Page 69 - Building A Succesful Board-Test Strategy

Test Methods 55

In many companies, a rigidly departmentalized organizational structure
aggravates these cost escalations by hiding costs of insufficient test comprehen-
siveness at a particular manufacturing stage. Each department minimizes its own
costs, passing problems on to the next stage until they fall out at system test or,
worse, at warranty repair. The classic example is the increased cost of design-for-
testability, including engineering time, additional board components, and extra
testing for testability circuitry. Design activities cost more, assembly may cost more,
but test costs are much lower. Designers often contend that their extra work ben-
efits other departments to the detriment of their own. Adding inspection to pre-
screen traditional test introduces costs as well. Again, a department that merely
looks out for its own interests rather than considering overall costs will not adopt
the extra step,
Combating this attitude requires attacking it from two directions. Manageri-
ally, sharing any extra costs or cost reductions incurred at a particular stage among
all involved departmental budgets encourages cooperation. After all, the idea is
that total benefits will exceed total costs. (Otherwise, why bother?)
Addressing the cultural barriers between designers or manufacturing people
and test people is both more difficult and more important. Historically, design engi-
neers have regarded test responsibilities as beneath them. They do the "important"
work of creating the products, and someone else has to figure out how to make
them reliably. This cavalier "over-the-wall" attitude ("throw the design over the wall
and let manufacturing and test people deal with it") often begins at our most ven-
erable engineering educational institutions, where students learn how to design but
only vaguely understand that without cost-effective, reliable manufacturing and test
operations, the cleverest, most innovative product design cannot succeed in the
real world.
Fortunately, like the gradual acceptance of concurrent-engineering principles,
there is light at the end of the educational tunnel. People such as Ken Rose at Rens-
selaer Polytechnic Institute in Troy, New York, are actively encouraging their stu-
dents to recognize the practical-application aspects of their work. For many
managers of engineers already in industry, however, cultivating a "we're all in this
together" spirit of cooperation remains a challenge. Averting test people's histori-
cal "hands-off" reaction to inspection equipment requires similar diligence.

2.2 A Brief (Somewhat Apocryphal) History of Test

Early circuit boards consisted of a handful of discrete components distrib–
uted at low density to minimize heat dissipation. Testing consisted primarily of
examining the board visually and perhaps measuring a few connections using an
ohmmeter or other simple instrument. Final confirmation of board performance
occurred only after system assembly.
Development of the transistor and the integrated circuit in the late 1950s
precipitated the first great explosion of circuit complexity and component den-
sity, because these new forms produced much less heat than their vacuum-tube
predecessors. In fact, IBM's first transistorized computer, introduced in 1955,