Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


solution to design challenge (ubga, NSMD pads, heavy traces)

Views: 2407

#84659

solution to design challenge (ubga, NSMD pads, heavy traces) | 18 March, 2020

I am about to order a stencil and I can already see that I'm going to have issues. I'm hoping someone has a recipe for this scenario that we haven't tried yet.

The parts are 16 mil pitch bga's, the pads are NSMD (10 mil dia. with 12.6 mil mask openings) and there are 9 mil traces connected to the pads.

In the past with this same scenario, using a 4 mil nano-coated stencil with 1:1 apertures, my experience has been that the paste flows onto the traces and leaves me with insufficient solder in the connections.

My fear of increasing the aperture diameter to over 1:1 or going to a 5 mil stencil stems from the potential for bridging, particularly on the side of the pads where there are no traces.

Thoughts?

reply »

#84670

solution to design challenge (ubga, NSMD pads, heavy traces) | 19 March, 2020

We are trying to picture it here. Which side doesn't have traces connected to it? Which one does? An X-ray picture will be best. Gerber screenshot will be second best.

reply »

#84672

solution to design challenge (ubga, NSMD pads, heavy traces) | 19 March, 2020

Imagine a circle. Now imagine a line attached to that circle. A line almost as wide as the circle. Or look at a regular BGA pattern now imagine if the traces were almost as wide as the pads. I have not seen any quite as bad as he describes but I have seen some ugly ones. We have only done a few and do not have a good solution.

reply »

#84675

solution to design challenge (ubga, NSMD pads, heavy traces) | 19 March, 2020

Stephen, you are spot on with your description.

Here's a .jpg with paste layer, copper, and solder mask.

Attachments:

reply »

#84692

solution to design challenge (ubga, NSMD pads, heavy traces) | 23 March, 2020

Steve,

On 16mil pitch I would try 12mil diameter on 5mil stencil, unless the BGA ball size is something unexpectedly huge.

reply »

#84693

solution to design challenge (ubga, NSMD pads, heavy traces) | 24 March, 2020

My experience with our process is that those parameters would result in a big bridging problem. I'm working with a stencil designer at our fabricator on an asymmetrical aperture design that hopefully balances forces enough to keep the parts centered. If we remain in business, I will report back.

reply »

#84701

solution to design challenge (ubga, NSMD pads, heavy traces) | 27 March, 2020

Did you try 0.12 or 0.1mm nano coated stencil @ 10 mil aperture? A soak time type profile with long reflow time could help.

reply »

dual pot selective soldering

Manufacturing Software