what the hell is this on the solder joint?

[tron9000]

So got a simple one that came in yesterday:

A set of highly accurate scales from a semiconductors lab, measures down to 0.01g. It has a VFD 8-digit readout and the problem with it was the leading number after the decimal point was flicking between 0 and 8 and going a bit brighter when it does so, naturally we assumed that the pin on the driver chip that was controlling that segment was going a bit funny.

So we took out the driver board with the intention of just whipping out the IC and and swapping it with the one that was driving the other half of the display, and socketing them whilst we're at it. The IC's are COP470 - long since gone out of production we believe from searching the internet.

Dates on the IC's put it around 30 years old (1984-85).

We stuck it under the viewer first for a nosey and what we found seriously baffled us:




This we believe to be the offending pin (testing pending). It not tin whiskers, they are way too big. Its been used regularly in a clean room for the past 30 years - has anyone else seen anything like this?

[Srbel]

Tin whiskers?

http://www.tinwhisker.us/

[grumpydoc]

Tin whiskers?

http://www.tinwhisker.us/

Possibly but there's stuff on the board as well which is some way from the joint.

It looks like crystalized rosin.

[crispy_tofu]

It not tin whiskers, they are way too big.

Tin whiskers?

...

It looks like crystalized rosin.

Hmmm, is that flux residue along the joint? 

[Srbel]

I didn't see the rest of the text under the photographs...

[Ian.M]

The last time I saw something similar that wasn't tin whiskers, it was contamination from a leaky NiCd backup battery.  Collect a small sample of the whiskers and put on a watch glass or microscope slide with a drop of IPA.  If they dissolve, they are Rosin or similar.  If not, evaporate to near dryness and try a drop of water.  If they dissolve when heated they are some sort of tin or lead salt, probably due to ambient humidity + acids from flux residue.

[voltz]

Looks to me as if a flux bubble in the joint exploded. Hence the bits of flux around the joint on the pcb. Probably a defect in manufacture. If it was tin whiskers or some kind of fungus growth, it would be local only. Something went pop. How weird! Nice.

[tron9000]

The last time I saw something similar that wasn't tin whiskers, it was contamination from a leaky NiCd backup battery.  Collect a small sample of the whiskers and put on a watch glass or microscope slide with a drop of IPA.  If they dissolve, they are Rosin or similar.  If not, evaporate to near dryness and try a drop of water.  If they dissolve when heated they are some sort of tin or lead salt, probably due to ambient humidity + acids from flux residue.

Well can't do that now, the "whiskers" are no longer there cos teh joint got reflowed! oh well.

There might be some left on the board somewhere, will give it a go with the IPA.

Nope they're gone, but apparently they melted away when re-soldered, so possibly flux - very strange effect though - it is like a bubble burst, we believe its more like a growth...

It's been in a room where the environment is monitored & controlled very closely. So the humidity will be near ideal lab conditions.

[tron9000]

HA! searching for Tin whiskers: Tin Whiskers Brewing Co. - hmmm Short Circuit Stout anyone?

[GreyWoolfe]

HA! searching for Tin whiskers: Tin Whiskers Brewing Co. - hmmm Short Circuit Stout anyone?

I will take a pint.

[SeanB]

Gold plated pin with purple plague, from a gold lead intermetallic formed in the joint. That pin probably had too little solder from new from a flux bubble, and over the years the gold reacted to make that purple ball and the gold whiskers.

[tron9000]

Gold plated pin with purple plague, from a gold lead intermetallic formed in the joint. That pin probably had too little solder from new from a flux bubble, and over the years the gold reacted to make that purple ball and the gold whiskers.

So your saying the whiskers are made of gold or are gold coloured? Cos if the former then they wouldn't have melted when the joint was re-soldered.

[Ian.M]

Its possible.  Gold readily dissolves in Sn/Pb or Sn based solders if its thin enough.