I have to agree with Alain's answer below with one additional caveat. One of
he primary mechanisms for panel warpage is the different coefficients of
thermal expansion of the different raw materials within the board (fiber
glass, epoxy resin, nd copper, will all expand and contract at slightly
different rates). Add to that any unbalanced construction and there will
definetely be a problem. If however, a construction is largely well balanced
and some warpage is still for whatever reason noted a bake cycle may very
well help. The trick would be to apply light pressure to flatten (we use
steel weights), bring the temperature of the panels up past the Tg (glass
transition temp) hold at temp for several hours, and then allow for a very
slow cool down under weight back through the Tg. A cool down of less than 4
° F per min. should help in avoiding re-introducing the sort of stresses
which will lend to warpge.
This is a key component in the lamination process of multilayer boards.
Slow, gentle cool down.
Of course as Alain noted, some will likely debate this a well. Looking
foreward to other feedback.
Bob Dube
QA Mngr
ES&D
-----Original Message-----
From: Alain Savard <[log in to unmask]>
To: [log in to unmask] <[log in to unmask]>
Date: Thursday, December 02, 1999 8:06 AM
Subject: Re: [TN] PCB Warping and Laminate Stack-Up
>Hi Kathy,
>
>1) True
>2-1) True
>2-2) Unlikely... you can have a balanced lay-up without using the same
stuff
>all the way through... The center of the board should be almost a mirror
and
>the distribution materials should be almost the same on both side of the
>mirror.
>3) Most common causes I've observed for warpage are lay-ups and layer
>designs, uneven copper distribution in 'mirrored' layers can also
>contribute. An unbalanced design may cause some serious warpage.
>
>This answer should get the ball rolling. People will disagree.
>
>Alain "not a guru" Savard, B.Sc.
>Chemical Process Analyst
>CAE Electronics Ltd.
>e-mail: [log in to unmask]
>
>-----Original Message-----
>From: Kathy Palumbo
>
>Hey techies,
>
>I have three questions for all you PCB technology guru's out there.
>
>1) First of all, I was provided (in a PCB technology training seminar) the
>following information:
>Because of the way the laminate material is made (it follows the textile
>industry equipment), there is always more yarn (glass fibers) in the warp
>direction, then in the fill direction. Because of this you will have a
more
>stable board in one direction then the other. When running the smaller
size
>panels (12" X 36") a lay-up called a "cross-ply layup" of the pre-preg
>material would make the board more stable, and less likely to have warping
>issues.
>Can anyone tell me if this is correct or not?
>
>2) I was also informed of the following information:
> The warping of the boards will never go away, especially by running
>the boards through more thermal cycles. This will only make it worse. The
>Epoxy in the laminate material continues to cross link and with each
thermal
>cycle the epoxy cross links more and more. This means everytime you run
the
>board through an oven you lose some of its ductility.
> The board is in its relaxed state while in the warped state, and
>trying to flatten the board out during a heating process will cause even
>more problems, because as soon as the board is heated up again it will go
>back to its relaxed (or warped) state again.
> Usually the warping of the board is caused by the following
>problems:
> 1) Bad design -- uneven stack up of the layers
> 2) The PCB house did not use the same Pre-Preg (stage B) laminate
>material as the Cored (stage C) material.
>Can anyone tell me if this is correct or not?
>
>3) Can anyone tell me what the other causes for warping of a PCB are?
>
>As always, thanks in advance for your feedback!
>
>##############################################################
>TechNet Mail List 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://www.ipc.org/html/forum.htm) for
additional
>information.
>If you need assistance - contact Gayatri Sardeshpande at [log in to unmask] or
>847-509-9700 ext.5365
>##############################################################
>
##############################################################
TechNet Mail List 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://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Gayatri Sardeshpande at [log in to unmask] or
847-509-9700 ext.5365
##############################################################
|
