Ed has made some good points about coupons and the misconceptions about
their significance. I agree that a good coupon does not necessarily mean a
good PCB because there are many other important parameters which can only be
checked on the PCB itself.
However, I consider they are very important because we do a whole raft of
tests which are not specific to the PCB design but to the process. These
include daily sections checking for side to side and through hole
distribution from our direct plate and pattern plate processes, and these
sections include test pieces which have been soldered and thermal shocked.
There are solderability tests before and after accelerated ageing, and a
severe test of thermal shock for 5 or 10 cycles to test integrity of inner
layer connections looking at change in resistance through a series of
interconnects. There are also controlled impedance traces to test
consistency across the batch, and a  contamination test coupon which has to
be put in a humidity chamber with constant voltage applied looking for
breakdown in resistance over 21 days.  These tests could not be carried out
on samples from the variety of PCB's being processed day to day because they
require a standard design for function and comparison.

Plating thickness must be checked on the PCB's because coupons on the edge
of the panel will have higher readings than coupons at the centre of the
panel and the important consideration is the max and min values which have
to be measured with a gauge non-destructively. X-Ray inspection of
registration, AOI of circuit patterns, final Electrical Test, and hole size
and dimensional checks are all carried out on the PCB itself to determine
quality, not on coupons.  I like to define the Quality Assurance laboratory
as checking the process quality with coupons sampled from each days
production, and production inspection operations checking the PCB quality of
each and every batch. This division of responsibility is important to keep
clear.

On the original question about placement of coupons, I see no reason for the
scrap biscuit panel border not to be used for positioning coupons which
could then be used by the customer for pre-production tests. However, I
would question the economics of having so much scrap in a biscuit panel and
try to persuade them to reduce their borders as much as possible. I have
seen some incredibly wasteful biscuit panels because the customer wanted to
have a standard panel regardless of the shape and size of the individual
pcb. Basically, the cost per panel square inch is pretty much the same for a
process whether it has a pcb on it or is scrap material so keep it tight as
possible. Big bare laminate borders are also bad news for plating
distribution over the panel.

Cheers
Paul Gould
Teknacron Circuits Ltd

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet 
################################################################
Please visit IPC web site (http://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################