solder fillet peeling

#25041

solder fillet peeling | 2 July, 2003

Not sure how else to describe this. We've received some boards (1/2" x 3.5" x .0325" board built on panels of 24) with numerous resistors showing separation of the solder from the metallization at the top of the joint.

There are 26 0805's on the board (about 1/2 are resistors, half caps) and all of the defects are on resistors. They are not all of the same value. This board is a purchased part so I can't assess the process.

It's probably something that's fairly common but I've never seen it, and I can't seem to find any images of it anywhere (Coombs, Klein Wassink, Prasad).

Can contamination of a component be such that it allows for what amounts to perfectly fine looking wetting but still not form an acceptable intermetallic? Could this be a no-lead issue with the component metallization compounded by insufficient heating?

Thanks....

#25053

solder fillet peeling | 3 July, 2003

So, there is a crack or gap between: * Solder and component termination. * Component termination and component body.

Correct? Describe: * Coarseness / smoothness of the metal on the termination underneither the crack. * Solder left on the termination.

That this occures in a wide variety of components, possibly from different suppliers; indicates a mechanical flexing of the board caused the problem. It may have been caused by poor handling.

Problems with this theory are: * Panel scoring or whatever used to aid in seperating the panel should absorb the stress from handling. * Capacitors were not affected.

Q1: Can contamination of a component be such that it allows for what amounts to perfectly fine looking wetting but still not form an acceptable intermetallic? A1: It's possible, but as you say, unlikely. Q2: Could this be a no-lead issue with the component metallization compounded by insufficient heating? A2: No. That you have a smooth, well wetted connection indicates the metals alloyed and heated well.

#25059

solder fillet peeling | 3 July, 2003

So, there is a crack or gap between: * Solder and component termination.

Bingo

Correct? Describe: * Coarseness / smoothness of the metal on the termination underneither the crack. * Solder left on the termination.

Unfortunately at 45X I can't really tell. It really looks like the separation is taking place right at the interface between the termination and the fillet. Metallurgically speaking (and this is a language I speak very poorly, it looks light it could be right on the intermetallic or on the termination side of it.

My hunch is that we have bad plating on the terminations, and that secondary processing (this is a daughter module that gets machine pinned after reflow then waved soldered onto the board) or depaneling just provides the stress to open them up.

What "bad plating" means is where I'm scratching my head.

I'm going to flex some boards to failure and see if I can get a better view.

#25060

solder fillet peeling | 3 July, 2003

I snapped a board in half at a cracked solder joint, and the component pulled free of the solder and still has a very thin layer of metal on the termination end.

Based on the construction spec.(solder over nickel over palladium), it looks like it parted at the palladium/nickel interface. Whether it parted because of excessive mechanical stress or a weak bond, I have no idea.

Should the bond there be stronger than the solder? We certainly aren't seeing the same problems with caps, although we have just started looking closely at the cap bodies for cracking there.

#25066

solder fillet peeling | 4 July, 2003

If you think you have an intermetallic, soldering as probably done correctly. In that case, expect the solder to fail in the solder, just above the intermetallic. This the weakest portion of the solder, because the tin has be depleted to form the intermetallic. Remainting tin will be the weakest portion of the connection. Lead has no strength. It's there for it's low cost, good looks, and to lower the melting point.

You've told us nothing that doesn't reinforce the mechanicial stress idea. It's possible that because or orientation, fabrication, etc that the capacitors are more compliant, in this ace, than the resistors.

Happy forth. Got two 24s of PBR for 15 total plus a 2 rebate. Cheaper than Coke.

#25067

solder fillet peeling | 4 July, 2003

You should see a coarse silverish / grayish coat on the termination that indicates that it soldered. We have to opinion [don't know] about the strength of solder connection versus the strength of component terminations.

#25085

solder fillet peeling | 7 July, 2003

It's definately soldering because the fillet is pulling the solder plating (I believe, from what I can see under available magnification) off of the termination, and some of the nickel plating as well.

#25086

solder fillet peeling | 7 July, 2003

If it's soldering, you have good plating.

MA/NY DDave

#25093

solder fillet peeling | 7 July, 2003

After reading the dialog I think you should send this to a lab that can do cross sections a little etching and raise the magnification.

YiEngr, MA/NY DDave

This message was posted via the Electronics Forum @

MA/NY DDave

#25107

solder fillet peeling | 9 July, 2003

Saw a Cross Section picture and thought of you

You noted above that you couldn't find any pictures or descriptions of your problem in a few well known sources.

It probably is not what you are experiencing yet a soldering defect called land lifting or fillet lifting due to CTE mismatch has a similar appearance to what you are describing. Again probably not your situation by the sound of things, YET who knows.

YiEngr, MA/NY DDave

This message was posted via the Electronics Forum @