Types of Testing

VLSI testing can be classified into four types depending

upon the specific purpose it accomplishes

• Characterization
• Production
• Burn-in
• Incoming Inspection
CHARACTERIZATION
- Also known as design debug or verification testing, this
form of testing is performed on a new design before it is
sent to production. The purpose is to verify that the
design is correct and the device will meet all
specifications.
- Use of specialized tools such as scanning electron
microscopes (SEM) and electron beam testers, and
techniques such as artificial intelligence (AI) and expert
systems, can be effective.
- A characterization test determines the exact limits of
device operating values.
- Yield is the fraction (or percentage) of acceptable parts
among all fabricated parts.
- Characterization testing may be done on the chips
rejected during the production test or in the field.
This essentially means repetitively applying functional tests and
measuring various DC or AC parameters, as we vary different
variables such as (the supply voltage).
“Shmoo Plot”
PRODUCTION
- Every fabricated chip is subjected to production tests
which are less comprehensive than characterization
tests yet they must enforce the quality requirements by
determining whether the device meet specifications.
- The vectors may not cover all possible functions and
data patterns but must have a high coverage of modeled
faults.
- Production tests are typically short but verify all relevant
specifications of the device. It is an outgoing inspection
test of each device, and is not repetitive.
- We test whether some device-under-test (DUT)
parameters are consistent with the device specifications
under normal operating conditions. We test either at the
speed required by the application of the device or at the
speed guaranteed by the supplier.
BURN-IN
- All devices that pass production tests are not identical.
When put to actual use, some will fail very quickly while
others will function for a long time.
- Burn-in ensures reliability of tested devices by testing,
either continuously or periodically, over a long period of
time, and by causing the bad devices to actually fail.
Correlation studies show that the occurrence of
potential failures can be accelerated at elevated
temperatures.
- Infant mortality failures and Freak failures
- During burn-in, we subject the chips to a combination of
production tests, high temperature, and over-voltage
power supply. In practice, a manufacturer must balance
economic considerations against the device reliability.
In any case, the elimination of infant mortality failures
is considered essential.
INCOMING INSPECTION
- System manufacturers perform incoming inspection on
the purchased devices before integrating them into the
system.
- Testing can be either similar to production testing, or
more comprehensive than production testing, or even
tuned to the specific systems application. Also, the
incoming inspection may be done for a random sample
with the sample size depending on the device quality
and the system requirement.
- The most important purpose of this testing is to avoid
placing a defective device in a system assembly where
the cost of diagnosis may far exceed the cost of
incoming inspection.
Types of Tests

• The first test, known as wafer sort or probe, differentiates

potentially good devices from defective ones. After this, the
wafer is scribed and cut, and the potentially good devices are
packaged. Also, during wafer sort, a test site characterization is
performed. Specially designed tests are applied to certain test
sites containing specific test patterns. These are designed to
characterize the processing technology through measurement
of parameters such as gate threshold, polysilicon field
threshold, bypass, metal field threshold, poly and metal sheet
resistances, contact resistance, etc.

• In general, each chip is subjected to two types of tests:

• 1. Parametric Tests
• 2. Functional Tests
Parametric Tests
- DC parametric tests include shorts test, opens test,
maximum current test, leakage test, output drive current
test, and threshold levels test. AC parametric tests
include propagation delay test, setup and hold test,
functional speed test, access time test, refresh and
pause time test, and rise and fall time test.
Functional Tests
- These consist of the input vectors and the corresponding
responses. They check for proper operation of a verified
design by testing the internal chip nodes. Functional
tests cover a very high percentage of modeled (e.g.,
stuck type) faults in logic circuits and their generation is
the main topic of this tutorial. Often, functional vectors
are understood as verification vectors, which are used to
verify whether the hardware actually matches its
specification.
- Functional tests may be applied at an elevated temperature to
guarantee specifications. For example, testing may be done at 85° C
to guarantee 70° operation. This is called guardbanding. Another
application is in speed binning to grade the chips according to
performance. This may be done by applying the tests at several
voltages and at varying timing conditions (e.g., clock frequency).
- VLSI chip testing, in many ways, resembles the testing of other
digital equipment such as circuit boards, but with differences. Circuit
boards consist of previously-tested components. A primary objective
of board testing is to check the printed wiring and the contacts
between wires and components. It is possible to perform a bare-board
testing of interconnections before the components are inserted. After
the components (such as chips) are in place, in-circuit testing is often
used to verify the performance of individual components, although
the bed-of-nails fixture is becoming obsolete. Finally, functional
testing determines whether or not individual components, possibly
designed with different technologies, function as a system and
produce the expected response
Test Specifications and Test Plan
The device specification document initiates the development
activity, and contains the following information:
• Functional Characteristics – Algorithms to be implemented,
I/O signal characteristics (timing waveforms, signal levels,
etc.), data and control signal behavior, clock rate.
• Type of Device – Logic, microprocessor, memory, analog,
etc.
• Physical Characteristics – Package, pin assignments, etc.
• Technology – CMOS (or gate array), custom, standard cell,
etc.
• Environmental Characteristics – Operating temperature
range, supply voltage, humidity, etc.
• Reliability – Acceptance quality level (defective parts per
million), failure rate per 1,000 hours, noise characteristics, etc.
 Testers
The basic purpose of a tester is to drive the inputs and to
monitor the outputs of a device-under-test. Testers are
popularly known as ATE (automatic test equipment.) Fast-
changing VLSI technology has driven the development of
modern ATE. Selection of ATE for a VLSI device must
consider the specifications of the device. Major factors are
speed (clock rate of the device), timing (strobe) accuracy,
number of input/output pins, etc. Other considerations in
selecting a tester are cost, reliability, serviceability, ease of
programming, etc.
Test Programming
Once the device-under-test has been mounted in the
tester, three things are needed to conduct the test. These
are the test program, the digital test vectors, and the
analog test waveforms. Until recently, test programs were
written manually. However, the use of CAD tools is now
becoming widespread in this area.
As shown in the figure, the device specifications spawn
several activities. The results of these activities are
required for the test program. An automatic test
program generation system (often called TPG) requires
three types of inputs:

(1) Tester specification and the information on the types

of tests is obtained from the test plan.
(2) Physical data on the device (pin locations, wafer map,
etc.) are obtained from the layout.
(3) Timing information on signals and test vectors (inputs
and expected responses) are obtained from simulators.
Test Data Analysis
The test data obtained from the ATE serves three
purposes. First, it helps to accept or reject the device-
under-test. Second, it provides useful information about
the fabrication process. And third, it provides information
about design weaknesses.