VLSI Testing - DFT and Scan
-
Upload
abdul-rasheed -
Category
Documents
-
view
239 -
download
0
Transcript of VLSI Testing - DFT and Scan
-
8/12/2019 VLSI Testing - DFT and Scan
1/35
Copyright 2001, Agrawal & BushnellCopyright 2001, Agrawal & Bushnell
VLSI Testing
DFT and Scan
VLSI Testing
DFT and Scan
Definitions
Ad-hoc methods
Full Scan design Partial scan
Boundary scan
Summary
-
8/12/2019 VLSI Testing - DFT and Scan
2/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 2Copyright 2001, Agrawal & Bushnell 2
DefinitionsDefinitions
Design for testabil ity (DFT) refers to those design
techniques that make test generation and testapplication cost-effective.
DFT methods for digital circuits:
Ad-hoc methods
Structured methods: Scan
Partial Scan
Built-in self-test (BIST)
Boundary scan
DFT method for mixed-signal circuits: Analog test bus
-
8/12/2019 VLSI Testing - DFT and Scan
3/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 3Copyright 2001, Agrawal & Bushnell 3
Ad-Hoc DFT MethodsAd-Hoc DFT Methods Good design practices learnt through experience are used as
guidelines:
Avoid asynchronous (unclocked) feedback.
Make flip-flops initializable.
Avoid redundant gates. Avoid large fanin gates.
Provide test control for dif ficult-to-control signals.
Avoid gated clocks. Consider ATE requirements (tristates, etc.)
Design reviews conducted by experts or design auditing tools.
Disadvantages of ad-hoc DFT methods:
Experts and tools not always available. Test generation is often manual with no guarantee of high fault
coverage.
Design iterations may be necessary.
-
8/12/2019 VLSI Testing - DFT and Scan
4/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 4
Difficulties in Seq. ATPGDifficulties in Seq. ATPG
Poor initializability.
Poor controllability/observability of state variables. Gate count, number of flip-flops, and sequential depth
do not explain the problem.
Cycles are mainly responsible for complexity.
An ATPG experiment:
Circuit Number of Number of Sequential ATPG Fault
gates flip-flops depth CPU s coverage
TLC 355 21 14* 1,247 89.01%
Chip A 1,112 39 14 269 98.80%
* Maximum number of flip-flops on a PI to PO path
-
8/12/2019 VLSI Testing - DFT and Scan
5/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 5Copyright 2001, Agrawal & Bushnell 5
Scan DesignScan Design Circuit is designed using pre-specified design rules.
Test structure (hardware) is added to the verifieddesign:
Add a test control (TC) primary input.
Replace flip-flops by scan flip-flops (SFF) and connect to form
one or more shift registers in the test mode. Make input/output of each scan shift register
controllable/observable from PI/PO.
Use combinational ATPG to obtain tests for all testable
faults in the combinational logic. Add shift register tests and convert ATPG tests into
scan sequences for use in manufacturing test.
-
8/12/2019 VLSI Testing - DFT and Scan
6/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 6Copyright 2001, Agrawal & Bushnell 6
Scan Design RulesScan Design Rules
Use only clocked D-type of flip-flops for all state
variables.
At least one PI pin must be available for test; more
pins, if available, can be used. All clocks must be controlled from PIs.
Clocks must not feed data inputs of flip-flops.
-
8/12/2019 VLSI Testing - DFT and Scan
7/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 7Copyright 2001, Agrawal & Bushnell 7
Correcting a Rule ViolationCorrecting a Rule Violation All clocks must be controlled from PIs.
Comb.
logic
Comb.
logic
D1
D2
CK
Q
FF
Comb.
logic
D1
D2
CK
Q
FF
Comb.
logic
-
8/12/2019 VLSI Testing - DFT and Scan
8/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 8Copyright 2001, Agrawal & Bushnell 8
Scan Flip-Flop (SFF)Scan Flip-Flop (SFF)D
TC
SD
CK
Q
QMUX
D flip-flop
Master latch Slave latch
CK
TC Normal mode, D selected Scan mode, SD selected
Master open Slave open
t
t
Logicoverhead
-
8/12/2019 VLSI Testing - DFT and Scan
9/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 9Copyright 2001, Agrawal & Bushnell 9
Level-Sensitive Scan-Design
Flip-Flop (LSSD-SFF)
Level-Sensitive Scan-Design
Flip-Flop (LSSD-SFF)
D
SD
MCK
Q
Q
D flip-flop
Master latch Slave latch
t
SCK
TCK
SCK
MCK
TCK
N
ormal
mode
MCK
TCK Scan
mode
Logic
overhead
-
8/12/2019 VLSI Testing - DFT and Scan
10/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 10Copyright 2001, Agrawal & Bushnell 10
Adding Scan StructureAdding Scan Structure
SFF
SFF
SFF
Combinational
logic
PI PO
SCANOUT
SCANIN
TC or TCK Not shown: CK orMCK/SCK feed all
SFFs.
-
8/12/2019 VLSI Testing - DFT and Scan
11/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 11Copyright 2001, Agrawal & Bushnell 11
Comb. Test VectorsComb. Test Vectors
I2I1 O1 O2
S2S1 N2N1
Combinational
logic
PI
Present
state
PO
Next
state
SCANINTC
SCANOUT
-
8/12/2019 VLSI Testing - DFT and Scan
12/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 12Copyright 2001, Agrawal & Bushnell 12
Comb. Test VectorsComb. Test Vectors
I2I1
O1 O2
PI
PO
SCANIN
SCANOUT
S1 S2
N1 N2
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0TC
Dont care
or random
bits
Sequence length= (ncomb + 1) nsff+ ncomb clock periods
ncomb = number of combinational vectors
nsff= number of scan flip-flops
-
8/12/2019 VLSI Testing - DFT and Scan
13/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 13Copyright 2001, Agrawal & Bushnell 13
Testing Scan RegisterTesting Scan Register Scan register must be tested prior to application
of scan test sequences. A shift sequence 00110011 . . . of length nsff+ 4 in
scan mode (TC = 0) produces 00, 01, 11 and 10
transitions in all fl ip-flops and observes the result
at SCANOUT output. Total scan test length:
(ncomb + 2) nsff+ ncomb + 4 clock periods.
Example: 2,000 scan flip-flops, 500 comb. vectors,total scan test length ~ 106 clocks.
Multiple scan registers reduce test length.
-
8/12/2019 VLSI Testing - DFT and Scan
14/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 14Copyright 2001, Agrawal & Bushnell 14
Multiple Scan RegistersMultiple Scan Registers Scan flip-flops can be distributed among
any number of shift registers, each having
a separate scaninand scanoutpin. Test sequence length is determined by the
longest scan shift register.
Just one test control(TC) pin is essential.
SFFSFF
SFF
Combinational
logic
PI/SCANIN PO/
SCANOUTMU
X
CK
TC
-
8/12/2019 VLSI Testing - DFT and Scan
15/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 15Copyright 2001, Agrawal & Bushnell 15
Scan OverheadsScan Overheads IO pins: One pin necessary.
Area overhead:
Gate overhead = [4 nsff/(ng+10nff)] x 100%where ng = comb. gates; nff = flip-flopsExample ng = 100k gates, nff= 2k flip-flopsoverhead = 6.7%.
More accurate estimate must consider scan wiringand layout area.
Performance overhead:
Multiplexer delay added in combinational path;approx. two gate-delays.
Flip-flop output loading due to one additionalfanout; approx. 5 - 6%.
-
8/12/2019 VLSI Testing - DFT and Scan
16/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 16Copyright 2001, Agrawal & Bushnell 16
Hierarchical ScanHierarchical Scan Scan flip-flops are chained within
subnetworks before chaining subnetworks.
Advantages: Automatic scan insertion in netlist
Circuit hierarchy preserved helps in
debugging and design changes
Disadvantage: Non-optimum chip layout.
SFF1
SFF2 SFF3
SFF4SFF3SFF1
SFF2SFF4
Scanin Scanout
Scanin
Scanout
Hierarchical netlist Flat layout
-
8/12/2019 VLSI Testing - DFT and Scan
17/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 17Copyright 2001, Agrawal & Bushnell 17
Optimum Scan LayoutOptimum Scan Layout
IO
pad
Flip-
flop
cell
Interconnects
Routing
channels
SFF
cell
TC
SCANIN
SCAN
OUT
Y
XX
Y
Active areas: XY and XY
-
8/12/2019 VLSI Testing - DFT and Scan
18/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 18Copyright 2001, Agrawal & Bushnell 18
Scan Area OverheadScan Area OverheadLinear dimensions of active area:
X = (C + S) / r
X = (C + S + S) / r
Y = Y + ry = Y + Y(1--) / TArea overhead
XY--XY
= -------------- x 100%XY
1--= [(1+s)(1+ -------) 1] x 100%
T
1--= (s + ------- ) x 100%
T
y = track dimension, wire
width+separationC = total comb. cell width
S = total non-scan FF cell
width
s = fractional FF cell area
= S/(C+S) = SFF cell width fractionalincrease
r = number of cell rows
or routing channels = routing fraction in activearea
T = cell height in track
dimension y
-
8/12/2019 VLSI Testing - DFT and Scan
19/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 19Copyright 2001, Agrawal & Bushnell 19
Example: Scan LayoutExample: Scan Layout 2,000-gate CMOS chip
Fractional area under flip-flop cells, s = 0.478
Scan flip-flop (SFF) cell width increase, = 0.25 Routing area fraction, = 0.471 Cell height in routing tracks, T = 10
Calculated overhead = 17.24%
Actual measured data:
Scan implementation Area overhead Normalized clock rate
______________________________________________________________________
None 0.0 1.00
Hierarchical 16.93% 0.87
Optimum layout 11.90% 0.91
-
8/12/2019 VLSI Testing - DFT and Scan
20/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 20Copyright 2001, Agrawal & Bushnell 20
ATPG Example: S5378ATPG Example: S5378
Original
2,781
179
0
0.0%
4,603
35/49
70.0%
70.9%
5,533 s
414
414
Full-scan
2,781
0
179
15.66%
4,603
214/228
99.1%
100.0%
5 s
585
105,662
Number of combinational gates
Number of non-scan flip-flops (10 gates each)
Number of scan flip-flops (14 gates each)
Gate overhead
Number of faults
PI/PO for ATPG
Fault coverage
Fault efficiency
CPU time on SUN Ultra II, 200MHz processor
Number of ATPG vectors
Scan sequence length
-
8/12/2019 VLSI Testing - DFT and Scan
21/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 21
Partial-Scan DefinitionPartial-Scan Definition A subset of flip-flops is scanned.
Objectives: Minimize area overhead and scan
sequence length, yet achieve requiredfault coverage
Exclude selected flip-flops from scan: Improve performance
Allow limited scan design rule violations
Allow automation: In scan flip-flop selection In test generation
Shorter scan sequences
-
8/12/2019 VLSI Testing - DFT and Scan
22/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 22
Partial-Scan ArchitecturePartial-Scan Architecture
FF
FF
SFF
SFF
Combinationalcircuit
PI PO
CK1
CK2 SCANOUT
SCANIN
TC
-
8/12/2019 VLSI Testing - DFT and Scan
23/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 23
A Partial-Scan MethodA Partial-Scan Method
Select a minimal set of flip-flops for scan
to eliminate all cycles.
Alternatively, to keep the overhead low
only long cycles may be eliminated.
In some circuits with a large number ofself-loops, all cycles other than self-loops
may be eliminated.
-
8/12/2019 VLSI Testing - DFT and Scan
24/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 24
The MFVS ProblemThe MFVS Problem For a directed graph find a set of vertices with
smallest cardinality such that the deletion of thisvertex-set makes the graph acyclic.
The minimum feedback vertex set(MFVS) problemis NP-complete; practical solutions use heuristics.
A secondary objective of minimizing the depth ofacyclic graph is useful.
1 2
3
4 5 6
L=3
1 2
3
4 5 6L=2
L=1
s-graphA 6-flip-flop circuit
-
8/12/2019 VLSI Testing - DFT and Scan
25/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 25
Partial Scan ExamplePartial Scan Example Circuit: TLC
355 gates
21 flip-flops
Scan Max. cycle Depth* ATPG Fault sim. Fault ATPG Test seq.
flip-flops length CPU s CPU s cov. vectors length
0 4 14 1,247 61 89.01% 805 805
4 2 10 157 11 95.90% 247 1,249
9 1 5 32 4 99.20% 136 1,382
10 1 3 13 4 100.00% 112 1,256
21 0 0 2 2 100.00% 52 1,190
* Cyclic paths ignored
-
8/12/2019 VLSI Testing - DFT and Scan
26/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 26
Test Length StatisticsTest Length Statistics Circuit: TLC
200
100
00 50 100 150 200 250
Numb
er
offau
lts
200
100
00 5 10 15 20 25
Numbe
r
offaults
200
100
00 5 10 15 20 25
Number
offaults
Without scan
9 scan flip-flops
10 scan flip-flops
Test
length
Test
length
Test
length
-
8/12/2019 VLSI Testing - DFT and Scan
27/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 27
Partial vs. Full Scan:
S5378
Partial vs. Full Scan:
S5378
Original
2,781
179
0
0.0%4,603
35/49
70.0%
70.9%
5,533 s
414
414
Full-scan
2,781
0
179
15.66%4,603
214/228
99.1%
100.0%
5 s
585
105,662
Number of combinational gates
Number of non-scan flip-flops
(10 gates each)
Number of scan flip-flops
(14 gates each)
Gate overhead
Number of faults
PI/PO for ATPG
Fault coverage
Fault efficiency
CPU time on SUN Ultra II200MHz processor
Number of ATPG vectors
Scan sequence length
Partial-scan
2,781
149
30
2.63%
4,603
65/79
93.7%
99.5%
727 s
1,117
34,691
-
8/12/2019 VLSI Testing - DFT and Scan
28/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 28
Random-Access Scan (RAS)Random-Access Scan (RAS)
POPI
Combinational
logic
RAM
nff
bits
SCANOUTSCANIN
CK
TC
ADDRESS
ACK
Address scan
register
log2 nffbits
Address decoder
SEL
-
8/12/2019 VLSI Testing - DFT and Scan
29/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 29
RAS Flip-Flop (RAM Cell)RAS Flip-Flop (RAM Cell)
Scan flip-flop
(SFF)
Q To comb.
logic
D
SD
From comb. logic
SCANIN
TCCK
SELSCANOUT
-
8/12/2019 VLSI Testing - DFT and Scan
30/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 30
RAS ApplicationsRAS Applications Logic test:
Reduced test length Reduced scan power
Delay test: Easy to generate single-input-change
(SIC) delay tests.
Advantage: RAS may be suitable for certainarchitecture, e.g., where memory is implemented
as a RAM block.
Disadvantages:
Not suitable for random logic architecture High overhead gates added to SFF, address
decoder, address register, extra pins and routing
-
8/12/2019 VLSI Testing - DFT and Scan
31/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 31Copyright 2001, Agrawal & Bushnell 31
Boundary Scan (BS)
IEEE 1149.1 Standard
Boundary Scan (BS)
IEEE 1149.1 Standard
Developed for testing chips on a printed circuitboard (PCB).
A chip with BS can be accessed for test from theedge connector of PCB.
BS hardware added to chip: Test Access port (TAP) added
Four test pins
A test controller FSM
A scan flip-flop added to each I/O pin. Standard is also known as JTAG (Joint Test
Action Group) standard.
-
8/12/2019 VLSI Testing - DFT and Scan
32/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 32Copyright 2001, Agrawal & Bushnell 32
Boundary Scan Test LogicBoundary Scan Test Logic
-
8/12/2019 VLSI Testing - DFT and Scan
33/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 33Copyright 2001, Agrawal & Bushnell 33
SummarySummary
Scan is the most popular DFT technique:
Rule-based design
Automated DFT hardware insertion
Combinational ATPG
Advantages:
Design automation
High fault coverage; helpful in diagnosis
Hierarchical scan-testable modules are easily combined
into large scan-testable systems Moderate area (~10%) and speed (~5%) overheads
Disadvantages:
Large test data volume and long test time
Basically a slow speed (DC) test
Variations of scan:
Partial scan
Random access scan (RAS)
Boundary scan (BS)
-
8/12/2019 VLSI Testing - DFT and Scan
34/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 34Copyright 2001, Agrawal & Bushnell 34
Problems to SolveProblems to Solve What is the main advantage of scan method?
Given that the critical path delay of a circuit is 800ps andthe scan multiplexer adds a delay of 200ps, determine theperformance penalty of scan as percentage reduction in theclock frequency. Assume 20% margin for the clock period
and no delay due to the extra fanout of flip-flop outputs.
How will you reduce the test time of a scan circuit by afactor of 10?
-
8/12/2019 VLSI Testing - DFT and Scan
35/35
Copyright 2001, Agrawal & Bushnell DFT and Scan 35Copyright 2001, Agrawal & Bushnell 35
SolutionsSolutions What is the main advantage of scan method?
Only combinational ATPG (with lower complexity) is used.
Given that the critical path delay of a circuit is 800ps and the scanmultiplexer adds a delay of 200ps, determine the performance penaltyof scan as percentage reduction in the clock frequency. Assume 20%margin for the clock period and no delay due to the extra fanout of flip-flop outputs.
Clock period of pre-scan circuit = 800+160 = 960ps
Clock period for scan circuit = 800+200+200 = 1200ps
Clock frequency reduction = 100(1200-960)/1200 = 20%
How will you reduce the test time of a scan circuit by a factor of 10?
Form 10 scan registers, each having 1/10th the length of a single scanregister.