Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

EDD Calibration Trim

Download as pdf or txt
Download as pdf or txt
You are on page 1of 23

wp_EDDL calibrationtrim rH

2015-09-03 00:26:00

Intelligent Device Management Tutorial: Calibration


In the early days of smart transmitters the concept of range setting from a central location (“remote
calibration”) and re-ranging without applying input was revolutionary. It took years of education to
be accepted and understood. Calibration can be carried out using a handheld communicator in the
field, a laptop on the bench in the workshop, or from intelligent device management (IDM)
software as part of an asset management system. Electronic Device Description Language (EDDL)
is the technology used by device manufacturers to define how the system shall display the device
information and functions to the technician. EDDL makes calibration of smart transmitters and
other intelligent devices easier thanks to user guidance such as wizards and help, and unparalleled
consistency of use.

This tutorial explains the common principles of calibration, re-ranging, and trim as they apply to
various kinds of transmitters. The detail procedure varies slightly depending on the measurement
done, sensing principle, and each manufacturer. Refer to the device instruction manual. But once
the common principles are understood, the product manual becomes easier to understand.

1 Calibration
By definition, the term “calibrate” means several different things:
1) Set the range (scale)
2) Trim (correct) the sensor (transducer) reading or current output against a standard
3) Simply compare the sensor (transducer) reading or current output against a standard to see how
large the error is without correcting (trim) it. This is often done in five points, increasing and
decreasing. If the error is too large, the transmitter may be trimmed or replaced.

ANSI/ISA–51.1 Definition of Terms


Calibrate: To ascertain outputs of a device corresponding to a series of values of the quantity
which the device is to measure, receive, or transmit. Data so obtained are used to:
1) determine the locations at which scale graduations are to be placed;
2) adjust the output, to bring it to the desired value, within a specified tolerance;
3) ascertain the error by comparing the device output reading against a standard.

2 Calibrating Intelligent Transmitters


The term "calibration" in the context of smart/intelligent transmitters is often misunderstood. In the
days of analog transmitters calibration meant applying a physical input and turning the trim
potentiometers to adjust the transmitter so that the analog output current becomes correct according
to the desired measurement range. Once smart transmitters appeared, this "calibration" process was
divided into three parts:
• Sensor trim
• Range setting (re-ranging)
• Current trim

The reason for separating these functions is that the range can be changed without applying a
physical input. This was a huge time and cost saver and one of the major reasons for the rapid
adoption of smart transmitters. However, do not confuse "sensor trim" with "range setting". Both
are part of calibration, but two very different things. In the view of many, range setting is more like
configuration than calibration.

www.eddl.org 1
2.1 Sensor Trim (Digital Trim)
Over time all sensors drift. Depending on the type of sensor it may be due to extreme pressure or
temperature, vibration, material fatigue, contamination, or other factors. Sensor reading may also be
offset due to mounting position.

Sensor trim is used to correct the digital reading as seen in the device local indicator LCD and
received over the digital communication. For instance, if pressure is 0 bar but transmitter reading
shows 0.03 bar, then sensor trim is used adjust it back to 0 bar.

Sensor trim can also be used to optimize performance over a smaller range than was originally
trimmed in the factory.

The basic principle for calibration (sensor trim) of all transmitters is the same:
1. Apply a known input
2. Tell the transmitter what it is
3. The transmitter calculates internal correction factors
4. The transmitter uses the new correction factors to compute a new correct measurement reading

Sensor trim requires the technician to apply a physical input to the transmitter. Therefore the
technician must either do sensor trim in the field at the process location, or the transmitter has to be
brought back into the workshop to perform sensor trim. This applies to 4-20 mA/HART,
WirelessHART, FOUNDATION fieldbus, as well as PROFIBUS transmitters. Sensor trim in the
field is easiest done using a handheld communicator connected to the running bus which is
supported by 4-20 mA/HART, WirelessHART, and FOUNDATION fieldbus. For PROFIBUS-PA
the trim command can either be sent from the control system, or the transmitter can temporarily be
disconnected from the running bus to perform the sensor trim.

Typically there are three forms of sensor trim:


• Zero sensor trim
• Lower sensor trim
• Upper sensor trim

Zero trim requires the physical input applied to be zero, this is often used with pressure transmitters
For best accuracy perform sensor trim in two points, close to lower range value and upper range
value. This is where lower and upper sensor trim is used. A known physical input is applied to the
transmitter to perform the sensor trim, the technician keys in the applied value (on a computer or
handheld communicator) communicated to the transmitter, allowing the transmitter to correct itself.
The physical input values applied for lower and upper sensor trim respectively are stored in the
transmitter memory and are referred to as Lower Sensor Trim Point and Upper Sensor Trim Point
respectively. EDDL plays an important role in sensor trim because a “wizard” created by the device
manufacturer guides the user step by step through the process.

Sensor trim requires a very accurate input to be applied. The factory calibration equipment is
usually more accurate than the portable calibrators at site. Moreover, transmitters these days are
typically very stable. Therefore, sensor trim of brand new transmitters is rarely done at
commissioning.

Note that sensor trim is done in firmware in the transmitter microprocessor; it is not done in the
sensor itself. The trim is really a mathematical function, adjusting numerical bias and gain factors.
That is, it is the sensor reading after the A/D conversion which is trimmed, not the sensor hardware

Sensor trim is the aspect of calibration which this white paper focuses on. That is:
• Pressure calibration

www.eddl.org 2
• Flow calibration
• Temperature calibration
• Level calibration
• Etc.

2.1.1 Sensor Trim Points


The purpose of the (CAL_POINT) parameters is to tell at which points sensor trim was last done,
and to perform sensor trim points sensor trim if needed. If the sensor trim points parameters are 0
and 360 mbar this means these are the points at which it was calibrated (sensor trim). The
transmitter may still be able to measure -600 to +600, but remember the transmitter is now
extrapolating so it may not be full accuracy, but it may be OK anyway. This is not uncommon. If
sensor trim is performed at -600 and +600 greater accuracy may be achieved.

Note that the sensor trim points are not just “set”; they are NOT range configuration parameters.
These two parameters are written when sensor trim is performed. The transmitter then remembers
these points were the trim was made. Typically there is a sensor trim wizard (“method”) that takes
the technician step by step through the calibration process and it is this sensor trim wizard that
writes the sensor trim point parameters.

2.2 Range Setting (Re-range)


Range setting (re-ranging) refers to setting the scale for the 4 mA and 20 mA points. This scale is
usually referred to as "calibrated range" or "calibration range". That is, at what input shall the
transmitter analog output be 4 mA; Lower Range Value (LRV) often referred to as "zero" meaning
0%, and at what input shall it be 20 mA; Upper Range Value (URV), sometimes called "full scale"
meaning 100%. Note that the term "span" is not the same as URV. Span is the magnitude of
difference between URV and LRV. For instance, if LRV is 20 and URV is 100, the span is 80.
Since Fieldbus, PROFIBUS, and WirelessHART do not use 4-20 mA, range setting is not required
for such devices in most applications.

Note that calculating what the output current value should be is done in firmware in the transmitter
microprocessor. It is a mathematical function.

Internally the 4-20 mA/HART transmitter computes:

Percentage = (PRIMARY_VARIABLE - LRV) / (URV - LRV) * 100 [%]

Analog Current = (PRIMARY_VARIABLE - LRV) / (URV - LRV) * 16 + 4 [mA]

www.eddl.org 3
DCS Range

100 kPa

80 kPa
Operator Console PV = 70 kPa
60 kPa

40 kPa
Analog Output
Current 20 kPa Analog Input Current
4 mA 8 mA 12 mA 16 mA 20 mA
100% / 20 mA
0% 25% 50% 75% 100%

75% / 16 mA
Current = 14 mA

50% / 12 mA

25% / 8 mA

0% / 4 mA Transmitter Range
20 40 60 80 100
kPa kPa kPa kPa kPa
Sensor
PRIMARY_VARIABLE
= 70 kPa
Figure 1 Range/scaling of transmitter and indication

Internally the 4-20 mA control system, recorder, or indicator computes:

Percentage = (Current - 4) / 16 * 100 [%]

PV = (Current - 4) / 16 * (URV - LRV) + LRV [E.U.]

The analog 4-20 mA output of a transmitter is limited to the LRV to URV range. Thus the analog
output does not benefit from the full LSL to USL capability of the sensor.

Actual Process
Saturates at 20 mA
250.00

200.00
Saturates at 4 mA
150.00

100.00

50.00 4 mA = 0 inH2O
20 mA = 150 inH2O
0.00
0 50 100 150 200 250 300 350 400
‐50.00

Figure 2 In an analog signal system the measurement is confined within range values

However, FOUNDATION fieldbus, PROFIBUS, and WirelessHART transmitters as well as the


digital output of 4-20 mA/HART transmitters are not limited to the LRV to URV range, but can
benefit from full LSL to USL capability of the sensor.

www.eddl.org 4
Does not saturate at range
250.00

200.00

150.00
Utilizes full sensor limits
100.00

50.00 LSL = -250 inH2O


USL = 250 inH2O
0.00
0 50 100 150 200 250 300 350 400
‐50.00

Figure 3 In a digital bus signal system the measurement enjoys the full sensor limits

History Lesson: Rangeability and Turn-down Ratio

Rangeability is the ratio of the smallest permitted span to the Upper Sensor Limit. For instance, if
the Upper Sensor Limit is 80 kPa and the minimum span is 2 kPa, then the turndown ratio
(rangeability) of that transmitter is 40:1.

Transmitter range setting is done without applying input, and therefore can be done remotely from a
central location. For instance, set range of pressure transmitter to get 4 mA when input is 0 bar and
20 mA when pressure is 40 bar. Range must be set within the Lower Sensor Limit (LSL) and Upper
Sensor Limit (USL). Transmitters usually also have a minimum span to be observed. The difference
between URV and LRV must exceed the minimum span, or else the analog output resolution and
percentage accuracy is too poor since with a small span the quantization error from the sensor A/D
converter gets amplified too much. The transmitter will reject range setting and sensor trim attempts
not observing the LSL, USL, and minimum span limits.

Error When Attempting to Reverse the Range

In some applications, particularly DP level, it is necessary to reverse the range such that the LRV >
URV. For instance, the range may have been set -87 to -1 mbar initially, but the application may
actually require the range to be -1 to -87 mbar. That is the range has to be reversed:

LRV = -87 mbar Change to: LRV = -1 mbar


URV = -1 mbar URV = -87 mbar

To make this range change the technician would typically make the mistake of directly keying in -1
mbar for LRV which would be rejected since both URV and LRV would then be -1 mbar leaving a
span of 0 mbar which would violate the minimum span. Or, the technician would attempt to directly
key in -87 mbar for URV which would also be rejected since both URV and LRV would then be -
87 mbar also violating the minimum span. The solution is to first key in an intermediate value. For
instance, setting the LRV to -40, then setting URV to -87, and lastly setting LRV to the final -1
mbar.

The same procedure applies if the sensor trim points are to be reversed

The sensor limits depend mostly on physical restrictions of the sensor. The sensor limits cannot be
changed, therefore they are always read-only. Different sensors have different sensor limits. For
example, various RTDs and thermocouples have different sensor limits. In temperature applications,
a sensor type with sufficient sensor limits has to be selected to accommodate the range of the
application. Since the range limits are physical and cannot be changed, to get wider sensor limits to
www.eddl.org 5
accommodate a wider range, it is necessary to purchase a new sensor. Similarly, pressure
transmitters have a selection of sensor modules with different range limits from the lowest “draft”
up to very high pressures. To get wider sensor limits to accommodate a wider range, it is necessary
to purchase a new sensor.

ANSI/ISA–51.1 Definition of Terms


zero elevation: For an elevated-zero range, the amount the measured variable zero is above the
lower range-value.

zero suppression: For a suppressed-zero range, the amount the measured variable zero is
below the lower range-value.

History Lesson: Elevation and Suppression

When the lower range value (4 mA point) is above zero, for instance a range of 20 to 100, is called
zero suppression. When the lower range value (4 mA point) is below zero, for instance range of -25
to +100, or -100 to 0, or -100 to -20, this is called zero elevation. That is, elevation and suppression
relates to the lower range value (the 4 mA of the transmitter range).

Elevation and suppression are chiefly used in differential pressure measurement since this make it
possible to measure liquid levels with wet legs and remote seals etc. In smart transmitters it is all
done in firmware and nobody ever asks about it anymore. The ability of a transmitter to do
elevation and suppression was stated in % and relates to rangeability (turndown). A popular analog
pressure transmitter had a 6:1 rangeability and therefore used to sport 600% elevation and 500%
suppression. Modern smart pressure transmitters have capabilities in order of several thousand
percent so it is no longer discussed.

Range setting is only applicable to transmitter with 4-20 mA analog output. That is, for 4-20
mA/HART transmitters, not for pure digital solutions like FOUNDATION fieldbus (FF) or
WirelessHART transmitters. The reason being that FF and WirelessHART transmitters has no 4-20
mA analog output, therefore there is no need to set 4 mA and 20 mA range points. For 4-20 mA
systems the range is set in both the transmitter and controller. For FF and PROFIBUS the range is
set in the controller, and need not be set in the transmitter which can lead to some confusion for
beginners. The only exception for FF, WirelessHART, and PROFIBUS transmitters may be for
differential pressure (DP) flow and level measurement where the end-points of the DP scale (e.g. 0-
250 inH2O in XD_SCALE) and corresponding flow or level scale (e.g. 0-400 bbl/day in
OUT_SCALE). This also enables DP transmitters to locally indicate in flow or level units. FF and
PROFIBUS devices have provision for setting a range in the transmitter even though it may not be
used for the application.

4-20 mA
Transmitter A DCS EU
m
0
2
-
Sensor EU % Current 4 Current % Range %
Sensor Range PID
Trim Trim Trim (AI)

Bus
Transmitter Valve / Controller
AI s
EU
u
Sensor B EU Range &
Sensor
Trim PID

Figure 4 An analog signal system requires range, current trims, and scaling. A digital bus
system does not

However, the nominal operating range has to be specified also for FOUNDATION fieldbus and
WirelessHART transmitters for sizing purposes when purchasing, such that the device supplier can
www.eddl.org 6
pick the appropriate sensor model. There is also a need to select the desired engineering unit in the
device. The DCS may need a range set in database as scaling end-points for bargraphs and trend and
will also need a range for PID control even though there is no range in the FOUNDATION fieldbus or
WirelessHART device. In control applications, level is usually expressed in percentage of full tank.

The output of both the FF transducer block and the AI function block is engineering unit. For most
applications there is no need to set range in either block in order to get the PV. However, many
systems use the range in the FF transmitter AI block to scale the faceplates bargraphs. A narrower
range may optionally be set to increase the resolution of the faceplate bargraph. If a range is set in
the AI block, the percentage of range can be seen from the FIELD_VAL parameter.

Transmitter
Transd
E.U. E.U.
Sensor ucer AI
Block

Figure 5 Digital transmitters internally operate in engineering units

4-20 mA Range Mismatch


If the range set in the transmitter does not match the range in the DCS, alarms, controls, and
indication will not function correctly. Therefore it is important to document any range changes such
that the correct range is set when transmitters are replaced.

DCS Range

100 kPa

80 kPa
Operator Console PV = 76 kPa

60 kPa

40 kPa
Analog Output
Current 20 kPa Analog Input Current
4 mA 8 mA 12 mA 16 mA 20 mA
100% / 20 mA
0% 25% 50% 75% 100%

75% / 16 mA Current = 15.2 mA

50% / 12 mA

25% / 8 mA

0% / 4 mA Transmitter Range
0 100
kPa kPa
Sensor
PRIMARY_VARIABLE
= 70 kPa
Figure 6 In this example a mismatch in LRV results in erroneous indication

Range in Instrument Specification Forms


Range is not required to be set in transmitters using pure digital communication, but it is required to
be specified in instrument specification forms (ISA20) such as with bid documents for sensor
element selection. For instance, the pressure transmitter supplier needs to know the pressure range
(and other process data) to select the sensor module for the application. The temperature transmitter
vendor needs to know the temperature range (and other process data) to select the temperature
element model for the application.

www.eddl.org 7
There are typically two ways to set the range of the transmitter:
• Direct numeric value entry
• To applied input

2.2.1 Direct numeric value entry


Direct numeric value entry means the desired lower and upper range values are simply keyed in
from device software or handheld field communicator, and sent to the transmitter, for instance,
keying in 20 to 100 kPa.

2.2.2 To applied input


Range setting to applied input requires a physical input corresponding to the desired range value to
be applied to the transmitter. This is sometimes used in level measurement applications. Because
the mounting (datum) of the level transmitter plays a part in the range, the range shall be adjusted at
site, it cannot be done in a lab. Basically it is a zero cancelation such as DP wet leg. For instance,
first the tank is emptied to its lower level and then the “set PV LRV command” is sent to the
transmitter to set the lower range value to whatever the input happens to be. For instance for a DP
level transmitter, if the pressure is 20 kPa when the tank is empty (the pressure tap is slightly below
the datum), this becomes the new lower range value, thus ensuring the reading is 0% and analog
output current is 4 mA whenever the tank is empty. Conversely, next the tank is filled to its upper
level and then the “set PV URV” command is sent to the transmitter to set the upper range value to
whatever the input happens to be. For instance, if the pressure is 100 kPa when the tank is full, this
becomes the new upper range value, thus ensuring the reading is 100% and analog output current is
20 mA whenever the tank is full. In between the reading is linear. Note that the technician need not
know what the physical input is, just that the tanks is full and empty respectively. EDDL plays an
important role in range setting to applied input because a “wizard” created by the device
manufacturer guides the user step by step through the process.

The set PV LRV command is also common to cancel wet-leg for DP transmitters in all kinds of
application including flow. The set PV LRV and set PV URV commands are equivalent to pushing
the 'zero' and 'span' buttons respectively found on some transmitters.

ANSI/ISA–51.1 Definition of Terms


range: The region between the limits within which a quantity is measured, received, or transmitted,
expressed by stating the lower and upper range-values. For example:
a) 0 to 150°F
b) –20 to +200°F
c) 20 to 150°C
range-value, lower: The lowest value of the measured variable that a device is adjusted to measure.
range-value, upper: The highest value of the measured variable that a device is adjusted to
measure.
range-limit, lower: The lowest value of the measured variable that a device can be adjusted to
measure.
range-limit, upper: The highest value of the measured variable that a device can be adjusted to
measure.
span: The algebraic difference between the upper and lower range-values. For example:
Range 0 to 150°F, Span 150°F
Range –20 to 200°F, Span 220°F
Range 20 to 150°C, Span 130°C
FOUNDATION fieldbus uses the term “scale” for “range”

www.eddl.org 8
History Lesson: Non-interactive zero and span
One of the major time saving benefits that came with microprocessor-based transmitters was non-
interactive zero and span:

When lower range value is set to applied input, the span (i.e. the difference between upper range
value and lower range value) is maintained. That is, the upper range value is shifted by the same
amount as the lower range value. For instance, if the original range was 0-100 kPa, and lower range
value is set to applied input when the applied input is 20 kPa, the new range will be 20-120 kPa,
maintaining the original span of 100 kPa. That is, the upper range value was also shifted by 20 kPa
just like the lower range value. However, if upper range value is set to applied input, the lower
range value zero) is not changed. For instance, if the original range was 0-100 kPa, and upper range
value is set to applied input when the applied pressure is 80 kPa, the new range will be 0-80 kPa,
maintaining the original zero of 0 kPa. That is, the lower range value was not shifted.

That is, changing the zero does not affect the span, changing the span does not affect the zero. This
is non-interactive zero and span.

Corrupting the range to make 4-20 mA right


Don't use range setting to correct a sensor error. For example, a 4-20 mA/HART pressure
transmitter ranged 0-100 inH2O may due to mounting position get a shift in reading of 1 inH2O.
The analog output will be 4.16 mA when no pressure is applied, making the field indicator or HMI
show 1 inH2O.

A common mistake is to isolate, equalize, and vent the manifold and click the 'Set PV LRV'
command in the device management software or handheld field communicator, or pushing the 'zero'
and button on the device thus changing the range bringing the analog output to 4 mA.

However, checking the PV in the device management software or local transmitter indicator will
reveal the sensor reading with no pressure applied is still 1 inH2O and the range is changed to 1-101
in H2O. Two wrongs appear to make a right

The correct way to correct for shift or drift is to do a sensor trim

Corrupting the sensor reading to make 4-20 mA right


Don't use sensor trim to cancel a wet-leg. For example, a 4-20 mA/HART pressure transmitter
ranged 0-100 inH2O may when mounted below the tank read 1 inH2O. The analog output will be
4.16 mA when the level is zero (at the tank datum point), making the field indicator or HMI show 1
inH2O.

A common mistake is to click the 'Zero Trim' command in the device management software or
handheld field communicator thus changing the sensor reading bringing the analog output to 4 mA.

However, if the 4-20 mA/HART pressure transmitter is isolated and vented, and the PV checked in
the device management software or local transmitter indicator, it will show the sensor reading with
no pressure applied is -1 inH2O. Two wrongs appear to make a right

The correct way to cancel a wet-leg is to set the range

2.3 Transmitter Output Current Trim (Analog Trim)


It is rare for the analog output current circuitry of a 4-20 mA transmitter to drift. However, should
the analog output current be incorrect, use current trim to correct the analog output signal. For
instance, if the analog output current is 4.13 mA when it should be 4.00 mA, then current trim is
used to adjust it to 4 mA.
www.eddl.org 9
Current trim is used to match the transmitter analog output current to the current input of the analog
input (AI) card channel on the DCS. For instance, the transmitter may be reading 0.00% but the
DCS may show 0.13% because of differences in current calibration. The DCS may not support
current trim of channels in the AI and AO cards. If there is drift in the DCS input circuitry A/D
conversion or D/A conversion and output circuitry, current trim has to be done in each device
instead.

Current trim is only applicable to transmitter with 4-20 mA analog output. That is, for 4-20
mA/HART transmitters, not for FOUNDATION fieldbus (FF), PROFIBUS-PA, or WirelessHART
transmitters, the reason being pure digital transmitters have no 4-20 mA analog output.

Current trim requires the technician to measure the physical output current from the transmitter.
Therefore the technician must either do current trim in the field at the process location by
connecting a multimeter to the transmitter test terminals, or the transmitter has to be brought back
into the workshop to perform current trim. Current trim in the field is possible using a handheld
communicator. EDDL plays an important role in current trim because a “wizard” created by the
device manufacturer guides the user step by step through the process.

2.4 Trim Quick Reference


The difference between sensor trim, range setting, and current trim are summarized in the table
below

Task Local / Example 4-20 mA FOUNDATION PROFIBUS


Central HART Fieldbus
Sensor Local Correct the sensor Yes Yes Yes
trim reading to applied input.
For instance, if pressure
is 0 bar but transmitter
reading is 0.03 bar, then
sensor trim is used
adjust it to 0 bar.
Range Local or Set the 4 mA and 20 Yes No* No*
setting central mA points. For
instance, set range of
pressure transmitter to
get 4 mA when input is
0 bar and 20 mA when
pressure is 40 bar.
Current Local Correct the analog Yes No No
trim output current. For
instance, if the analog
output current is 4.13
mA when it should be 4
mA, then current trim is
used to adjust it to 4
mA.
* Range setting in transmitter with digital output only done for DP-flow and DP-level measurement

2.5 Range Values and Limit Summary


The relationship between range values and limits are summarized in the table below

Description

www.eddl.org 10
Description
LSL Lower Sensor Limit Lowest possible value for the 4 mA point
LRV Lower Range Value The 4 mA point
URV Upper Range Value The 20 mA point
USL Upper Sensor Limit Highest possible value for the 20 mA point
Span URV minus LRV
Zero Same as LRV
Turndown Span divided by URV

Analog Output
Current

Span

100% / 20mA

0% / 4 mA

Transmitter Input
LSL LRV URV USL
Zero
Figure 7 Range values and limits

3 Valve Positioner Setpoint Current Trim


Similarly, it is rare for the setpoint input current circuitry of a 4-20 mA positioner to drift. However,
should the input current sensing be incorrect, use current trim to correct the input signal. For
instance, if the current input reads 4.13 mA when it should read 4.00 mA, then current trim is used
to adjust it so that the setpoint reads correctly.

Current trim is used to match the positioner current input to the analog output current of the analog
output (AO) card channel on the DCS. For instance, the DCS PID output may be 0.00% but the
positioner setpoint may show 0.13% because of differences in current calibration.

Current trim is only applicable to positioners with 4-20 mA input. That is, for 4-20 mA/HART
positioners, not for FOUNDATION fieldbus (FF) positioners, the reason being pure digital
positioners have no 4-20 mA input.

Current trim requires the technician to connect a precision current source or to measure the physical
input to the positioner. Therefore the technician must either do current trim in the field at the valve
by connecting a multimeter to the positioner test terminals, or the valve has to be brought back into
the workshop to perform current trim. Current trim in the field is possible using a handheld
communicator.

www.eddl.org 11
Figure 8 Positioner manufacturer uses EDDL to illustrate connection for current trim
procedure

4 Valve Positioner Travel Stroking (Position Feedback Sensor Trim)


Stroking a valve positioner to find its fully opened and fully closed positions is in fact an automated
procedure to among other things trim (calibrate) the position transmitter feedback sensor. That is, it
is just like sensor trim for a pressure or temperature transmitter, only that a known reference need
not be connected, the positioner will automatically stoke the valve over its full travel to discover the
open and closed end-positions.

Figure 9 Automatic valve positioner travel calibration wizard rendered by the system from
EDDL

Likewise, the analog 4-20 mA actual valve position feedback current output of a 4-20 mA
positioner is calibrated just like a 4-20 mA transmitter. EDDL plays an important role in position
feedback sensor trim because a “wizard” created by the device manufacturer guides the user step by
step through the process.
www.eddl.org 12
Again, this process is not required for FOUNDATION fieldbus positioners or for position feedback
transmitters based on WirelessHART.

5 Sensor Trim Procedure


Plants have a great mix of transmitters for different kinds of measurements from different
manufacturers. Since all sensors drift, at some point in time all sensors need a trim. The procedure
for calibration depends on the type of transmitter:
• Pressure transmitter: apply pressure from calibrator or dead weight tester or the manifold
can be equalized for zero trim
• Temperature transmitter: apply milli-voltage or resistance from calibrator or resistance
decade box
• Flowmeter: has to be calibrated against a prover or master meter
• Valve position transmitter: stroke the valve fully opened and fully closed
• pH transmitter: put the pH sensor in buffer solutions

The procedure for sensor trim may also vary slightly from one manufacturer to the next depending
on the requirement for the particular sensor technology. EDDL plays an important role in sensor
trim because a “wizard” created by the device manufacturer guides the user step by step through the
process.

Some calibration is easier in the workshop, such as pH sensor buffering where the pH sensor has to
be put into with buffer solutions and distilled water. This is easier in the lab. Smart pH sensors have
a memory chip inside making it possible to calibrate the sensor in the lab and bring it to the field,
carrying the calibration offset and slope data inside its memory chip. Once connected, the pH
transmitter/analyzers upload the calibration data from the sensor memory.

6 Poor Hosts
The original DD technology from 1992 made it possible to calibrate all transmitters using the same
handheld field communicator or laptop software. Before DD only proprietary solutions existed.

The most basic content of a DD file is the Device Definition describing the blocks and the
parameters in the device, including limits, options, and help etc. The original DD technology from
1992 also included Business Logic such as "wizards" (aka "methods") which is a kind of script
created by the transmitter manufacturer to guide the technician through the sensor trim process.
Wizards thus make sensor trim easy. Wizards, conditionals, and help are explained further in the
section on EDDL. However, not all transmitters provided wizards in their DD file and not all
intelligent device management software supported wizards. That is, on many systems and for many
types of transmitters, sensor trim in the past was not so easy. For instance, FOUNDATION fieldbus
transmitters which in the past had to set the correct mode, in the correct block, write value to correct
parameter, and remember to return block mode. At the same time the technician also had to
remember to inform operations. All of this in addition to actual sensor trim task itself such as
applying input, or working the manifold etc. Lack of wizards was not a problem with the DD
technology itself, it was poor implementation in many early products.

www.eddl.org 13
EDDL

Device Business User


Definition Logic Interface
Description

•Blocks •Wizards •Menus


•Parameters •Conditionals •Graphics

Figure 10 Not all systems supported EDDL “Business Logic”, and the graphical User
Interface Description was not introduced until 2006.

The EDDL enhancement done in 2006 is a User Interface Description which includes graphics such
as trend charts, waveform graphs, dial gages, bargraphs, bar charts, and tables etc. Graphics, menu
system, wizards, and conditionals are now mandatory for all FOUNDATION fieldbus and 4-20
mA/HART compliant host systems. Therefore, make sure to use a control system which supports
enhanced EDDL.

7 Sensor Trim Made Easy


Sensor trim is usually done in the field since there is normally a need to access the transmitter to
perform sensor trim. For instance, for a pressure transmitter it is necessary to isolate, equalize, and
vent the manifold. For a temperature transmitter there is a need to connect a calibrator. In the field,
sensor trim is carried out using a small portable field communicator. In a workshop it may be done
using configuration software on a laptop computer or intelligent device management software part
of asset management system. EDDL (IEC 61804-3) is the only device integration that supports all,
not just laptop (see separate technical white paper on field work). Thanks to EDDL wizards, sensor
trim for FOUNDATION fieldbus is now just as easy as 4-20 mA/HART.

7.1 Interactive Wizards


Wizards, also known as EDDL methods, are used by transmitter manufacturers to guide tasks such
as calibration. Calibration capability may include wizards for:
• Sensor zero trim
• Sensor lower trim
• Sensor upper trim
• Set lower range value to applied input
• Set upper range value to applied input
• Current 4 mA trim
• Current 20 mA trim

Using wizards the transmitter manufacturer's expert embeds their calibration know-how in the
EDDL file taking the technician through the task step-by-step. This ensures the correct sequence is
followed, thereby simplifying tasks and reducing mistakes. Entries provided by the technician as
part of the procedure, such as applied input, are validated for plausibility before being accepted by
the transmitter avoiding gross errors. This same sequence will be followed regardless which EDDL
software the technician chooses to use.

Typical steps in a sensor zero trim calibration wizard for a pressure transmitter are:
• Instruct technician to tell operations to put the associated control loop in manual so control
is not upset when PV changes when sensor reading changes.
• inform the technician the sensor reading will change
• Instruct the technician to apply zero physical input (e.g. by isolate, equalize, and vent the
manifold)
• Instruct the technician to wait while the sensor reading stabilizes and is corrected by the
transmitter.
www.eddl.org 14
• Inform the technician the zero sensor trim was successful
• Instruct technician to tell operations the associated control loop can be put back in automatic

Because the sequence is driven according to the steps defined by the transmitter manufacturer, all
technicians will follow that same sequence when calibrating that type of transmitter. All
transmitters will be calibrated the same way every time. Transmitter manufacturers use conditionals
to make the wizards more intelligent, for instance only ask for user input which is relevant based on
prior selection, only display valid options, and evaluate plausibility (if it is reasonable) of any user
input.

Figure 11 Calibration wizard is created by the device manufacturer's expert to guide the user

7.2 Consistent
EDDL (www.eddl.org) has a multilingual dictionary including standard text for user prompts in the
wizards, error messages, parameter labels, and even help. This dictionary is references by
transmitter manufacturers, thus providing consistency making work more intuitive and reducing
errors because transmitters of different type from different manufacturers will use the same message
(refer to separate white paper on consistent look & feel).

7.3 Context Sensitive Help


The transmitter manufacturer’s factory experts share their know-how in the form of help text
embedded into the EDDL file for their device. This includes help for configuration/setup and
diagnostics parameters, status indicators, and wizards such as for sensor trim. The help for any
wizard or parameter can be pulled up at the click of a button. This simplifies work by minimizing
the need to refer to manuals. Even if the information is buried deep in the manual, it's easily
accessed from the device management software thanks to EDDL.

www.eddl.org 15
Figure 12 Transmitter know-how provided by the device manufacturer's expert

8 Better Calibration Results


EDDL reduces sensor trim mistakes and enforces work consistency among technicians. Calibration
work can be completed faster as the need to carry and refer to manuals and procedures is reduced.
Transmitters perform better when properly calibrated

Other device integration technologies do not work on handheld field communicators, thus are not as
convenient and portable. Other device integration technologies do not have a standard dictionary
and therefore do not provide the consistency and ease of use.

Mixing EDDL with other device integration technologies will make calibration more difficult. If
transmitters are used only with EDDL they are all calibrated the same way. If EDDL is mixed with
other technologies there will be more than one way of doing calibration. Moreover, the drawbacks
and limitations of other technologies will be introduced into the system (refer to white paper on
system administration).

9 Pre-Commissioning Calibration
If there is a need to "calibrate" (perform sensor trim) or not the transmitters before installation is a
common question. Don’t calibrate for the sake of calibrating. Since sensor trim is done in the
factory when the transmitter is manufactured, to trim the sensor or not before installation depends
on if it will drift during shipment. These days transmitters are very rugged and don't drift during
shipment. Hence there is typically no need to perform sensor trim before installation.

The 4-20 mA/HART, FOUNDATION fieldbus, WirelessHART, and PROFIBUS version of a


device use the same sensor. Therefore they have the same basic accuracy and stability. The 4-20
mA current loop adds a little inaccuracy and drift, but less than the sensor. Thus doing sensor trim
or not before installation does not depend on the protocol, it's how rugged the sensor is.

Pressure transmitters needs zero trim to correct mounting position effect.

Usually, this question only comes up for pressure and temperature transmitters - presumably
because portable calibrators exist for temperature and pressure. There is no discussion about site
www.eddl.org 16
calibration of the flowmeters for gas or liquid, and not for many other transmitters - presumably
because it would be very much more difficult to do.

9.1 Loop Check


Loop check moved to a separate tutorial document. See www.eddl.org

10 Beyond Sensor Trim


EDDL-based device management software goes beyond sensor trim and range setting.

10.1 Manuals and Notes


EDDL-based device management software is also capable of giving access to related
documentation. Instruction manuals from the transmitter manufacturer and plant procedures can be
opened at the click of a button.

Figure 13 Intelligent Device Management (IDM) software gives access to documentation like
instruction manuals

10.2 Audit Trail


EDDL-based IDM software also supports audit trail functionality of sensor trim and range changes
etc. Since the audit trail function is independent of the EDDL file, EDDL-based IDM software
supports audit trail for all devices, not just some.

www.eddl.org 17
Figure 14 IDM software based on EDDL supports audit trail for all devices, not just some

When used with documenting calibrators, the calibration records include "as found" / "as left"; that
is percentage error graph at five points before and after sensor trim was performed.

10.3 Maintenance Schedule


EDDL-based device management software is also able to schedule maintenance tasks such as
inspection and calibration.

11 Documenting Calibrators
A calibrator can be used to perform sensor trim or current trim, but is not required for range setting.
Calibrators are primarily used for pressure and temperature sensor trim. For smart transmitters using
4-20 mA/HART, FOUNDATION fieldbus, or PROFIBUS-PA, a documenting calibrator is a good
option as it enables “paperless calibration”.

11.1 Communication
Traditionally an instrument technician on a calibration job carries two tools: a portable calibrator for
the input to the transmitter and to measure the analog output plus a handheld field communicator to
send the trim commands.

A documenting calibrator combines the functionality of a portable calibrator with that of a handheld
field communicator into a single tool. The portable documenting calibrator communicates digitally
with the intelligent transmitters and is able to retrieve all the identification information in the
transmitter as well as range information and sensor limits etc. The documenting calibrator also
retrieves the measured value digitally from the transmitter. The documenting calibrator not only
compares the sensor reading against the standard, but is also able to send the trim commands to the
transmitters when the technician performs sensor trim or current trim so as to achieve an accurate
reading. Documenting calibrators supporting 4-20 mA/HART, FOUNDATION fieldbus, and
PROFIBUS-PA are available. Some documenting calibrators are based on EDDL supporting
wizards, thus making the instrument technician’s job easier.

Therefore, instead of having to carry two tools, the instrument technician carries a single tool.

www.eddl.org 18
11.2 Record Keeping
Traditionally an instrument technician on a calibration job carries two tools also carries a clipboard
with a calibration form where the calibration job is documented.

A documenting calibrator can automatically document the calibration performed, thus also
eliminating the need to carry a clipboard for writing down all the data by hand in the field. The
documenting calibrator internally stores the input points applied and the corresponding output
generated by the transmitter in the calibration process. These data points can be stored for both “as
found” before trim and “as left” after trim was done based on the transmitter's measured value
digitally preserving full precision. Identification information such as manufacturer, device type,
version, and serial number and include as part of the calibration result documentation without
requiring manual data entry. This saves time by avoiding manual data entry and mistakes.
Documenting calibrators store the calibration data internally, until they are docked and the data
automatically uploaded to (synchronized with) the audit trail in the Intelligent Device Management
(IDM) software or their dedicated calibration management software. The software permits the
calibration report to be generated and printed.

Manual data recording and data entry is not required, reducing the burden on technicians and
minimizing mistakes. While handwritten paperwork documentation for the calibration may be
illegible or incomplete, data from a documenting calibrator is always legible and complete for
record keeping.

11.3 Calibration Route


A calibration route is the order in which instruments around the plant will be visited and calibrated.
Generally, those instruments in one area of the plant, but it could be by instrument type or by
calibration due date, whichever is most efficient. The route list of devices to be calibrated is created
in the Intelligent Device Management (IDM) software or their dedicated calibration management
software, and then downloaded to the documenting calibrator.

12 Single Solution
Because EDDL is a text file from which the device page graphics is rendered by the device
management software, the calibration wizards and help are rendered the same way for all devices
regardless of protocol, manufacturer, or type. This consistency achieved thanks to EDDL makes
calibration easier and intuitive (see separate technical white paper on consistency of use). No other
technologies can provide a comparable result.

13 Conclusion
For plants that are looking for an easy solution to calibrating multiple types and versions of
transmitters, EDDL technology is a perfect match. EDDL meets the need of plants to calibrate all
types of devices from a single software application using a single technology while at the same time
making it fast and easy to keep the system up to date with new device versions. Plants should
upgrade existing DD systems to EDDL with enhancements to enjoy the greater ease of use afforded
by the standard graphical display.

Calibration trim is just one of the ways in which intelligent device management software based on
EDDL reduces maintenance & operational cost. Please refer to other technical white papers on
advanced diagnostics, configuration/setup, and system administration etc.

14 Questions
Post your questions on transmitter calibration and intelligent device management with the EDDL
group on Linkedin:
http://www.linkedin.com/groups?gid=3736433

www.eddl.org 19
15 References
EDDL Brochure and Technical Description on www.eddl.org site

Jonas Berge, "Fieldbuses for Process Control: Engineering, Operation, and Maintenance", ISA,
2002, ISBN 1-55617-760-7

ANSI/ISA–51.1–1979 (R1993) “Process Instrumentation Terminology”, Reaffirmed 26 May 1995,


ISBN 0-87664-390-4

Appendix A: FF Parameters
Systems based on enhanced EDDL shows range, calibration, and other information organized in
hierarchical menu trees, tabs, and frames with human readable labels etc. However, older systems
based on traditional DD (without EDDL enhancements) typically show parameters in list or table
form, displayed with a cryptic parameter name. These are the parameters:

Data Block Parameter Name


Lower Range Value (LRV) AI function XD_SCALE.EU_0
Upper Range Value (URV) AI function XD_SCALE.EU_100
Converted AI function OUT_SCALE.EU_0
Converted AI function OUT_SCALE.EU_100
PV% AI function FIELD_VAL
Lower Sensor Limit (LSL) Transducer SENSOR_RANGE.EU_0
Upper Sensor Limit (USL) Transducer SENSOR_RANGE.EU_100
Minimum Span Transducer CAL_MIN_SPAN
Lower Sensor Trim Point Transducer CAL_POINT_LO
Upper Sensor Trim Point Transducer CAL_POINT_HI
*Lower Range Value (LRV) Transducer PRIMARY_VALUE_RANGE.EU_0
*Upper Range Value (URV) Transducer PRIMARY_VALUE_RANGE.EU_100

In most devices, the transducer block PRIMARY_VALUE_RANGE cannot be configured from the
transducer block. If at all used, transmitter range is configured in the AI function block parameter
XD_SCALE which automatically updates the transducer block PRIMARY_VALUE_RANGE.
However, some transmitter may require the range to be configured manually in both places. That is,
there are two types of FF transmitters:
1. Transmitters where range can only be set in the AI block XD_SCALE parameter (not in the
transducer block) making BLOCK_ERR due to XD_SCALE mismatch impossible. This is easy
to use.
2. Devices where range can be set in both the XD_SCALE parameter in the AI block and in the
PRIMARY_VALUE_RANGE parameter in the transducer block making BLOCK_ERR due to
XD_SCALE mismatch possible. This can be confusing..

Internally the FF AI block computes:

FIELD_VAL = (PRIMARY_VARIABLE - XD_SCALE::0) / (XD_SCALE::100 - XD_SCALE::0) *


100 [%}

PV = (FIELD_VAL / 100) * (OUT_SCALE::100 - OUT_SCALE::0) + OUT_SCALE::0

Note the AI block must be in Out of Service mode in order to change XD_SCALE or OUT_SCALE

www.eddl.org 20
OUT_SCALE

400 bbl/day

300 bbl/day

200 bbl/day
Square Root
100 bbl/day

0 bbl/day XD_SCALE
0 62.5 125 182.5 250
inH2OinH2OinH2OinH2OinH2O

If the pressure transmitter is measuring pressure or DP (L_TYPE is ‘direct’), there is no need to set
XD_SCALE. Setting XD_SCALE is only required for DP flow and DP level applications where the
transmitter shall convert and scale a DP into flow or level. When L_TYPE is ‘direct’ the pressure
reading will be seen in the transducer block and on the AI block PV and OUT

Appendix B: History of 3-15 psi, 10-15 mA, and 4-20 mA signals


Digital HART communication is superimposed over the analog 4-20 mA signal. Before 4-20 mA
the signals were 3-15 psi and 10-50 mA. These signals all have a “live zero” (3 psi, 10 mA, and 4
mA) rather than a “dead zero” (0 psi, 0 mA) which can be used to detect a severed pneumatic tube
or signal wire. The live zero can also be used to power a 2-wire loop powered device. 3-15 psi, 10-
50 mA, and 4-20 mA etc. all have a ratio of 1:5 (20% bias). Nobody remembers the exact reason
why the signal ranges 3-15 psi, 10-50 mA, and 4-20 mA were selected and there appears to be no
definite source documenting this decision such as minutes of meeting from a standards committee.
Some research has uncovered the following technical reasons why these signal ranges were chosen.

3-15 psi pneumatic signal


Pneumatic instruments operate on the flapper-nozzle (baffle-nozzle) principle. It appears 3-15 psi
with the 1:5 ratio was chosen because it the most linear portion on the curve for the movement of
the flapper (baffle) and the backpressure resulting in the nozzle.

10-50 mA analog electronic signal


Early analog electronic instruments used magnetic amplifiers. 10 mA live zero was chosen as it is
the lowest at which instrument based on magnetic amplifiers could operate. Maintaining the 1:5
ratio the signal was chosen 10-50 mA.

4-20 mA analog electronic signal


With the introduction of the transistor it became possible to make devices operating on 4 mA. Thus
the signal was chosen 4-20 mA maintaining the 1:5 ratio.

With the introduction of microprocessors in instrumentation and digital communication, the HART
protocol was introduced superimposed over the 4-20 mA signal. The HART commands supported
by the device, how device information is displayed, and the wizards to make tasks like calibration
easy are described using EDDL.

Appendix C: “Failure mode alarm” and NAMUR NE43


4-20 mA instrumentation and controls usually support a signal range slightly below 4 mA and
above 20 mA. For transmitters, current values below 4 mA and above 20 mA are used to signal a
fault such as a thermocouple burnout or other sensor failure. The transmitter can be configured for
failure indication low or high. Unfortunately manufacturers use different signal levels to indicate
failure which prevents tight analog signal integration and interpretation in single loop controllers,
control systems, and safety systems. Some transmitters may use 3.75 mA while others may use 3.6
mA or less. Some transmitter may use 21.75 mA or more while other use 23 mA. This
inconsistency of signal levels for failure indication makes it difficult to take full advantage of the
www.eddl.org 21
failure information in control strategies. The NAMUR NE43 “Standardization of the Signal Level
for the Failure Information of Digital Transmitters” recommendation was created to standardize
failure indication from transmitters and interpretation in control systems to enable better analog
integration. NE43 defines 3.8-20.5 mA as a valid (‘Good’) measurement value where 3.8-4 and 20-
20.5 mA indicates saturation. A signal of <3.6 mA or >21 mA indicates a transmitter failure
(‘Bad’).

mA %
3.6 -2.50 Failure low
3.8 -1.25 Saturated low
4.0 0.00 Lower Range Value
20.0 100.00 Upper Range Value
20.5 103.13 Saturated high
21.0 106.25 Failure high

Analog Current

106.25% / 21 mA Failure High


103.13% / 20.5 mA Saturated High
100% / 20 mA

Valid range

0% / 4 mA Saturated Low
-1.25% / 3.8 mA
-2.50% / 3.6 mA Failure Low
0 mA

By using transmitters and systems that both conform to the NE43 recommendation, it is possible to
flag faults to the operators and control strategies. However, note that all device errors, severe and
trivial, are flagged the same way so the operator cannot tell the difference and that if any error
occurs it is flagged and the measurement value is not provided.

FOUNDATION fieldbus and PROFIBUS-PA transmitters use fast digital communication with
separate status indication for each measurement including measurement validity flagged in real-
time as ‘Good’, ‘Bad’, or ‘Uncertain’. This allows operators and control strategies to severe
problems from trivial issues. This allows the control strategy to put the loop in manual in case of
failure, with the option to trip. For non-severe issues the value is still displayed with ‘Uncertain’
status.

Smart Valve Positioners


Smart valve positioners are not in the scope of NAMUR NE43. However, signals <4 mA and >20
mA also have specific meanings. Control systems and single loop controllers with 4-20 mA output
use a similar scheme to achieve tight shut-off for control valves. That is, they may set current <<4
mA or >>20 mA. Some control systems set the output current to 0 mA to achieve tight shutoff. This
is impractical in the case of smart valve positioners since they need 3.6 mA to operate and will be
completely switched off at 0 mA. Therefore make sure to configure the control system or single
loop controller to provide at least 3.6 or 3.8 mA for tight shutoff to ensure that the smart valve
positioner can continue to operate and respond to HART communication.

Appendix D: Bench Calibration Tools


The same handheld field communicator is used to calibrate 4-20 mA/HART and Fieldbus
transmitters.

www.eddl.org 22
Alternatively, the same Intelligent Device Management (IDM) software or configuration software is
used to calibrate 4-20 mA/HART and Fieldbus transmitters. A 4-20 mA/HART transmitter requires
the laptop to be fitted with a HART-USB interface while a Fieldbus transmitter requires a Fieldbus-
USB interface. A PROFIBUS-PA device requires a PA-USB interface.

A 4-20 mA/HART transmitter has to be powered by a DC power supply and 250 ohm resistor, all
connected in series. A Fieldbus or PROFIBUS-PA transmitter has to be powered by a fieldbus
power supply, connected in parallel. Portable fieldbus power battery packs are convenient both for
field work and on the bench. A fieldbus transmitter could also be powered by the Fieldbus-USB
interface eliminating the need for separate power.

Thanks to enhanced EDDL graphics and wizards, the procedure for calibrating the 4-20 mA/HART
version of a device is the same as calibrating the Fieldbus version of that same device.

Appendix E: Low Cut-Off


Low cut-off is used in flow metering in order to avoid irratic control and false totalization. It is
predominantly used for DP flow meters. The low cur-off function turns off the square root function
below the cut-off point. This is done because the square root function has very high gain closer to
zero so any measurement noise gets amplified and results in irratic control and false totalization etc.
All DP transmitters have this function. In the past this was a "hard" cut-off making flow zero below
the cutoff point. Today "soft" cutoff is more common, linearly interpolating the flow between zero
and the cut-off point below the cut-off.

In some 4-20 mA devices the cut-off point is specified on DP while in other transmitters the cut-off
is defined on flow. There is a big difference in that a 4% cut-off in DP units corresponds to a 20%
cutoff in flow units.

www.eddl.org 23

You might also like