I totally agree with you, Wayne, for more "normal" soldering configurations, ie, on circuit boards with pads.

But when the application is flat-packs on top of TO-X can transistors, voiding criteria gets thrown out the window.

Graham is simply trying to find out if X-ray or some other inspection method can be used to evaluate the SJs formed with this particular configuration.

It's quite possible that SJ's with large voids are perfectly acceptable from a reliability standpoint for this design, perhaps even desired!

I believe someone with the good X-ray will be able to get a good image of the SJ, but the machine would have to have the ability to view from an obtuse angle due to the shadowing and the cluttering of the internal flat-pack and transistor features (as well as the substrate circuitry) stacked above and below the SJ image. Having done a lot of component failure analysis work, I can attest that X-ray inspection of SJs under components is no simple task. I just did it on an assembly where the relays were sometimes failing. Trying to get a good "identifiable" image of the relay reed positions, the rockers, internal relay circuit, the relay solder joints, etc. was difficult, especially because they were .050" apart from each other in a matrix. 

But with the right X-ray machine, I believe Graham's configuration can certainly be done.



-----Original Message-----

From: TechNet [mailto:[log in to unmask]] On Behalf Of Wayne Showers

Sent: Thursday, August 30, 2018 1:46 PM

To: [log in to unmask]

Subject: Re: [TN] qualifying a heatsink attach process



In the process you have described 2-D xray is sufficient as you are looking for sufficient coverage and that you do not have large voids.  a solid connection is best, followed by small voids, then baby Swiss.  If you see large voids (you should not see this if you are using prefroms), the process is not stable.  You do need to make sure you profile to ensure that you do not have excessive heat draw and retention from the heatsink itself.