Phil,

This question of delamination, was it on mulit-layer or pth? I had problems with 
undercured laminate in the 1970's and 80's which led us to bake all incoming 
materials as a precaution. In one case, the under cure was so bad the sheets of 
laminate came out of the oven with a bubble in the middle. Now our laminate 
suppliers are asked to post bake all material and we have not had a problem for 
years. If we did, I know the first thing to suspect.

My understanding is that if there is any partially polymerised resin in the 
laminate which has not fully cross-linked then, at solder flow temperatures, it 
will form shorter chains and leave some spare water molecules free to exert 
internal pressure as it turns to steam. This can cause localised delamination in 
areas well away from any holes where perhaps moisture can escape more easily.

The same can occur in multi-layers if the hot cycle is not long enough to 
complete the cross-linking. From my experience, moisture absorption will not 
cause a problem with delamination if the pcb is made with good laminate and 
processed correctly.

Paul Gould
EMail [log in to unmask]

In your message dated Friday 27, September 1996 you wrote :
> The extraction of volatiles from printed boards produces some intereseting
> stories. 
> 
>  I have experienced the return of most of the weight loss to samples within a
> few hours after baking; however, it seemed to arrange itself differently.  In
> a test where delamination was the prime consideration, the samples were baked
> and then let set for 2 days after baking at 70%RH to recover.  The baked and
> reabsorbed samples were considerably more resistant to delam. than samples
> which not previously baked.  Several theories were proposed, such as; it was
> not the same volatile, and there was not a large agglomeration of water drops
> as might have been in the unbaked samples, and that the reabsorbed moisture
> was more widely dispersed, or that it absorbed in the surface layers and came
> out during preheat.
> 
> Another oddity with polyimide.  When boards were baked at 110 deg C for 6
> hours they would have a certain weight loss.  If other samples from the same
> lot were baked at 150 deg C the weight loss was much larger; however after
> letting them set for a couple of days to readsorb water from the air, they
> returned to nearly the same weight, but on rebaking at the two temperatures,
> the weigh loss was again higher for the higher temperature samples.. Theory:
> water is adsorbed in the pores and it takes a greater energy input to break
> the water away from the pores than it does to merely spit water into vapor,
> which occurs quickly at 100 deg C. and you do not get it all out at the BP of
> water. The solution was to bake less time at a higher temperature which also
> by theory caused lest intermetallic growth for the time/temp used.
> 
> I had asked surface energy people to explain some of the above  phenomena,
> and they asked a few questions: (1) are the volatiles, absorbed, adsorbed or
> occluded. (2) are they free floating or are they bonded the the material by
> such a H bonding or are they actually a portion of the polymers. (3) What is
> the pore density and what is the space between the polymer chains.  They
> wanted to do WVT testing, I did not have the answers or the funds to test so
> the problem was resolved by the "good old production solution" bake a shorter
> time at a higher temperature, if it does not work, bake longer .  If any of
> you have the answers to any of the above let me know.
> 
> Thanks 
> Phil Hinton
> [log in to unmask] 
> 
>   
> 
> ***************************************************************************
> * TechNet mail list is provided as a service by IPC using SmartList v3.05 *
> ***************************************************************************
> * To unsubscribe from this list at any time, send a message to:           *
> * [log in to unmask] with <subject: unsubscribe> and no text.        *
> ***************************************************************************
> 
> 

-- 
Paul Gould
EMail [log in to unmask]

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************