 Test
Re-Use: An Idea Whose Time Has Come
by Glenn Woppman
President and CEO
More and more test engineers and manufacturing
managers are realizing how far reaching the effects
of boundary-scan test re-use can be. And now, with recent
new developments, we can truly say that test re-use
is a concept whose time has come.
Another article
in this issue of Connect describes the new test re-use
solutions from ASSET and Agilent Technologies that dispel
many of the misconceptions of test re-use. For example,
you may have thought that test re-use holds most benefit
for high-volume manufacturers that have the luxury of
long manufacturing runs and high boundary-scan  test
content in their designs. Nothing could be farther from
the truth. Check out the new solutions
and find out why test re-use is now for everyone.
Why Test Re-Use?
Behind the move toward boundary-scan
test re-use are some significant economic benefits,
but the obvious benefits are just the tip of the iceberg.
Most would agree that a large amount of time and money
can be saved if tests developed during the design of
a product are migrated and re-used by manufacturing
and, later, field service. Rather than re-inventing
the wheel, so to speak, during each phase of a product’s
life cycle, the same boundary-scan tests developed during
the design phase can be re-used, eliminating the need
to develop them all over again. The cost savings from
this kind of a process are rather apparent.
But there are other benefits from test
re-use lurking below the surface, although they may
be harder to quantify.
For example, how do you economically
quantify time-to-market? That’s been a question
experts have grappled with for years. Even though it’s
tough to put a dollar figure on time-to-market, it’s
rather obvious that test re-use will only shorten the
time it takes to get a fully tested and fully supported
product to market. Removing from the product’s
critical path the time it would have taken manufacturing
and field service to develop tests significantly shortens
a product’s road to the marketplace.
The Momentum of Cost Savings
As cost savings from boundary-scan
test re-use begin to mount and the value of an accelerated
time-to-market is realized, manufacturers often start
thinking about increasing the amount of boundary-scan
test coverage in their products in order to keep this
momentum going. When this happens, even greater efficiencies
can be achieved.
For example, as boundary-scan test
coverage increases on a printed circuit board (PCB)
or assembly, test probes can be eliminated from the
in-circuit test (ICT) fixtures that are used to test
the board or assembly on an ICT system like the Agilent
3070. And the savings can be substantial for every probe
eliminated from an ICT fixture. Tellabs, for example,
estimated saving anywhere from $40,000 to $60,000 per
fixture by reducing the number of probes on their fixtures.
(For a customer success story on Tellabs’ experience
with ScanWorks and how it  helped
reduced the cost of ICT test fixtures, click
here.) Moreover, reducing the number of test probes
on a fixture cuts the time it takes to produce the fixture,
further shortening the product’s time-to-market.
The time saved by re-using tests could
also be applied in other ways to improve the quality
of the product. This would reduce the cost of warranty
returns as well as support services. Before tests from
development are re-used in manufacturing, they might
be fine tuned to achieve even greater test coverage.
This would also improve the quality of the manufactured
product and reduce follow-on costs.
Of course, closely monitoring the results
of boundary scan tests in manufacturing can reveal certain
trends that can also lead to cost reductions. For example,
if a design consistently fails a certain test more often
than any other, the order that the tests are applied
to that design might be altered so that the test which
generates the frequent failure is applied first. This
might eliminate time applying other tests if the design
is more likely to fail one test over others. The next
logical step would be for engineers to devote special
attention to identifying the cause of failure to rectify
the problem all together.
Achieving Higher Order Test
Efficiencies
Once an organization begins to grasp
the potential boundary-scan test re-use has for reducing
costs and improving efficiencies, it likely will want
to achieve higher levels of boundary-scan test coverage
in its designs as a means of ratcheting up its efficiencies
further. In a sense, a closed loop process develops.
That is, test re-use leads to greater efficiencies,
which lead to greater boundary-scan test coverage as
a way of increasing test re-use. The result is that
boundary-scan test re-use increases along with boundary-scan
test coverage.
After this self-perpetuating process
has run its course for some time, the organization will
begin to analyze how it can make the process itself
more efficient, reducing test costs further and shortening
time-to-market again. The first place where greater
efficiencies can be achieved would be during the initial
development of tests by the design department. If the
organization’s time-to-test, or the time it takes
to develop a boundary-scan test for a certain design,
can be shortened, the cost-of-test and time-to-market
will be slashed as well.
This is just what ASSET has done with
TopCAT (Topology and Cluster Analysis Technology) for
ScanWorks. To cut time-to-test, TopCAT begins by analyzing
the schematic of a PCB to identify all of the non-boundary
scan devices that are connected to boundary scan devices.
These non-boundary scan devices are candidates for boundary
scan cluster tests. Next, TopCAT automatically matches
the names of the non-boundary scan devices in the design's
netlist with device models archived on ASSET's web site
or stored in a model library within the user organization.
Once the device models have been retrieved by ScanWorks'
TopCAT technology, they are automatically included in
the interconnect test generation process. Lastly, TopCAT
optimizes the  configuration
of the device models in a test action for the highest
test coverage and to ensure the safety of the board.
(For a technical article in a previous issue of Connect
by ASSET’s chief technology officer on device
modeling, click here.)
Benchmarks have been performed with
ScanWorks’ TopCAT technology on a complex printed
circuit board (PCB) consisting of almost 9,000 nets,
over 40,000 solder joints, approximately 6,000 components
including more than 60 non-boundary-scan memory clusters
and over 40 non-boundary-scan logic clusters. The results
of the benchmarks showed that boundary-scan test development
time could be reduced to less than a day while achieving
very high test coverage. Previous to TopCAT, a suite
of boundary-scan tests for such a design could have
taken weeks to develop.
For more information on TopCAT, click
here.
The Time for Re-Use Is Now!
Test re-use points to a critical issue
for electronics manufacturers. That is, once you start
going down the boundary-scan road, you’ll have
many opportunities to reduce your overall test costs,
achieve higher operational efficiencies and cut the
time-to-market for your products. And that’s not
a bad proposition at all.
|
