We have boards that inadvertently ended up with hard gold finish. As a result we have 50 microinches of gold causing all kinds of headaches at assembly. We asked to have the extra gold stripped off but I was told it was not possible. Is there no way to dissolve the gold off? Otherwise we have to get another batch made and they're expensive. Mercury bath? Hot cyanide? Microtome? Have an immersion silver finish over the gold!!! Argghgghgh anything!!
getting rid of gold on pcb
Oct 12, 2007
51 Replies
50 microinches is a lot. But you could have them hot-dip solder coated and releveled, which would dissolve the gold and replace with solder. That might work, but then the gold usually has nickel underneath...
What's the problem with the gold?
John
Hi John, It's causing a lot of embrittlement as the gold dissolves in the solder. As it is a 100 mil thick board with 12 plane layers and 6 signal layers, the board sits in the heat soak stage for a long time, causing more gold to get dissolved. The boards barely pass visual inspection of the solder joints, and some 0402 ans 0201 parts just pop off easily. The few bare boards we have left we want to somehow get them to work. I'm trying to get the PCB fab to put immersion silver over the gold but I haven't heard back from them. Perhaps they died laughing. I like the solder idea, do you mean some sort of HASL ?? Or some sort of wave process? It's an unusual request I know.
What does one of those babies cost?
I was thinking of trying the regular HASL pass. I know that thin, like
10 uin, gold just dissolves in solder without much effect, and the solder seems to stick to whatever's underneath, nickel usually.John
I would have thought the same..... going through HASL should remove the gold. Maybe you need to do it a couple of times..... not sure what the fab plant will say about having their solder contaminated with gold though. But doesn't the molten gold float on top of the solder?
It should dissolve nicely. At high gold concentrations, the intermetallic is nasty brittle stuff, but diluted enough, the gold shouldn't matter. I have no idea how long it might take to dissolve 50 uin of gold.
John
Maybe I'll ask our assembly guys to put solder paste on the board and reflow it, then send the boards to the PCB shop to do a HASL? Is that the process with the hot-air knife and it'll blow away the excess solder? Oh well look, it's Friday night already. I'm off to pickle what few neurons I have left!
You're an idiot. The dissolution of the Gold is a BAD thing, dumbass.
It causes embrittlement, and no, after it gets above a certain amount, it does not go away!
You should get a clue, before you spew.
Brasso has been available in the US for at least 50 years...and is still available.
Maybe you'd like to start over, and make a little bit of sense next time.
John
I have no experience in this area so I'll just rattle off the wacky ideas:
1) Sandblasting (copper microspheres?) 2) Planar (as with lumbar) 3) Wet abrasive and lapping (mirror polishing) 4) Air blasting at 1063C (Au melting point) (Cu melts at 1083C and Ni at 1455C..but PCB will burn) 5) EDM engraving 6) 3:1 HCL:HNO3 etching 7) Chlorine gas etching or Chlorine in solutionD from BC
Gold will alloy slightly - around 2% if i remember correctly.
...and do not forget environmentally and people friendly Mercury as well as acidic Cyanides...
Wouldn't chlorine would be the choice toxin. If it gets loose, it's heavier than air and will find something to react with. It's far from being an inert gas and isn't found naturally in it's elemental form. I'm guessing the resulting chlorine compounds are going to be more friendly than mercury compounds or cyanide just sitting around not reacting with objects.
If chlorine etching is used on the gold..it should be checked out what temperature should be used. The colder the temperature...the less reaction to copper. D from BC
You could probably rub the gold off with a scotchbrite pad and household cleanser.
John
I'd try Brasso, don't know what the US equivalent is.
formatting link
Martin
The US equivalent is Brasso.
John
Yup.. Does sound like that'll work and perhaps worth a try..
Gold is a relatively soft metal. Using a pad that is just harder than the gold but softer than the copper (or plating on copper) will help reduce wear on the copper. I know scotchbrite scuffs copper which means something in the scotchbrite is harder than copper..
Household cleansers (ex Javex) contain chlorine compounds but I don't know if that will react with gold.. (I'm not a chemist!) However chlorine gas will. How fast..I dunno..
It's interesting, it takes only 1% chlorine to saturate water. Max solubility is at 9.6C.. Anymore and it gases. This can be a good thing too. I suspect chlorine has much more reactivity with gold at 9.6C temp than with copper.. So.. chilled water saturated with chlorine and etch the board is my guess... I have no idea if that'll work..but seems like a start.
Joke...Perhaps you could just plop the pcbs in a chlorinated public swimming pool for a few days.. (I dunno if that'll work..) D from BC
Will your stupidity ever cease?
I find it nearly unbelievable some of the horseshit you uneducated twits spew.
Join the Discussion
Have something to add? Share your thoughts — no account required.
Didn't find your answer?
Ask the community — no account required