We search for industry news, so you don't need to.
Ask the Experts  
Questions Index  ■  Submit a Question  ■  Experts Panel  ■  Join the Panel

August 20, 2013

How to Eliminate Via Pits

Is there a process or solder paste designed for filling vias that will not cure with a pit in the via material? The pit may require rework. We have tried several brands of paste and it runs out of the holes causing incomplete fill or in general the paste shrinks when cured. The hole sizes range from 0.005" to 0.025". 

D.F.

Experts Comments

You have two things working against you here. The first and most important is that almost all alloys shrink during solidification. For SnPb, for instance, that shrinkage is about 4% (per "Soldering in Electronics," RJ Klein Wassink, 1989). That alone will cause a big pit in the via. Then there is the fact that although a paste may be advertised as "90% metal content" that is by weight. By volume, it is probably just over 50%. So while you may fill the via with paste, you will wind up with a half-filled via once reflowed. There will be some internal void volume, and probably some un-filled volume at the ends where flux has escaped the joint.

The question is, why do you feel the pit is a problem? As far as I know there is no specification that states such a thing, and I can't imagine a reliability impact. If you are filling the vias to eliminate air passing through, then severe pits *might* occasionally result in leakage, however in the hole sizes you specify, I can't imagine it happening more than very rarely.
image
Fritz Byle
Process Engineer
Astronautics
Fritz's career in electronics manufacturing has included diverse engineering roles including PWB fabrication, thick film print & fire, SMT and wave/selective solder process engineering, and electronics materials development and marketing. Fritz's educational background is in mechanical engineering with an emphasis on materials science. Design of Experiments (DoE) techniques have been an area of independent study. Fritz has published over a dozen papers at various industry conferences.
Solder paste are not designed/developed as a via fill materials and should not be used as a via fill.
Mahendra Gandhi
SME - PWB Technologies
Northrop Grumman
Mahendra Gandhi has been working in interconnect industry since 1972.
Yes, there are processes that can be used. I assume you are trying to achieve flat filled vias in the center of BGA pads, using via-in-pad technology, and you are having solder voiding issues with the little dimple that sometimes remains in the middle of the pad. Via-in-pad should only be performed by a PWB fabricator with extensive experience, as it is not a technology for the willing but inexperienced.

Typically, microvias are filled either with copper (called plated-up vias) or with a hard conductive epoxy such as Peters PP2795. Read all about via fill in these two articles.

http://www.all4-pcb.us/viafilling.htm
http://www2.dupont.com/MCM/en_US/assets/downloads/prodinfo/Plugging_Vias_wConductive_Ink.pdf

After via filling is done, a process called planarization is performed which is supposed to leave the top of the vias with a smooth, hard, and flat surface. Unfortunately it quite often leaves a small dimple due to outgassing that occurs during curing of the via-fill epoxy. After solder paste is printed on the pad and BGA is placed and reflowed, the dimple causes outgassing into the BGA sphere during reflow. I have seen this where it was so bad that the outgassing actually exploded the BGA solder ball, causing all kinds of uncontrolled solder splatters and missing solder balls under the BGA, clearly evident in the X-ray. At best, you will have a void in the finished BGA ball, at worst you will have a mess.  

I had this issue happen to me (more than once), and of course the PWB vendor blamed it on the reflow profile we were using. Stuck with the PWBs and a very tight schedule, I developed the following process to fix the pads with the dimples in the vias.

First, I cleaned the PWB with de-ionized water in the in-line cleaner. Then I cleaned the BGA areas with via-in-pads with medical-grade IPA.

Using a mini-micro stencil from www.mini-microstencil.com, I manually printed a very thin layer of Pb90Sn10 solder paste onto the BGA pads. The stencil was .003" thick. The PWB pad diameters were .020". I used a stencil aperture diameter of .015". I then reflowed the bare PWB to coat the via-in-pads with a very thin layer of Pb90 solder. This filled the dimples and provided a smooth and fairly flat pad.

Later, we printed paste and soldered using the normal process, nothing was changed. The standard Sn63Pb37 solder wetted perfectly to the high-temperature Pb90-coated pads. There were no blowouts and there was no voiding seen whatsoever. Upon microsectioning, we saw no lead-rich areas within the BGA balls, which indicated that the Pb90 solder never reached liquidus temperatures during the subsequent production reflow. The finished CCAs passed reliability testing with better results than the originals.

Warning! This process may not be allowed for CCAs that must be RoHS-compliant, but I believe there are certain exemptions made for Pb90 solder, such as those on PowerPC and other CCBGAs and CBGAs, for which this would qualify. You need to check with your customer and the current RoHS requirements.
Richard D. Stadem
Advanced Engineer/Scientist
General Dynamics
Richard D. Stadem is an advanced engineer/scientist for General Dynamics and is also a consulting engineer for other companies. He has 38 years of engineering experience having worked for Honeywell, ADC, Pemstar (now Benchmark), Analog Technologies, and General Dynamics.
Submit a Comment

Your Name


Your Email


Company Name


Country


Comments




Authentication

Please type the number displayed into the box. If you receive an error, you may need to refresh the page and resubmit the information.