq /
RotTorq /
HSShftTq HSS torque (this is constant along the shaft) LSShftMxa / (kN·m)
LSShftMxs /
LSSGagMxa /
LSSGagMxs
Same sign as
HSShftPwr HSS power (kW)
HSShftTq
HSS torque coefficient (unavailable if CompAero
HSShftCq N/A (-)
is False)
HSS power coefficient (unavailable if CompAero
HSShftCp N/A (-)
is False)
Positive reflects
power extracted and
negative represents a
GenTq Electrical generator torque (kN·m)
motoring-up
situation or power
input
Same sign as
GenPwr Electrical generator power (kW)
GenTq
Electrical generator torque coefficient (unavailable
GenCq N/A (-)
if CompAero is False)
Electrical generator power coefficient (unavailable
GenCp N/A (-)
if CompAero is False)
Always positive
HSS brake torque (i.e., the moment applied to the (indicating
HSSBrTq (kN·m)
HSS by the brake) dissipation of
power)

FAST User's Guide 115 Last updated August 2005 for version 6.0
 Table 38. Output Parameters for Rotor-Furl Bearing Loads.
Other
Name Description Convention Units
Name(s)
About the rotor-furl
RFrlBrM Rotor-furl bearing moment (kN·m)
axis (see Figure 17)

Table 39. Output Parameters for Tail-Furl Bearing Loads.

Other
Name Description Convention Units
Name(s)
About the tail-furl
TFrlBrM Tail-furl bearing moment (kN·m)
axis (see Figure 17)

Table 40. Output Parameters for Tail Fin Aerodynamic Loads.

Other
Name Description Convention Units
Name(s)
About the tail fin z-
Tail fin angle of attack. This is the angle between axis, which is the
the relative velocity of the wind-inflow at the tail fin axis in the tail fin
TFinAlpha (deg)
center-of-pressure and the tail fin chordline. plane normal to the
(unavailable if CompAero is False) chordline (see Figure
19)
Tail fin dimensionless lift coefficient (unavailable if
TFinCLift N/A (-)
CompAero is False)
Tail fin dimensionless drag coefficient (unavailable
TFinCDrag N/A (-)
if CompAero is False)
Tail fin dynamic pressure, equal to ½•Rho•Vrel2
where Vrel is the relative velocity of the wind-inflow
TFinDnPrs N/A (Pa)
at the tail fin center-of-pressure (unavailable if
CompAero is False)
Directed along the
tail fin x-axis, which
Tangential aerodynamic force at the tail fin center- is the axis along the
TFinCPFx (kN)
of-pressure (unavailable if CompAero is False) chordline, positive
towards the trailing
edge (see Figure 19)
Directed along the
tail fin y-axis, which
Normal aerodynamic force at the tail fin center-of-
TFinCPFy is orthogonal to the (kN)
pressure (unavailable if CompAero is False)
tail fin plane (see
Figure 19)

FAST User's Guide 116 Last updated August 2005 for version 6.0
 Table 41. Output Parameters for Tower-Top, Yaw-Bearing Loads.
Other
Name Description Convention Units
Name(s)
Rotating (with nacelle) tower-top / yaw bearing Directed along the
YawBrFxn (kN)
shear force xn-axis
Rotating (with nacelle) tower-top / yaw bearing Directed along the
YawBrFyn (kN)
shear force yn-axis
Directed along the
YawBrFzn YawBrFzp Tower-top / yaw bearing axial force (kN)
zn- and zp-axes
Tower-top / yaw bearing fore-aft (nonrotating) shear Directed along the
YawBrFxp (kN)
force xp-axis
Tower-top / yaw bearing side-to-side (nonrotating) Directed along the
YawBrFyp (kN)
shear force yp-axis
Rotating (with nacelle) tower-top / yaw bearing roll
YawBrMxn About the xn-axis (kN·m)
moment
Rotating (with nacelle) tower-top / yaw bearing
YawBrMyn About the yn-axis (kN·m)
pitch moment
YawBrMzp About the zn- and
YawBrMzn Tower-top / yaw bearing yaw moment (kN·m)
YawMom zp-axes
YawBrMxp Nonrotating tower-top / yaw bearing roll moment About the xp-axis (kN·m)

YawBrMyp Nonrotating tower-top / yaw bearing pitch moment About the yp-axis (kN·m)

Table 42. Output Parameters for Tower Base Loads.

Other
Name Description Convention Units
Name(s)
Directed along the
TwrBsFxt Tower base fore-aft shear force (kN)
xt-axis
Directed along the
TwrBsFyt Tower base side-to-side shear force (kN)
yt-axis
Directed along the
TwrBsFzt Tower base axial force (kN)
zt-axis
Tower base roll (or side-to-side) moment (i.e., the
TwrBsMxt About the xt-axis (kN·m)
moment caused by side-to-side forces)
Tower base pitching (or fore-aft) moment (i.e., the
TwrBsMyt About the yt-axis (kN·m)
moment caused by fore-aft forces)
TwrBsMzt Tower base yaw (or torsional) moment About the zt-axis (kN·m)

FAST User's Guide 117 Last updated August 2005 for version 6.0
 Table 43. Output Parameters for Local Tower Loads*.
Other
Name Description Convention Units
Name(s)
Local tower roll (or side-to-side) moment of tower About the local xt-
TwHt1MLxt (kN·m)
gage 1 (unavailable if NTwGages = 0) axis
Local tower pitching (or fore-aft) moment of tower About the local yt-
TwHt1MLyt (kN·m)
gage 1 (unavailable if NTwGages = 0) axis
Local tower yaw (or torsional) moment of tower About the local zt-
TwHt1MLzt (kN·m)
gage 1 (unavailable if NTwGages = 0) axis
Local tower roll (or side-to-side) moment of tower About the local xt-
TwHt2MLxt (kN·m)
gage 2 (unavailable if NTwGages < 2) axis
Local tower pitching (or fore-aft) moment of tower About the local yt-
TwHt2MLyt (kN·m)
gage 2 (unavailable if NTwGages < 2) axis
Local tower yaw (or torsional) moment of tower About the local zt-
TwHt2MLzt (kN·m)
gage 2 (unavailable if NTwGages < 2) axis
Local tower roll (or side-to-side) moment of tower About the local xt-
TwHt3MLxt (kN·m)
gage 3 (unavailable if NTwGages < 3) axis
Local tower pitching (or fore-aft) moment of tower About the local yt-
TwHt3MLyt (kN·m)
gage 3 (unavailable if NTwGages < 3) axis
Local tower yaw (or torsional) moment of tower About the local zt-
TwHt3MLzt (kN·m)
gage 3 (unavailable if NTwGages < 3) axis
Local tower roll (or side-to-side) moment of tower About the local xt-
TwHt4MLxt (kN·m)
gage 4 (unavailable if NTwGages < 4) axis
Local tower pitching (or fore-aft) moment of tower About the local yt-
TwHt4MLyt (kN·m)
gage 4 (unavailable if NTwGages < 4) axis
Local tower yaw (or torsional) moment of tower About the local zt-
TwHt4MLzt (kN·m)
gage 4 (unavailable if NTwGages < 4) axis
Local tower roll (or side-to-side) moment of tower About the local xt-
TwHt5MLxt (kN·m)
gage 5 (unavailable if NTwGages < 5) axis
Local tower pitching (or fore-aft) moment of tower About the local yt-
TwHt5MLyt (kN·m)
gage 5 (unavailable if NTwGages < 5) axis
Local tower yaw (or torsional) moment of tower About the local zt-
TwHt5MLzt (kN·m)
gage 5 (unavailable if NTwGages < 5) axis

*
These loads are for the nodes you specify with the TwrGagNd input array.

FAST User's Guide 118 Last updated August 2005 for version 6.0
 Table 44. Output Parameters for Platform Loads.
Other
Name Description Convention Units
Name(s)
Directed along the
PtfmFxt Platform horizontal surge shear force (kN)
xt-axis
Directed along the
PtfmFyt Platform horizontal sway shear force (kN)
yt-axis
Directed along the
PtfmFzt Platform vertical heave force (kN)
zt-axis
Directed along the
PtfmFxi Platform horizontal surge shear force (kN)
xi-axis
Directed along the
PtfmFyi Platform horizontal sway shear force (kN)
yi-axis
Directed along the
PtfmFzi Platform vertical heave force (kN)
zi-axis
PtfmMxt Platform roll tilt moment About the xt-axis (kN·m)

PtfmMyt Platform pitch tilt moment About the yt-axis (kN·m)

PtfmMzt Platform yaw moment About the zt-axis (kN·m)

PtfmMxi Platform roll tilt moment About the xi-axis (kN·m)

PtfmMyi Platform pitch tilt moment About the yi-axis (kN·m)

PtfmMzi Platform yaw moment About the zi-axis (kN·m)

FAST User's Guide 119 Last updated August 2005 for version 6.0
 These predictions were generated by FAST (v4.00, 09-Jul-2002) on 09-Jul-2002 at 09:38:47.

The aerodynamic calculations were made by AeroDyn (12.46, 23-May-2002).

FAST certification test #1 for AWT-27CR2 with many DOFs.

Time uWind Azimuth TeetDefl RootMyc1 RootMxc1 RotTorq YawBrMzn TTDspFA

(sec) (m/sec) (deg) (deg) (kN·m) (kN·m) (kN·m) (kN·m) (m)

10.000 1.039E+01 1.180E+01 1.031E+00 3.533E+01 2.039E+01 3.613E+01 -2.280E+00 4.922E-02

10.020 1.039E+01 1.831E+01 9.697E-01 3.642E+01 2.085E+01 3.558E+01 -1.996E+00 4.920E-02

10.040 1.039E+01 2.482E+01 8.946E-01 3.632E+01 2.235E+01 3.525E+01 -2.426E+00 4.920E-02

10.060 1.039E+01 3.134E+01 8.081E-01 3.538E+01 2.447E+01 3.514E+01 -3.286E+00 4.920E-02

10.080 1.039E+01 3.785E+01 7.116E-01 3.473E+01 2.672E+01 3.517E+01 -4.282E+00 4.918E-02

10.100 1.039E+01 4.436E+01 6.067E-01 3.503E+01 2.868E+01 3.526E+01 -5.124E+00 4.913E-02

10.120 1.039E+01 5.088E+01 4.943E-01 3.604E+01 3.011E+01 3.541E+01 -5.681E+00 4.906E-02

10.140 1.039E+01 5.739E+01 3.751E-01 3.707E+01 3.110E+01 3.565E+01 -5.993E+00 4.900E-02

10.160 1.039E+01 6.391E+01 2.498E-01 3.759E+01 3.191E+01 3.600E+01 -6.148E+00 4.897E-02

10.180 1.039E+01 7.042E+01 1.198E-01 3.769E+01 3.271E+01 3.642E+01 -6.184E+00 4.896E-02

10.200 1.039E+01 7.694E+01 -1.301E-02 3.777E+01 3.353E+01 3.684E+01 -6.121E+00 4.896E-02

10.220 1.039E+01 8.345E+01 -1.463E-01 3.813E+01 3.424E+01 3.720E+01 -5.908E+00 4.895E-02

10.240 1.039E+01 8.997E+01 -2.775E-01 3.868E+01 3.465E+01 3.745E+01 -5.546E+00 4.893E-02

10.260 1.039E+01 9.649E+01 -4.041E-01 3.916E+01 3.468E+01 3.764E+01 -5.027E+00 4.892E-02

10.280 1.039E+01 1.030E+02 -5.241E-01 3.939E+01 3.431E+01 3.777E+01 -4.365E+00 4.892E-02

10.300 1.039E+01 1.095E+02 -6.357E-01 3.942E+01 3.365E+01 3.787E+01 -3.590E+00 4.895E-02

10.320 1.039E+01 1.160E+02 -7.378E-01 3.945E+01 3.276E+01 3.791E+01 -2.757E+00 4.899E-02

10.340 1.039E+01 1.226E+02 -8.296E-01 3.959E+01 3.170E+01 3.791E+01 -1.938E+00 4.902E-02

10.360 1.039E+01 1.291E+02 -9.103E-01 3.985E+01 3.045E+01 3.788E+01 -1.210E+00 4.905E-02

10.380 1.039E+01 1.356E+02 -9.793E-01 4.009E+01 2.902E+01 3.783E+01 -6.359E-01 4.907E-02

10.400 1.039E+01 1.421E+02 -1.036E+00 4.028E+01 2.743E+01 3.777E+01 -1.897E-01 4.909E-02

10.420 1.039E+01 1.486E+02 -1.079E+00 4.039E+01 2.570E+01 3.772E+01 1.200E-01 4.912E-02

10.440 1.039E+01 1.551E+02 -1.107E+00 4.055E+01 2.384E+01 3.768E+01 2.992E-01 4.914E-02

10.460 1.039E+01 1.617E+02 -1.121E+00 4.054E+01 2.179E+01 3.762E+01 2.539E-01 4.917E-02

Figure 30. Sample output file.

FAST User's Guide 120 Last updated August 2005 for version 6.0
 This linearized model file was generated by FAST (v6.00c-jmj, 15-Apr-2005) on 15-Apr-2005 at 11:24:16.

The aerodynamic calculations were made by AeroDyn (12.57, 29-Sept-2004).

FAST model of a 1.5 MW 3-bladed upwind baseline turbine.

Some Useful Information:

Type of steady state solution found Trimmed collective blade pitch (TrimCase = 3)

Period of steady state solution (sec) 2.93212E+00

Iterations needed to find steady state solution 34

Displacement 2-norm of steady state solution (rad) 9.01316E-05

Velocity 2-norm of steady state solution (rad/s) 4.84390E-05

Number of equally-speced azimuth steps, NAzimStep 4

Order of linearized model, MdlOrder 1

Number of active (enabled) DOFs 4 ( 8 states)

Number of control inputs, NInputs 2

Number of input wind disturbances, NDisturbs 1

Number of output measurements 5

Order of States in Linearized State Matrices:

Row/column 1 = Variable speed generator DOF (internal DOF index = DOF_GeAz)

Row/column 2 = 1st flapwise bending-mode DOF of blade 1 (internal DOF index = DOF_BF(1,1))

Row/column 3 = 1st flapwise bending-mode DOF of blade 2 (internal DOF index = DOF_BF(2,1))

Row/column 4 = 1st flapwise bending-mode DOF of blade 3 (internal DOF index = DOF_BF(3,1))

Row/column 5 to 8 = First derivatives of row/column 1 to 4.

Order of Control Inputs in Linearized State Matrices:

Column 1 = electrical generator torque (N·m) 7.95744E+03 op

Column 2 = rotor collective blade pitch (rad) 3.41938E-01 op

Order of Input Wind Disturbances in Linearized State Matrices:

Column 1 = horizontal hub-height wind speed (m/s) 1.80000E+01 op

Order of Output Measurements in Linearized State Matrices:

Row 1 = GenTq (kN·m)

Row 2 = GenPwr (kW)

Row 3 = BldPitch1 (deg)

Row 4 = OoPDefl1 (m)

Row 5 = IPDefl1 (m)

Linearized State Matrices:

----------------------------- Azimuth = 0.00 deg -----------------------------

op State | op | A - State | B - Input | Bd - Dstrb

Derivativs | States | Matrix | Matrix | Matrix

2.143E+00 | 4.712E+00 | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 1.000E+00 0.000E+00 0.000E+00 0.000E+00 | 0.000E+00 0.000E+00 | 0.000E+00

2.222E-01 | 4.928E-01 | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 1.000E+00 0.000E+00 0.000E+00 | 0.000E+00 0.000E+00 | 0.000E+00

-3.294E-02 | 6.283E-01 | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 1.000E+00 0.000E+00 | 0.000E+00 0.000E+00 | 0.000E+00

-1.825E-01 | 4.442E-01 | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 1.000E+00 | 0.000E+00 0.000E+00 | 0.000E+00

8.973E-05 | 2.143E+00 | 3.515E-05 1.380E-01 1.373E-01 1.374E-01 1.087E-01 8.897E-03 9.439E-03 6.190E-03 | -3.027E-05 1.867E+00 | -7.305E-03

8.147E-03 | 2.222E-01 | 6.668E+00 -7.072E+01 -3.559E+00 -3.562E+00 -7.968E+01 -8.571E+00 -2.437E-01 -1.607E-01 | 7.851E-04 -8.277E+02 | 1.282E+01

-4.493E-01 | -3.294E-02 | -2.788E+00 -3.578E+00 -7.014E+01 -3.563E+00 -8.379E+01 -2.273E-01 -8.909E+00 -1.590E-01 | 7.847E-04 -8.522E+02 | 1.308E+01

3.781E-01 | -1.825E-01 | -4.118E+00 -3.579E+00 -3.560E+00 -7.020E+01 -6.411E+01 -2.279E-01 -2.445E-01 -6.834E+00 | 7.847E-04 -7.148E+02 | 1.124E+01

op Output | This colmn | C - Output | D - Trnsmt | Dd - DTsmt

Measurmnts | is blank | Matrix | Matrix | Matrix

7.957E+00 | | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 | 1.000E-03 0.000E+00 | 0.000E+00

1.425E+03 | | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 6.650E+02 0.000E+00 0.000E+00 0.000E+00 | 1.791E-01 0.000E+00 | 0.000E+00

1.959E+01 | | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 | 0.000E+00 5.730E+01 | 0.000E+00

4.459E-01 | | -2.134E-06 9.048E-01 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 | 0.000E+00 -2.092E-01 | 0.000E+00

-2.091E-01 | | -4.269E-07 -4.243E-01 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 | 0.000E+00 -4.459E-01 | 0.000E+00

[lines deleted]

----------------------------- Azimuth = 270.00 deg -----------------------------

op State | op | A - State | B - Input | Bd - Dstrb

Derivativs | States | Matrix | Matrix | Matrix

2.143E+00 | 3.142E+00 | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 1.000E+00 0.000E+00 0.000E+00 0.000E+00 | 0.000E+00 0.000E+00 | 0.000E+00

-6.944E-02 | 4.126E-01 | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 1.000E+00 0.000E+00 0.000E+00 | 0.000E+00 0.000E+00 | 0.000E+00

2.108E-01 | 5.462E-01 | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 1.000E+00 0.000E+00 | 0.000E+00 0.000E+00 | 0.000E+00

-1.409E-01 | 6.073E-01 | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 1.000E+00 | 0.000E+00 0.000E+00 | 0.000E+00

2.973E-05 | 2.143E+00 | 8.486E-04 1.374E-01 1.379E-01 1.373E-01 1.021E-01 6.150E-03 9.160E-03 6.822E-03 | -3.027E-05 1.795E+00 | -4.572E-03

5.245E-01 | -6.944E-02 | -3.978E-01 -7.029E+01 -3.577E+00 -3.560E+00 -6.344E+01 -6.790E+00 -2.369E-01 -1.769E-01 | 7.851E-04 -7.094E+02 | 1.113E+01

-1.025E-01 | 2.108E-01 | 5.913E+00 -3.563E+00 -7.069E+01 -3.559E+00 -8.454E+01 -1.609E-01 -8.656E+00 -1.756E-01 | 7.852E-04 -8.561E+02 | 1.272E+01

-4.959E-01 | -1.409E-01 | -6.108E+00 -3.565E+00 -3.577E+00 -7.011E+01 -7.481E+01 -1.555E-01 -2.372E-01 -7.050E+00 | 7.845E-04 -7.755E+02 | 1.121E+01

op Output | This colmn | C - Output | D - Trnsmt | Dd - DTsmt

Measurmnts | is blank | Matrix | Matrix | Matrix

7.957E+00 | | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 | 1.000E-03 0.000E+00 | 0.000E+00

1.425E+03 | | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 6.650E+02 0.000E+00 0.000E+00 0.000E+00 | 1.791E-01 0.000E+00 | 0.000E+00

1.959E+01 | | 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 | 0.000E+00 5.730E+01 | 0.000E+00

3.734E-01 | | -4.269E-07 9.048E-01 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 | 0.000E+00 -1.751E-01 | 0.000E+00

-1.751E-01 | | 1.921E-06 -4.243E-01 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 | 0.000E+00 -3.734E-01 | 0.000E+00

Figure 31. Sample linearization model file.

FAST User's Guide 121 Last updated August 2005 for version 6.0
 This summary information was generated by FAST (v6.00c-jmj, 15-Apr-2005) on 15-Apr-2005

at 16:37:53.

FAST certification Test #01: AWT-27CR2 with many DOFs with fixed yaw error and steady

wind.

Turbine features:

Downwind, two-bladed rotor with teetering hub.

Rigid foundation.

The model has 13 of 22 DOFs active (enabled) at start-up.

Enabled First flapwise blade mode DOF.

Enabled Second flapwise blade mode DOF.

Enabled Edgewise blade mode DOF.

Enabled Rotor-teeter DOF.

Enabled Drivetrain rotational-flexibility DOF.

Enabled Generator DOF.

Disabled Rotor-furl DOF.

Disabled Tail-furl DOF.

Disabled Yaw DOF.

Enabled First tower fore-aft bending-mode DOF.

Enabled Second tower fore-aft bending-mode DOF.

Enabled First tower side-to-side bending-mode DOF.

Enabled Second tower side-to-side bending-mode DOF.

Disabled Platform horizontal surge translation DOF.

Disabled Platform horizontal sway translation DOF.

Disabled Platform vertical heave translation DOF.

Disabled Platform roll tilt rotation DOF.

Disabled Platform pitch tilt rotation DOF.

Disabled Platform yaw rotation DOF.

Enabled Computation of aerodynamic loads.

Disabled Computation of aeroacoustics.

Time steps:

Structural (s) 0.00400000

Aerodynamic (s) 0.00400000

Some calculated parameters:

Hub-Height (m) 42.672

Flexible Tower Length (m) 41.980

Flexible Blade Length (m) 12.573

Rotor mass properties:

Rotor Mass (kg) 2200.608

Rotor Inertia (kg-m^2) 41755.996
Blade 1 Blade 2
------- -------
Mass (kg) 435.304 435.304
Second Mass Moment (kg-m^2) 15434.709 15434.709
First Mass Moment (kg-m) 2120.280 2120.280
Center of Mass (m) 4.871 4.871

Additional mass properties:

Tower-top Mass
(kg) 7216.038

Tower Mass
(kg) 36907.137

Turbine Mass
(kg) 44123.176

Mass Incl. Platform

(kg) 44123.176

Figure 32. Sample summary file.

FAST User's Guide 122 Last updated August 2005 for version 6.0

 Interpolated tower properties:

Node TwFract HNodes DHNodes TMassDen FAStiff SSStiff

(-) (-) (m) (m) (kg/m) (Nm^2) (Nm^2)

1
0.024 1.000 1.999 879.160 1.564E+10 1.564E+10

2
0.071 2.999 1.999 879.160 1.564E+10 1.564E+10

3
0.119 4.998 1.999 879.160 1.564E+10 1.564E+10

[lines deleted]

19 0.881 36.982 1.999 879.160 1.564E+10 1.564E+10

20 0.929 38.981 1.999 879.160 1.564E+10 1.564E+10

21 0.976 40.980 1.999 879.160 1.564E+10 1.564E+10

Interpolated blade 1 properties:

Node BlFract RNodes DRNodes AeroCent StrcTwst BMassDen FlpStff EdgStff

(-) (-) (m) (m) (-) (deg) (kg/m) (Nm^2) (Nm^2)
1 0.050 1.813 1.257 0.250 10.500 58.496 2.760E+07 8.618E+07
2 0.150 3.070 1.257 0.250 10.404 49.105 1.582E+07 9.772E+07
3 0.250 4.327 1.257 0.250 9.852 48.912 1.043E+07 1.065E+08
4 0.350 5.585 1.257 0.250 9.096 42.735 6.827E+06 8.982E+07
5 0.450 6.842 1.257 0.250 7.692 36.024 4.463E+06 6.640E+07
6 0.550 8.099 1.257 0.250 5.613 30.272 2.821E+06 4.508E+07
7 0.650 9.356 1.257 0.250 3.575 24.996 1.677E+06 2.751E+07
8 0.750 10.614 1.257 0.250 1.990 20.925 8.835E+05 1.819E+07
9 0.850 11.871 1.257 0.250 1.012 16.201 3.821E+05 9.658E+06
10 0.950 13.128 1.257 0.250 0.395 9.536 1.030E+05 2.780E+06

Interpolated blade 2 properties:

Node BlFract RNodes DRNodes

AeroCent StrcTwst BMassDen FlpStff EdgStff

(-) (-) (m) (m)

(-) (deg) (kg/m) (Nm^2) (Nm^2)

1 0.050 1.813 1.257

0.250 10.500 58.496 2.760E+07 8.618E+07

2 0.150 3.070 1.257

0.250 10.404 49.105 1.582E+07 9.772E+07

3 0.250 4.327 1.257

0.250 9.852 48.912 1.043E+07 1.065E+08

4 0.350 5.585 1.257

0.250 9.096 42.735 6.827E+06 8.982E+07

5 0.450 6.842 1.257

0.250 7.692 36.024 4.463E+06 6.640E+07

6 0.550 8.099 1.257

0.250 5.613 30.272 2.821E+06 4.508E+07

7 0.650 9.356 1.257

0.250 3.575 24.996 1.677E+06 2.751E+07

8 0.750 10.614 1.257

0.250 1.990 20.925 8.835E+05 1.819E+07

9 0.850 11.871 1.257

0.250 1.012 16.201 3.821E+05 9.658E+06

10 0.950 13.128 1.257

0.250 0.395 9.536 1.030E+05 2.780E+06

Figure 32. Sample summary file (concluded).

FAST User's Guide 123 Last updated August 2005 for version 6.0
 This file was generated by AeroDyn(12.56, 24-Sep-2003) in FAST (v4.31, 03-Oct-2003) on 03-Oct-2003 at 16:08:06.

Inputs read in from aerodyn.ipt:

AWT-27CR aerodynamic parameters for FAST certification test #1.

SI Units for input and output

BEDDOES Dynamic stall model [Beddoes]

NO_CM Aerodynamic pitching moment model [NO Pitching Moments calculated]

DYNIN Inflow model [Dynamic Inflow]

SWIRL Induction factor model [Normal and Radial flow induction factors calculated]

0.005 Convergence tolerance for induction factor

[Not Used] Tip-loss model

[Not Used] Hub-loss model

"Wind/AWT27/Shr12_30.wnd" is the Hub-height wind file

42.672 Wind reference (hub) height, m

0.3 Tower shadow centerline velocity deficit

1 Tower shadow half width, m

2.432 Tower shadow reference point, m

1.225 Air density, kg/m^3

1.4639e-5 Kinematic air viscosity, m^2/'sec

0.004 Time interval for aerodynamic calculations, sec

10 Number of airfoil files used. Files listed below:

"AeroData/AWT27/AWT27_05.dat"

"AeroData/AWT27/AWT27_15.dat"

"AeroData/AWT27/AWT27_25.dat"

"AeroData/AWT27/AWT27_35.dat"

"AeroData/AWT27/AWT27_45.dat"

"AeroData/AWT27/AWT27_55.dat"

"AeroData/AWT27/AWT27_65.dat"

"AeroData/AWT27/AWT27_75.dat"

"AeroData/AWT27/AWT27_85.dat"

"AeroData/AWT27/AWT27_95.dat"

10 Number of blade elements per blade

RELM(m) Twist(deg) DR(m) Chord(m) File ID Elem Data RELM and Twist ignored by ADAMS (but placeholders must be

present)

1.81265 5.8 1.2573 0.859 1

3.07 5.2 1.2573 1.045 2

4.32725 4.66 1.2573 1.145 3

5.58455 3.73 1.2573 1.124 4

6.84185 2.64 1.2573 1.054 5

8.1 1.59 1.2573 0.976 6

9.35645 0.73 1.2573 0.885 7

10.61375 0.23 1.2573 0.775 8

11.87105 0.08 1.2573 0.651 9

13.12835 0.03 1.2573 0.493 10

Hub-height wind file info:

Initial horizontal wind speed = 12 mps

Initial wind direction = 30 deg

Initial vertical wind speed = 0 mps

Initial horiz. wind shear coeff. = 0

Initial power law vert. wind shear coeff. = 0.2

Initial linear vert. wind shear coeff. = 0

Initial gust wind speed = 0 mps

BEDDOES DYNAMIC STALL PARAMETERS:

CN SLOPE 6.0090 6.0190 6.0280 6.0340 6.1280 6.2040 6.2710 6.2580 6.1560 6.1180
STALL CN (UPPER) 1.8090 1.8130 1.8170 1.8200 1.7430 1.6770 1.6180 1.7910 1.7490 1.8400
STALL CN (LOWER) -1.0000 -1.0000 -1.0000 -1.0000 -1.0000 -1.0000 -1.0000 -1.0000 -1.0000 -1.0000
ZERO LIFT AOA -4.2450 -4.2560 -4.2730 -4.2850 -3.2970 -2.4840 -1.7820 -1.4010 -1.2780 -1.2320
MIN DRAG AOA 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
MIN DRAG COEFF 0.0107 0.0101 0.0094 0.0090 0.0080 0.0071 0.0064 0.0062 0.0064 0.0065

VORTEX TRANSIT TIME FROM LE TO TE 11.00000

PRESSURE TIME CONSTANT 1.700000

VORTEX TIME CONSTANT 6.000000

F-PARAMETER TIME CONSTANT 3.000000

Figure 33. Sample AeroDyn options file.

FAST User's Guide 124 Last updated August 2005 for version 6.0
 REFERENCES

1. Laino, D.J.; Hansen, A.C. “User’s Guide to the Wind Turbine Dynamics Computer Software AeroDyn.” Salt Lake City,
Utah: Windward Engineering, LC, August 2001.
2. Bossanyi, E.A. “GH Bladed Version 3.6 User Manual.” Document 282/BR/010 Issue 12. Garrad Hassan and Partners
Limited, 2003.
3. Manjock, A. “Evaluation Report: Design Codes FAST and ADAMS® for Load Calculations of Onshore Wind Turbines.”
Report No. 72042. Humburg Germany: Germanischer Lloyd WindEnergie GmbH, May 26, 2005.
4. Wilson, R.E.; Walker, S.N.; Heh, P. “Technical and User’s Manual for the FAST_AD Advanced Dynamics Code.”
OSU/NREL Report 99-01. Corvallis, Oregon: Oregon State University, May 1999.
5. Buhl, Jr. M.L.; Wright; A.D.; Pierce, K.G. “FAST_AD Code Verification: A Comparison to ADAMS.” The 20th ASME
Wind Energy Symposium, Reno, Nevada, January 8–11, 2001. NREL/CP-500-28848. Golden, Colorado: National
Renewable Energy Laboratory, 2001.
6. Jonkman, J.M.; Buhl, Jr. M.L. “New Development’s for the NWTC’s FAST Aeroelastic HAWT Simulator.” The 23rd ASME
Wind Energy Symposium, Reno, Nevada, January 5–8, 2004. NREL/CP-500-35077. Golden, Colorado: National Renewable
Energy Laboratory, 2004.
7. Buhl, Jr. M.L. “Installing NWTC Design Codes on PCs Running Windows NT®.” National Renewable Energy Laboratory,
http://wind.nrel.gov/designcodes/papers/setup.pdf. Last modified Dec. 22, 2000; accessed April 5, 2002. NREL/EP-500-
29384. Golden, Colorado.
8. IEC/TS 61400-13 ed. 1 “Wind Turbine Generator Systems – Part 13: Measurement of Mechanical Loads.” International
Electrotechnical Commission (IEC), 2001.
9. Buhl Jr., M.L. “A Simple Mode-Shape Generator for Both Towers and Rotating Blades.” NWTC Design Codes (Modes),
http://wind.nrel.gov/designcodes/preprocessors/modes/. Last modified April 29, 2002; accessed July 9, 2002.
10. Malcolm, D.J. “How a Teetered Rotor with Delta-3 Really Works.” The 2000 ASME Wind Energy Symposium, January 10-
13, 2000, Reno, Nevada. New York, New York: American Institute of Aeronautics and Astronautics, 2000.
11. Stol, K.A.; Bir, G.S. “SymDyn User’s Guide,” NREL/EL-500-33845. Golden, Colorado: National Renewable Energy
Laboratory, 2003.
12. Elliot, A.S.; Wright, A.D. “ADAMS/WT User’s Guide,”
http://wind.nrel.gov/designcodes/simulators/adamswt/docs_v2.0/index.html. Last modified December, 1998; accessed June
13, 2003.
13. Laino, D.J.; Hansen, A.C. “User’s Guide to the Computer Software Routines AeroDyn Interface for ADAMS®.” Salt Lake
City, Utah: Windward Engineering, LC, September 2001.
14. Laino,
D., “IECWind: A Program to Create IEC Wind Data Files.” NWTC Design Codes (IECWind),
http://wind.nrel.gov/designcodes/preprocessors/iecwind/. Last modified June 19, 2001; accessed July 9, 2002.
15. L
aino, D., “WindMaker: A Program to Create IEC Wind Data Files.” NWTC Design Codes (WindMaker),
http://wind.nrel.gov/designcodes/preprocessors/windmaker/. Last modified May 12, 2000; accessed July 9, 2002.
16. Jonkman, B.J.; Buhl Jr., M.L. “TurbSim User’s Guide,” http://wind.nrel.gov/designcodes/preprocessors/turbsim/turbsim.pdf.
Last modified May 26, 2005; accessed August 11, 2005.
17. Buhl Jr., M.L. “SNwind User’s Guide,” http://wind.nrel.gov/designcodes /preprocessors/snwind/snwind.pdf. Last modified
October 22, 2001; accessed July 9, 2002.
18. Kelley, N.D., “SNLWIND-3D: A Stochastic, Full-Field, Turbulent-Wind Simulator for Use with BLADED and the
Aerodyn-Based Design Codes (YawDyn, FAST_AD, and ADAMS®).” NWTC Design Codes (SNLWIND-3D),
http://wind.nrel.gov/designcodes/preprocessors/snlwind3d/. Last modified July 20, 2000; accessed July 9, 2002.
19. Moriarty, P.J.; Migliore, P.G. “Semi-Emperical Aeroacoustic Noise Prediction Code for Wind Turbines”, NREL/TP-500-
34478. Golden, Colorado: National Renewable Energy Laboratory, 2003.

FAST User's Guide 125 Last updated August 2005 for version 6.0
 Form Approved
REPORT DOCUMENTATION PAGE OMB No. 0704-0188
The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,
gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this
collection of information, including suggestions for reducing the burden, to Department of Defense, Executive Services and Communications Directorate (0704-0188). Respondents
should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a
currently valid OMB control number.
PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ORGANIZATION.
1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To)
October 2005 Technical Report
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
FAST User's Guide—Updated August 2005 DE-AC36-99-GO10337
5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

J.M. Jonkman and M.L. Buhl, Jr. NREL/TP-500-38230
5e. TASK NUMBER
WER5.3115
5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION

National Renewable Energy Laboratory REPORT NUMBER
1617 Cole Blvd. NREL/TP-500-38230
Golden, CO 80401-3393

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR'S ACRONYM(S)

NREL

11. SPONSORING/MONITORING
AGENCY REPORT NUMBER

12. DISTRIBUTION AVAILABILITY STATEMENT

National Technical Information Service
U.S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
13. SUPPLEMENTARY NOTES

14. ABSTRACT (Maximum 200 Words)

The FAST (Fatigue, Aerodynamics, Structures, and Turbulence) Code is a comprehensive aeroelastic simulator
capable of predicting both the extreme and fatigue loads of two- and three-bladed horizontal-axis wind turbines
(HAWTs). This document covers the features of FAST and outlines its operating procedures.

15. SUBJECT TERMS

FAST; Fatigue, Aerodynamics, Structures, and Turbulence; aeroelastic simulator; horizontal-axis wind turbine

16. SECURITY CLASSIFICATION OF: 17. LIMITATION 18. NUMBER 19a. NAME OF RESPONSIBLE PERSON
OF ABSTRACT OF PAGES
a. REPORT b. ABSTRACT c. THIS PAGE
Unclassified Unclassified Unclassified UL
19b. TELEPHONE NUMBER (Include area code)

Standard Form 298 (Rev. 8/98)

Prescribed by ANSI Std. Z39.18

F1147-E(12/2004)