On 06/14/2010 05:45 AM, Glenn Little WB4UIV wrote:
This is known as gold embrittlement.
The gold has to be removed before a good solder connection is made.
To remove the gold, tin the gold plated area.
Desolder the tinning.
Retin and remove the solder three times.
This should remove the gold which forms an amalgam with the solder.
After the gold has been removed, a reliable solder connection can be made.

This is a common failure mode of radio power amplifiers.

If you can undo the connection without adding heat, you will see a black
area between the gold plating and the solder.

I do not know why the manufacturers insist on gold plating leads that
are designed to be soldered.
Silver plating seems like a better solution.
In this case, it appears that pins were soldered that were not designed
to be soldered.

Glad that you found the problem.

Removing gold is the simple option, it just take time and effort to perform.

Gold and tin can under certain mixture relations from a gold-tin alloy which is brittle, this is the problem. When soldering, gold dissolves up into the tin blob very easily, that's why the above procedure work, and also why it can become a real problem. However, this is not an issue of the gold-tin relationship is sufficiently low on gold. When soldering BGA on gold-plated PCB, the amount of tin in the ball is given, but sufficiently thin gold plating is safe.

We had this problem in a time when the PCB maker didn't have proper control, but once they got that the issue disappeared. There is a huge difference between brittle and proper solder joints.

We still use gold on out board, and it works. We don't get any returns due to that failure mode. So, gold isn't that bad, but you need to be careful and aware. I have many old instruments (Tek, HP) that uses gold-plated boards among other things. None of them has failed due to that problem. Good that you localized that issue with the FTS1200 as I believe more people have that issue with them.

Cheers,
Magnus

_______________________________________________
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Reply via email to