Inge,
Some time ago I wrote a paper about trapped plating salts, typically nickel sulfamate solution that is trapped as the nickel barrier is plated. This happens when the termination metal glass frit has gaps between metal particles and the nickel bridges gaps trapping the plating solution that boils nicely during reflow. I still have a fun video showing parts hopping about on a hot plate. If Steve's web page is still available we could put it up for everyone's enjoyment.

John Maxwell
Sent from my iPad

On Feb 9, 2013, at 4:06 AM, Inge Hernefjord <[log in to unmask]> wrote:

> Bill,
> me again, while looking for references, it struck me that your resistor end
> terminations can be thickfilm made and not plated as I suggested.
> Thickfilm, whatever type Ag, AuAg, AuPd etc is sometimes porous as a
> sponge. Some are designed for attaching by adhesives and are less good for
> soldering, because the porosity sucks wet chemistry when you flux, and in
> the solder reflow all those volatiles want to come out, sometimes resulting
> in real outbursts. So, the quest ion is what terminations you got:  Au,
> Au/Ag, Ag, tinned, electroplated, electroless plated?
> 
> Inge
> 
> On 8 February 2013 17:55, Inge Hernefjord <[log in to unmask]> wrote:
> 
>> Bill, old & well documented. We may seemingly have to do with 'farting
>> chps' .Many  finish processes (in fact ALL) leave built-in gas e.g.
>> hydrogen. Normally the main part of voids in solder joints are from the
>> chemistry of the solder paste, and the addition of voids from termination
>> gas is negligable, but the opposite can happen if the component maker
>> doesn't have good control on the plating process, in fact, the outgassing
>> can be so violent that the slder looks like whipped cream, which is also
>> what we call it. A blowhole, even large ones, do not affect the solder
>> joint strength, but depending on what customer you have, it can be seen as
>> bad workmanship. What to do then? we simply performed single point
>> soldering of the bad joints or, if the bad guys were many, fluxed the board
>> and let it go through the soldering once more. You know, the gas generation
>> from the terminations can't fart a second time.
>> 
>> Inge
>> 
>> 
>> On 8 February 2013 03:56, <[log in to unmask]> wrote:
>> 
>>> Fellow TechNetters:  Recently having a problem with solder fillets on
>>> various SMT assemblies, we
>>> are observing an outgassing in the concave portion of the solder fillet
>>> of various resistor package
>>> sizes.  Specifically, 0603, 0805, and 2010 packages.  The outgassing is
>>> either a slight, hollow bump
>>> or a "blowhole" and always in the center of the fillet.  If we observe
>>> the condition, we switch the reel,
>>> and the problem generally goes away.
>>> 
>>> It is only happening with resistors, not any other SMT package.  The
>>> observations have now occurred
>>> in two separate facilities, with the reels of resistors having date codes
>>> from 2009 thru 2011.  We have not
>>> changed solder paste, using a no-clean paste from a major manufacturer.
>>> 
>>> Is anybody having issues with SMT resistors?  Can anyone shed any light
>>> on this absolutely random issue?
>>> 
>>> Your thoughts and comments are greatly appreciated.
>>> 
>>> THANK YOU,
>>> Bill
>>> 
>>> ______________________________________________________________________
>>> This email has been scanned by the Symantec Email Security.cloud service.
>>> For more information please contact helpdesk at x2960 or [log in to unmask]
>>> ______________________________________________________________________
> 
> 
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud service.
> For more information please contact helpdesk at x2960 or [log in to unmask] 
> ______________________________________________________________________

______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask] 
______________________________________________________________________