MOS Ka Test

CMOS Testing
ROCHESTER INSTITUTE OF TECHNOLOGY

MICROELECTRONIC ENGINEERING
CMOS Testing for the Student Run Factory

Dr. Lynn Fuller
Webpage: http://people.rit.edu/lffeee
Microelectronic Engineering
Rochester Institute of Technology
82 Lomb Memorial Drive
Rochester, NY 14623-5604
Tel (585) 475-2035
Email: Lynn.Fuller@rit.edu
Department webpage: http://www.microe.rit.edu

Revised 4-30-2015 cmostest.ppt
April 30, 2015, Dr. Lynn Fuller, Professor
Page 1
CMOS Testing
OUTLINE
Introduction
Test Structures
Test Equipment
Resistive Tests
Transistors
Integrated Circuits
Wafer Mapping

Page 2
CMOS Testing
INTRODUCTION
Completed factory wafers are tested as part of the manufacturing
process to capture data on processing parameters and transistor
characteristics. Some simple integrated circuits are tested to verify
functionality of more complex circuits. Uniformity is evaluated by
measuring some parameters, such as threshold voltage, over the
entire wafer.
These tests are broken into four tasks that can be completed in our
normal three hour laboratory time: 1. sheet resistance and contact
resistance, 2. transistors, 3. inverters, ring oscillator and op amp,
and 4. wafer map of NMOSFET threshold voltage.
More complex digital and analog integrated circuits are tested
outside of the factory by those who are interested.
Page 3
CMOS Testing
INTRODUCTION
Chip designs always include test structures for the factory in addition
to the specific integrated circuit or microsystem being fabricated. The
test structures are often located along the edges of the chip and might
even be removed after testing during the wafer sawing process. These
test structures are available as cells that can be easily added to the
integrated circuit or microsystem design. The test setups can be used
for all completed chips. Automated probing (rarely done at RIT) may
require different setups if the test structures are in different locations
on the wafer.
Three other documents provide additional details:
CMOSTestchip2008.pdf - description of latest factory chip
CMOSTEST_Manual.pdf details for test equipment operation
TestResults.pdf recent factory test results and template for
reporting test results
Page 4
CMOS Testing
TEST STRUCTURES
List of factory test structures:

Van Der Pauws for sheet resistance
CBKRs for contact resistance
NMOS and PMOS transistors of various sizes
Field Oxide NMOS and PMOS transistors
Inverter
Ring Oscillator
Op Amp
Metal serpentines
Via Chains

Page 5
CMOS Testing
VAN DER PAUWs AND CBKRs
NWELL PWELL N+
P+
N-POLY M1
P-POLY M2
2m M1toP+
2m M1toPoly
2m M1toN+
2m M1toM2
2m M1toN+
2m
M1toP+
4m M1toPoly
4m M1toM2
4m M1toP+
4m M1toP+
4m M1toN+
4m M1toN+
Page 6
CMOS Testing
SERPENTINES, COMBS, AND VIA CHAINS
To evaluate metal1, metal2, CC and Via layer quality.

Page 7
CMOS Testing
NMOS AND PMOS TRANSISTORS
1/4 2/4 4/4
8/4 16/4 32/4
1/8 2/8 4/8
8/8 16/8 32/8 1/32 2/32 4/32 8/32 16/32 32/32
Various
L/W
Ratios

NMOS 2/8
Page 8
PMOS 2/8
CMOS Testing
FIELD OXIDE NMOS AND PMOS FETS

Page 9
CMOS Testing
CMOS INVERTER
Vout
Vin
+V
Idd
PMOS
Vout
Vin
NMOS
CMOS
TRUTH TABLE
VIN
0
1
VOUT
1
0

W = 40 m
Ldrawn = 2.5m
Lpoly = 1.0m
Leff = 0.35 m
Page 10
CMOS Testing
INVERTER WITH PADS
INV/NOR4

Page 11
CMOS Testing
RING OSCILLATORS
17 Stage Un-buffered Output
L/W 8/16
4/16
L/W=2/30 Buffered Output
2/16
73 Stage

Page 12
37 Stage
CMOS Testing
OPERATIONAL AMPLIFIERS
Version 1
Page 13
CMOS Testing
TEST EQUIPMENT
Computer
HP4145
Semiconductor Paramater
Analyzer
IEEE
488
Switch Matrix
ICS (metrics)
Osprey
(video capture)
Microsoft Office
Ultracision
Semi-Automatic
Wafer Prober

Test Fixture
and
Manual Probe Station
Page 14
CMOS Testing
TEST EQUIPMENT
Semi-Automatic Prober
Automatic Prober

Page 15
CMOS Testing
TEST EQUIPMENT

Manual Prober
Page 16
CMOS Testing
STANDARD PROBE CARDS & SWITCH MATRIX

9
Numbers indicate switch matrix column

7 6 5 4
8 7 6 5
4
10 Pad
Probe
Card
12 Pad
Probe
Card
10 11 12 1
11

12 1 2 3
Probe Card
Page 17
CMOS Testing
SWITCH MATRIX
Switch Matrix
Indicator
On Button
Copy Button
Light Pen
The HP-4145 SMU-1 is connected to row 1, SMU-2 to row 2, SMU-3

to row 3 and SMU-4 to row 4. The probe pins are connected to the
columns. The switch matrix indicator above shows SMU1 connected
to col. 1 which is the 4th pin on the bottom row of the 12 pad probe
card (see previous page)
Page 18
CMOS Testing
PROBING A NMOS TRANSISTOR

9 8
8
S
7
D
7 6 5
10 11 12 1 2
Sub G
11 12

2/16 L/W NMOS FET
Page 19
CMOS Testing
ICS TEST SET UP for NMOSFET FAMILY OF CURVES

9 8
7 6 5
SMU1
S
Sub G
10 11 12 1 2
SMU2
SMU4
Test Name: NFAM

1 2 3 4 5 6 7 8 9 10 11 12
SMU3
SMU1
SMU2
SMU3
SMU4
Switch matrix
Sub G
SMU1
SMU2
SMU3
SMU4
Vds
Vgs
Com
Com
Sweep 0 - 5 V, 101 steps

11 Steps 0 - 5V
Vs = 0
0V
Page 20
CMOS Testing
GENERAL TEST INSTRUCTIONS

Test die in the center of the wafer, then upper left, upper right, lower
right, and lower left (about way between center and edge).
Extract parameter values from the test results.
Create a PowerPoint document from test_results.ppt master (see
example data powerpoint a few pages below) on Dr. Fullers
webpage at http://people.rit.edu/lffeee/labnotes.htm (save as) record
Lot#, Wafer#, Die location (center, top left,etc), pictures of die, test
results graphs, extracted parameters and comments. Email to Dr.
Fuller at Lynn.Fuller@rit.edu
Test 01 Van Der Pauw and CBKR. Record Average of five tests
Test 02 Transistors, test small transistors (L=2m for SMFL,
L=1m for Sub-CMOS and Adv-CMOS). Record results in
power point document.
Test 03 Inverters,
Ring Oscillator. Record Average of five tests.
Microelectronic
Engineering
Test 04 NMOS
VT
wafer map
Page 21
CMOS Testing

Substrate or Well Connections: Most of the test structures
incorporate diffusions. In Resistors, Van der Pauws and Transistors
the junctions between the diffusions and the substrate/wells are
normally never forward biased. As a result the test engineer needs to
evaluate the applied test voltages and connections to the
substrate/wells and connections to the diffusions to ensure proper
bias conditions.
For example: a P+ Van der Pauw in an N-type Well requires that the
Well connection always have the highest positive voltage that is
applied. If a separate (5th pad) connection is available (not often
because there are only 4 SMUs) that can be set to a high voltage.
Otherwise the substrate is normally connected to one of the four
pads of the Van der Pauw. This pad can be swept with positive
voltage thus keeping the substrate/well junction reverse biased.
Page 22
CMOS Testing

Each test requires you to:
1. Find the structure you want to test
2. Place the probes
3. Open the test by restoring the testname-1 (example PFAM-1) in
ICS, view the test setup to see what SMUs do what.
4. Set the switch matrix for the HP4145 SMUs to the probes you are
using, consistent with the test setup.
5. Edit the graph by making changes in the title, moving the cursors
to the correct location
6. Copy the plot using ctrl print screen, (paste into word, copy from
word to power point, crop and paste in correct location)
7. Extract the data, such as threshold voltage or LAMBDA and enter
the value in the data table in the powerpoint
8. Save the powerpoint, minimize the data plot on ICS
9. When done email the powerpoint to Lynn.Fuller@rit.edu
Page 23
CMOS Testing
TE01 VAN DER PAUW AND CBKR

Page 24
CMOS Testing
VAN DER PAUW STRUCTURES FOR SHEET RESISTANCE
V1
Rs
V1-V2
V2
(V1-V2)
Rs =
I
ln 2

V1
V2
Page 25
CMOS Testing
VAN DER PAUW WITH SUBSTRATE CONNECTIONS

N+/P
Poly
P+/N
SMU1
Note: connection to N has to be to the highest

voltage used to prevent P+/N junction from
being forward biased
P+

Page 26
CMOS Testing
VAN DER PAUW WITH SUBSTRATE CONNECTIONS

Page 27
CMOS Testing
HP 4145 OUTPUT PLOT

Page 28
CMOS Testing
CROSS BRIDGE KELVIN RESISTANCE TEST

STRUCTURES FOR CONTACT RESISTANCES
Gc
V1-V2
I
V1
V2
W2
W1
(V1-V2)
Rc =
I
I
1
Gc =
(V1-V2) W1 x W2
Page 29
ohms
mhos/m2
CMOS Testing
CBKR WITH SUBSTRATE CONNECTIONS

N+ in P well
P+ in N well
Note: connection to N+ has to be to the highest

voltage used to prevent N+/P junction from
being forward biased. SMU2 is swept from 0
to +5V, SMU3 is fixed at +5V
see next page.

Poly
SMU2
SMU3
+5V
N+
N+ in P well
Page 30
CMOS Testing
HP 4145 OUTPUT PLOT

Page 31
CMOS Testing
TE02 TRANSISTORS

Page 32
CMOS Testing
TRANSISTOR FAMILY OF CURVES, ID-VDS
Vds
D
Id
+G
Vgs
S
-
Vsub
Lambda = slope/Idsat

Page 33
CMOS Testing
TRANSISTOR LINEAR REGION VT, GM

+Ids
Vd = 0.1 Volt
D
Id
+5
+4
+Vgs
+3
+2
Non Saturation
Region
Vgs G
+Vds
Vsub
S
n
Vsub = 0
+Id
Body Effect
Vsub
gm
-1
-2
-3 volts
gm = Id/Vg
+Vg
Vto

D
n
p
Vsub
nMOSFET with Vt=1, since the

Drain is at 0.1 volts and the
source is at zero. Both drain and
source will be on at gate voltages
greater than 1.1 volt. the
transistor will be in the non
saturation region.
Page 34
CMOS Testing
TRANSISTOR SATURATION REGION VT, GM

+Ids
Id
Saturation Region
+5
+4
+Vgs
+3
+2
NMOS
Vgs=Vds
Vsub
Vsub = 0
+Id
+Vds
n
p
Body Effect
gm
-1
-2
-3 volts
gm = Id/Vg
+Vg
Vto

Vsub
nMOSFET with Vt=1, Drain end is

never on because Voltage Gate to
Drain is Zero. Therefore this
transistor is always in Saturation
Region if the gate voltage is above
the threshold voltage.
Page 35
CMOS Testing
NMOS ID-VGS TEST
Same test for field

oxide threshold voltage
gm = 0.16 mmho/16m
= 0.010 S/mm
Vt = 0.507 volts
Page 36
CMOS Testing
TRANSISTOR SUB THRESHOLD ID-VGS

Id (Amps)
Id
D
+
Vgs=Vds
10-2
10-3
10-4
10-5 Lights On
10-6
10-7
10-8
10-9
10-10
10-11
10-12
Sub Vt Slope
(mV/dec)
Vgs
Vt
The subthreshold characteristics are important in VLSI

circuits because when the transistors are off they should not
carry much current since there are so many transistors.
(typical value about 100 mV/decade)
Page 37
CMOS Testing
SUB THRESHOLD CHARACTERISTICS
Ion/Ioff = ~6 Decades
Swing = ~100 mV/dec

DIBL@1nA/m
= Vg/Vd
= 0 mV/V

Page 38
CMOS Testing
TE03 INTEGRATED CIRCUITS

Page 39
CMOS Testing
INVERTERS
VOUT
+V
VOUT
VIN
+V
Imax
Voh
Idd
Slope = Gain
VO
VIN
Idd
VoL
CMOS
VIN
+V
Vih
ViL
0 noise
margin=ViL-VoL
Microelectronic
Engineering
1 noise
margin=VoH-ViH
Vinv
Page 40
CMOS Testing
CMOS INVERTER VOUT VS VIN

Page 41
CMOS Testing
RING OSCILLATOR, td
Seven stage ring oscillator
with two output buffers
td = T / 2 N
td = gate delay
N = number of stages
T = period of oscillation
Buffer
Vout
T = period of oscillation
Page 42
Vout
CMOS Testing
RING OSCILLATOR MEASUREMENTS
73 Stage Ring at 5V, td = 0.712ns

Page 43
CMOS Testing
OPERATIONAL AMPLIFIER
+V
10
M11
M3
L/W
40/10
M4
2
10/20
M6
10/20
M10
40/10
5
M9
40/10
10/15
M2
M1
Vin-
10/15
10/15
Vin+
M7
M5
Vout
10/20
10/20
10/20
M8
-V
p-well CMOS
20
+V=+6
dimensions
L/W
(m/m)
Vin
Vout
+6
Slope = Gain
Vout
-6
-V=-6

-20mV
Vin
+20mV
Vos
Set up the HP 4145 to sweep the Vin from -20 mV to +20 mV

in 0.001V steps. Measure Gain and Input offset voltage.
Page 44
CMOS Testing
AC TEST RESULTS
ROCHESTER INSTITUTE OF TECHNOLOGY
MICROELECTRONIC ENGINEERING
LFF OPAMP.XLS FILE3B
LOT F960319 OPAMP TEST RESULTS - 1-29-97

Gain
dB
73.9794
73.53387
73.33036
70.31748
67.53154
65.48316
63.97314
62.41148
61.51094
60.34067
59.46256
58.68997
58.0618
57.50123
57.14665
56.5215
56.1236
55.56303
20
0
-20
Vout
V
10
9.5
9.28
6.56
4.76
3.76
3.16
2.64
2.38
2.08
1.88
1.72
1.6
1.5
1.44
1.34
1.28
1.2
0.02
0.002
0.0002
Vin
mV
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Op Amp Frequency Response
GBP = 500,000 Hz
80
70
60
50
Gain dB
Frequency
hZ
1
5
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
50000
500000
10000000
40
30
20
10
0
-10
-20
1
10
100
1000
10000
100000
1000000 10000000
Frequency Hz

Page 45
CMOS Testing
OPERATIONAL AMPLIFIER
Vos =-1.72 volts

Av = 2410
Page 46
CMOS Testing
TE04

Page 47
CMOS Testing
WAFER MAPS FOR MESA

nmos Vt target +1
000000000000000
000050505050000
000000000000000
005060507050700
000000000000000
004040304030300
000000000000000
004040404040400
000000000000000
004050405090600
000000000000000
005050606060700
000000000000000
000050505050000
000000000000000
Col 1
Col 15
Row 1
T
Row 15
row 1 is the first row in which a full die is located

column 1 is the first column in which a full die is located

Page 48
CMOS Testing
WAFER MAPS FOR MESA
Code
0
1
2
3
4
5
6
7
8
9
no die
value<(Target-70%)
(Target-70%)<value<(Target-50%)
(Target-10%)<value<(Target+10%)
(Target+10%)<value<(Target+30%)
(Target+70%)<value

Page 49
CMOS Testing
WAFER MAP
Example: Given a wafer
with test chips located as
shown and nmos threshold
voltage data encoded and
stored in MESA as shown.
Reconstruct a wafer map
using EXCELL spreadsheet.

Page 50
MOSFET IV CHARACTERISTICS
Lot Number = F050118
Process = SMFL CMOS
Wafer Number = D4
Product = DAC03
Date = 11-17-2006
NMOS L/W = 2/16

NMOS Id-Vds
Family of Curves
NFAM
NMOS Id-Vgs
NVT
NMOS Subthreshold
Characteristics
NSUB
NMOS Field VT
NFIELD
PMOS L/W = 2/16

PMOS Id-Vds
Family of Curves
PFAM
PMOS Id-Vgs
PVT
PMOS Subthreshold
Characteristics
PSUB
PMOS Field VT
PFIELD
MOSFET EXTRACTED PARAMETERS

Lot Number = F050118 Wafer Number = D4, Die Location R= , C=
PMOS
2/16
NMOS
2/16
VT
-1.51
1.36
Lambda (for Vgs = Vdd)
0.115
0.0417
V-1
Max gm / mm of channel width
21.3
31.3
S/mm
Idrive
54.4
93.8
A/m
Ion/Ioff @ Vd = 0.1V
Decades
Ion/Ioff @ Vd = 5V
Decades
Ioff @ Vd = 0.1V
5.9e-11
5.0e-10
A/m
Ioff @ Vd = 5V
5.9e-11
5.0e-10
A/m
Sub-Vt Slope @ Vd = 0.1V
90
190
mV/Dec
Sub-Vt Slope @ Vd = 5 V
90
190
mV/Dec
DIBL@1nA/m =Vg/Vd
mV/V
-23
23
Mask Length / Width
Field VT
Units
m
INVERTERS, VAN DER PAUW AND CBKR

Lot Number = F050118 Wafer Number = D4 , Die Location R= , C=
Inverter
Poly
Van Der Pauw
M1
Van Der Pauw
N+
Van Der Pauw
P+
Van Der Pauw
P-Well
Van Der Pauw
M1 to Poly
CBKR
M1 to N+
CBKR
M1 to P+
CBKR
EXTRACTED PARAMETERS FROM

INVERTERS, VAN DER PAUW AND CBKR
Lot Number = F050118 Wafer Number = D4 , Die Location R=
Contact Gs
P+
N+
poly
CBKR
42
mmho/m2
8
mmho/m2
37
mmho/m2
Ring Oscillator
# Stages
Period
td
Rhos
P+
N+
well
Poly
Al
73
104
0.712
Van der Pauw
115
5.8
614
20.0
0.0662
Vdd=5
V
nsec
nsec
Ohms
Ohms
Ohms
Ohms
Ohms
, C=
Inverter
VinL
2.4
VinH
VoL
0.4
VoH
4.5
Vinv
2.6
Imax
4.5
mA
Gain
-21.5
0=ViL-VoL
2.0
1=VoH-ViH
1.5
OpAmp
Gain
None
Offset
None
mVolts
GBW
None
Hz
M1-P+
M1-M2
VIA CHAIN
None
None
ohms
ohms
Lot Number = F050118

Wafer Number = D4
Family of curves for L=2m MOSFETs
2m/32m L/W NMOS AND PMOS
CMOS Testing
RESULTS
A test specification has been developed

Sheet Resistance of Poly, Metal, P+ DS, N+ DS, Well
Transistor Vt, gm, sub threshold slope, Lambda, field
Vt
ICs, inverter, ring oscillator, opamp
A history/data base has been built
Analog parametric tests for some devices has been
developed (much more to do)

Page 57
CMOS Testing
FUTURE WORK
More Automation
Improved Wafer Mapping
More Complete Testing
Improved Analog Testing
Improved Digital Testing

Page 58
CMOS Testing
CONCLUSION
Testing is very time consuming. It takes us 9 hours to
do all the specified tests and even then we only test a
few devices on a wafer.
Currently we test about 1% of the devices

Page 59
CMOS Testing
HOMEWORK CMOS TEST

1. How is Lambda, Vt and gm found from the transistor
family of curves? Show example calculations.
2. How are Vt and gm found from the Id-Vgs plot?
3. Explain how sub threshold swing, Ion/Ioff, and DIBL are
found. Show example calculations and correct units of
measure.
4. What is the significance of inverter noise margin.
5. What is the purpose of the ring oscillator test structure.

Page 60
CMOS Testing
MOSFET TEST RESULTS TEMPLATE

Page 61

MOS Ka Test

Uploaded by

Copyright:

Available Formats

MOS Ka Test

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MOS Ka Test

Uploaded by

Copyright:

Available Formats

CMOS Testing

ROCHESTER INSTITUTE OF TECHNOLOGY

CMOS Testing for the Student Run Factory

Rochester Institute of Technology

Revised 4-30-2015 cmostest.ppt

April 30, 2015, Dr. Lynn Fuller, Professor

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

April 30, 2015, Dr. Lynn Fuller, Professor

April 30, 2015, Dr. Lynn Fuller, Professor

List of factory test structures:

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

VAN DER PAUWs AND CBKRs

April 30, 2015, Dr. Lynn Fuller, Professor

SERPENTINES, COMBS, AND VIA CHAINS

To evaluate metal1, metal2, CC and Via layer quality.

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

NMOS AND PMOS TRANSISTORS

1/4 2/4 4/4

8/4 16/4 32/4

1/8 2/8 4/8

8/8 16/8 32/8 1/32 2/32 4/32 8/32 16/32 32/32

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

FIELD OXIDE NMOS AND PMOS FETS

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

INVERTER WITH PADS

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

L/W=2/30 Buffered Output

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

April 30, 2015, Dr. Lynn Fuller, Professor

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

Rochester Institute of Technology

STANDARD PROBE CARDS & SWITCH MATRIX

Numbers indicate switch matrix column

Rochester Institute of Technology

April 30, 2015, Dr. Lynn Fuller, Professor

The HP-4145 SMU-1 is connected to row 1, SMU-2 to row 2, SMU-3

April 30, 2015, Dr. Lynn Fuller, Professor

PROBING A NMOS TRANSISTOR

Rochester Institute of Technology

2/16 L/W NMOS FET

April 30, 2015, Dr. Lynn Fuller, Professor

ICS TEST SET UP for NMOSFET FAMILY OF CURVES

Test Name: NFAM

Sweep 0 - 5 V, 101 steps

April 30, 2015, Dr. Lynn Fuller, Professor

GENERAL TEST INSTRUCTIONS

GENERAL TEST INSTRUCTIONS