In response to these kind of questions, it ultimately depends on what
the customer wants, or the assembly house wants. Some of the things you cite
(all comps pin 1 same orientation, no caps on bottom) are done because
"that's the way we have always done it" mind set.
True, all comps in the same direction helps with inspection, but if the board
functionality suffers - then it doesn't matter how easy it is to inspect.
All we can do as PWB design professionals is point out to the customer during
design, what these trade-offs are let them decide the fate of the board.
After all, they are the ones footing the bill.
Putting caps on the bottom side (1206 pkgs and smaller) has not, to my knowledge,
been a problem from an assembly standpoint - there were issues a while ago that
the caps were known to get stressed/cracks while being subject to solder wave ops-
that appeared to manifest itself in later failures in the field.
Larger caps/tantalums (293a,b,c,d,e pkgs) should not be placed through solderwave ops
due to their size/height and the possibility of cracking.
And if these customers foresee their boards/products going into production in high
numbers then perhaps it DOES make sense that the inspection concerns outweigh the
extra effort/delay in getting the board design to artwork - as long as functionality
of the board is not compromised.
In a perfect world, the PWB designers would know so much about the details of
what the customers' board does - the electrical engineering - that we would be
"empowered" to make the decisions and be "fully trusted" to do what we know best
during the layout stages, instead of having to cater to "obsolete" design practices.
We have been routinely putting caps and resistors on the bottom without any problems.
I don't understand the reason for route and via keepouts under SMD parts other than if
it is a metal case (oscillators). We have via keepouts in the center of the 1206 so that
vias aren't placed there, but surely you can put traces through.
The glue dot is usually provided through a placement x.y file that we extract from the design.
Even though our 0,0 of the component is still at pin 1 in our lib, we have added a
text marker to the center of the body on a user layer and wrote a script that extracts that.
Sorry for the long message, but you've hit a hot button, because I've been dealing with
these sort of "design practices" issues for a while now and iam tired of trying to
"convince" engineers and factories that we are up-to-date on these trade-offs and we KNOW
which ones matter and which are merely "preferences".
Sometimes it feels like we are given a car to drive, but with handcuffs on.
Tom Kavendek
Lucent Technologies
Advanced Video Technologies Dept.
Murray Hill, N.J.
[log in to unmask]
|
