Excellent input Mike. Targeted all my concerns.  But, I did mislead.  The
foil is copper, nickel, gold. The gold is an electroplated flash. We are
measuring between .25um and .35um  of gold over the nickel. So, even with
the thin flash, the thin solder makes the gold a concern, our goal is 15um
to 20um of bulk solder for the interface.  

We are even having issues with Type 4 in a 50 um stencil. We scoop some of
it out, on some pads.  But, maybe a 75 um stencil . . . But, then I do, as
you mentioned face variation from wetting and spreading (now boys careful
with the comments). 

Fortunately, we are printing and reflowing first on an OSP pad, 

I think I'll try a thicker stencil with a smaller aperture. If it helps,
I'll try to remember to post back to this thread. 

Guy
 

-----Original Message-----
From: Mike Fenner [mailto:[log in to unmask]] 
Sent: Wednesday, October 23, 2013 2:03 PM
To: 'TechNet E-Mail Forum'; 'Guy Ramsey'
Subject: RE: [TN] requirement for thin solder joint.

Guy
Type 3 powder size won't hack it. In practice Type 4 particle size is not
that much different from type 3 from some suppliers, but in theory you
should be able to print it.
T5 would enable you to fiddle around with stencil thickness and reductions
but will be relatively expensive, could limit your effective supplier base
and alloy choice and likely will put out lead times. 
Assuming you get an even print you will end up with the right volume of
solder for a 25um reflowed deposit but you might have some problems with
actual reflowed height according to how the surfaces wet and the spread you
get. I imagine you are doing this lead free so this will tend to wet in
place on the non gold surface but will spread on the gold side. Depending on
the thickness of the foil. You haven't dimensioned that so you are anywhere
from getting a solder joint, dissolving Au completely to having high
intermetallics and so on depending on everything including time over
liquidus (as short as possible) and cool down rate( as quick as possible). A
laser zap as mentioned seems attractive. You might need to go to a tin free
alloy. No tin and no lead alloys do exist, but perhaps not in T5 powder. 
I imagine your assembly does not consist of just the one solder joint so
what are you doing with all the others, are they regular alloy joints?
There are some alloy and process combinations that would likely work without
the above issues but would need to have a better understanding of your
application than you might be prepared to share on an open forum.

Regards 
 
Mike Fenner
BS&P
M: +44 [0] 7810 526 317
T: +44 [0] 1865 522 663

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Guy Ramsey
Sent: Wednesday, October 23, 2013 3:04 PM
To: [log in to unmask]
Subject: [TN] requirement for thin solder joint.

Hi all, 

We would like to solder a thin gold foil, with some magic on top, to a PWB.


The foil and pad are round, about 560 um in diameter. 

We want the solder connection to end up less than 25 um thick.  So, we are
trying to print with a 50um thick stencil, and 20% reduction, the stencil
aperture is 460 um diameter.  So, area ratio, or aspect ratio look great,
printing shouldn't be a problem.  

 

Print results were inconsistent.  

 

This morning I realize that many of solder paste spheres are over half the
thickness of the stencil.  And the rule of thumb that five spheres should
fit across the aperture for good paste release.   Well, what about the
thickness of the stencil versus the diameter of the spheres? 

 

Has anyone out there wrestled with this problem?  Maybe I should be using a
thicker stencil, and push the aspect ratio a bit more. 

 

Thoughts? 

 



______________________________________________________________________
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] 
______________________________________________________________________