Does solder change when corroded?

[Alex Eisenhut]

The TRS-80 Model 102 I'm trying to repair had a bit of green gunk on the pcb from battery leakage. I removed a chip and tried to remove the old solder on the pads of the PCB, except that the solder doesn't melt. Like, at all.

I checked other clean joints on the PCB and the solder melts normally.

I'm trying not to damage the pads so I don't want to get too aggressive. I put RMA flux, I tried fresh solder, I tried 700 degrees, I tried to make good mechanical contact. Nothing. The blobs of solder look the same after with the divot of the IC lead still there.

Does the battery juice modify the solder into another alloy, I don't know, solder hydroxide or something?

I'm going to scratch the solder off but I don't want to rip off the pads, you know?

Anyone ever experience hard to melt solder after a battery juice leak?

[helius]

Copper hydroxide and copper carbonate. Some have reported success in using crushed aspirin tablets (acetylsalicylic acid) as flux to help melt through them.

[Alex Eisenhut]

Thanks but I'm not down at the copper yet, just the solder blobs. I'll take another stab at it later, I'm going to scrape it away.

[IanB]

The TRS-80 Model 102 I'm trying to repair had a bit of green gunk on the pcb from battery leakage. I removed a chip and tried to remove the old solder on the pads of the PCB, except that the solder doesn't melt. Like, at all.

Seems strange.

Can you scratch the surface of the solder to make it shiny? And can you have any success with the shiny surface using flux and new solder?

By the way, 700°F is not hot—not hot at all. 850°F would be hot. Have you tried that?

[Alex Eisenhut]

I don't want to damage the board. I'll try higher temps, and scratching the old solder. I stupidly already bought replacement parts...

[IanB]

You won't damage the board unless you apply heat for too long. Lower temperatures can work against you here as you may be tempted to leave the iron pressing on the solder for several seconds trying to get it to melt. Either the solder will melt or it won't. Time won't make much difference (unless there is a big heat sink like a ground plane). So set the iron to a very hot temperature, flux the solder blob and apply the iron to it. If it melts in a second or so you are good to go and can wick the older solder away. If it doesn't melt in a second or two it is not going to. In that case you may need to add additional amounts of low melting point solder in an attempt to dissolve it into a lower melting alloy.

[IanB]

A thought occurs to me that depending on the technology, they may have soldered the board using a two pass technique. They could perhaps have soldered surface mount components first using a high melting alloy in a reflow oven, and then soldered through hole components on a second pass through a wave solder machine at lower temperatures so that the SMT components wouldn't fall off.

If that were the case, you may have encountered a high melting alloy.

[Alex Eisenhut]

Come to think of it, it looks like solder "turned" to lead, ie the tin probably got eaten away. Lead has a much higher melting point than solder... I vaguely recall soldering wires to electrodes for water electrolysis experiments as a kid, and the solder turned into lead and it looked like this.
Make sense?

[IanB]

Come to think of it, it looks like solder "turned" to lead, ie the tin probably got eaten away. Lead has a much higher melting point than solder... I vaguely recall soldering wires to electrodes for water electrolysis experiments as a kid, and the solder turned into lead and it looked like this.
Make sense?

It's possible, though I think the chemicals could only attack the surface, not the interior.

Also pure lead melts at 621.4°F so you could still melt it with an ordinary soldering iron.

[Alex Eisenhut]

I guess it really was that simple. I used a K tip on my Hakko, turned up the heat all the way and blobbed on fresh solder with extra RMA. The TRS-80 PCB is very robust. Once I cleaned off the burnt flux residue, the pads look fresh and shiny.
Now to repeat the process and check traces.

[Alex Eisenhut]

If the batteries are still in the holder while leaking juice, is it possible that the tiny potential difference along the path of the leak causes current to flow through the leak path and somehow causing the tin to electromigrate out of the solder because of the chemicals?
You know, over years or decades?
I find this weird and fascinating.

[coppercone2]

is molten solder eutectic an azeotrope or does it seperate?

[IanB]

is molten solder eutectic an azeotrope or does it seperate?

When it's molten it won't separate. However, when it's crystalline it might behave differently. A metallurgist or materials scientist might be able to say more.

[coppercone2]

I meant to ask if it wont seperate by vapor pressure.

[SeanB]

The tin will be chemically removed by the alkali electrolyte, leaving behind the lead part of the matrix, which will also oxidise to make lead oxide.  Thus the white film and the crusty look. You need to first scrub off the oxide with a small nylon brush and some vinegar to remove the remaining NiCd KOH electrolyte, and then wash with water, then wash with IPA again to get all the water off and remove the majority of the by products. Then apply some flux (either rosin in some alcohol, or the unbuffered asprin as a powder in some alcohol to make a paste) to the board before adding fresh solder to make the remainder more or less eutectic solder again. Then remove and you will have a fresh pad to solder to again.