A thin board with no innerlayers will grow with moisture absorption, with
the degree depending on the material.  Thus you can get shrinkage by heating it
and driving off the mositure.  A thick multilayer made of typical
epoxy/fiberglass and copper voltage planes won't budge much with moisture so you
have to look elsewhere for sources of shrinkage.
     However, if a panel isn't 100% cured coming to assembly,  an above Tg
temperature exposure could cause a few hundred ppm shrinkage.  For practical
purposes the board could be coming to you sufficiently cured to be consider
fully cured, but with high temperature you could still raise the Tg a few
degrees more and impart a little further cure that results in noticeable panel
dimensional change.  It's also possible residual stresses from fabrication are
coming into play and causing the shrinkage.  I know this can happen if you bake
a board in an oven for a while - I don't know if the time/temperature exposure
of your assembly process is enough to make this a plausible explanation.
     If this is happening, you can learn to expect the change and live with it
(by adjusting your dimensions at assembly) or have the board house bake the
board above Tg and have it shrink in their house instead of yours.  They'd
probably have to go back and re-size their innerlayers to give you a board at
print dimensions that would hopefully now be stable through further heat
excursions.
          As far as I know specs are written around the dimensions (and
flatness) of the panel as delivered.  Hidden residual stresses or lack of total
cure that cause the panel to  shrink (or warp) in assembly aren't considered.
In some cases it might make sense to spec the board to be oversized as received
so that it ends up the right size when assembled.

Jonathan Whitcomb  IBM Endicott   607-757-1326   [log in to unmask]   fax
757-1156
Dept T49G bldg 257-2  1701 North St.  Endicott NY  13760     aisle H office X013
pager 2930

---------------------- Forwarded by Jonathan C Whitcomb/Endicott/IBM on 08/26/99
02:20 PM ---------------------------

Dave Hoover <[log in to unmask]> on 08/26/99 11:46:48 AM

Please respond to "TechNet E-Mail Forum." <[log in to unmask]>; Please respond to
      Dave Hoover <[log in to unmask]>

To:   [log in to unmask]
cc:
Subject:  Re: [TN] PCB material shrinkage





Tom,

Are these doublesided or multilayer PCB's????
Tell us alittle about the board.

Groovy
-----Original Message-----
From: Tom Oliver [mailto:[log in to unmask]]
Sent: Thursday, August 26, 1999 8:03 AM
To: [log in to unmask]
Subject: [TN] PCB material shrinkage


Hello
My first post and first question to the experts.
I am new to the electronic game most of my experience has been in mechanical
inspection and test.
My question: is there a rule of thumb on material shrinkage on board
material or a spec. Our boards are coming in with all the dimensions at
nominal but after wave they have shrunk down to the low limit and sometimes
below low limit. Most of our tolerances are +/- .003.
This is giving us headaches with interference problems when we put other
components on the boards.
Is it normal for the boards to shrink. Our arrays are approx. 4" x 12" with
25 separate boards on each array.
Thx's in advance
Tom Oliver
Eagle Comtronics
[log in to unmask]

##############################################################
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
##############################################################

##############################################################
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
##############################################################