Hi Mark!

Here's another *.PDF from Kemet that is pretty good:

http://www.kemet.com/KEMET/web/homepage/kechome.nsf/vapubfiles/f2106/$file/f2106.pdf

I know it's a long filename, but type it in without any spaces...it talks about MLC capacitor flex cracking...what happens with depanelizing.

One thing I must say though, a third, a third, and a third is pretty generic when specifying scores. Yeah, that works a lot of the time, but it can burn you more times than you would like...score depth really depends on laminate lay-up, how the glass fibers are oriented to each other within each layer, etc.

Also, one needs to verify that what you asked for is actually done. I've ordered fabs from a shop that specified a score (the general 1/3, 1/3, and a 1/3) and found out after we populated the assemblies, the score was 1/3 on one side, but barely kissed the surface on the other. So there I was with an exacto-knife going back over the scores so I could singulate the boards without fracturing parts...not fun. Luckily, this wasn't a high volume board.

-Steve Gregory-



Hi Mark, this is almost the same problem that we had with an 1812 multilayer
ceramic chip capacitor.  For some guidance I give up the following URLs.  Gotta
like that John Maxwell fella.

http://www.avxcorp.com/docs/techinfo/cracks.pdf

http://www.avxcorp.com/docs/techinfo/paramimp.pdf

http://www.avxcorp.com/docs/techinfo/assembly.pdf

http://www.avxcorp.com/docs/techinfo/smzero.pdf

http://www.kemet.com/KEMET/web/homepage/kechome.nsf/vapubfiles/f2111/$file/f2111
.pdf



Helped us, hope it helps others.

Regards,

Alex Krstic
NovAtel Inc.
Quality Engineering Services


-----Original Message-----
From: Mark E. Schumacher [mailto:[log in to unmask]]
Sent: Monday, April 15, 2002 2:57 PM
To: [log in to unmask]
Subject: [TN] Depanelization - Cracked Chip Caps

TechNet,
Does anyone out there have any good depanelization references or
guidelines?
Has anyone ever encountered similar problems with PCB depanelization per
below?

Problem -
We have encountered 1210 size chip capacitors failing, located near PCB
score edge, after depanelization with CAB Maestro 4M (automatic Pizza
cutter). Sectioned parts show classic flex cracks. PCBs are prescored with
30 deg and 1/3 web. The PCB edge has irregular shapes, therefore some
sections of the score line are routed. In addition, on advice from others,
we have added additional routes around the sensitive 1210 parts which were
cracking. As a consequence of these routes, we have made things worse, we
are now cracking 0805 parts (located at the end of the route) which never
were damaged before. The depanelizing cutter wheel impacts the PCB each
time it moves across a route and makes contact with the PCB. These added
routes just increased the number of impacts the PCB gets during
depanelization.

Questions -
1) Should routes around parts that are close to the score line be added
along the score line? Would I be better off getting rid of all routes in
the score line to eliminate cutter wheel impacts during depanelization? We
have noticed that we have no failures for PCBs which do not have routes in
the score line, (sore line is continuous),
2) What clearance from chip capacitors to score line should be followed as
a guideline?
3) How should parts be oriented along a score line, parrallel or
perpendicular?


Mark