Is this a gold brazed plate on CPU?

[NobleMetalsRecovery]

Can anyone help me figure from past experience if this type of CPU has gold in the solder that holds the lid onto the ceramic CPU? If you look at the pictures you may be able to see that the solder is gold plated on the outside of the CPU where it contacts the ceramic and metal interface. Also, the solder is gold plated inside the CPU, under the metal lid. The solder itself is a gray color. Is it likely to contain gold?, silver? indium, etc.

 

[samuel-a]

AuSn brazing is common for ceramic CPU's lids (and leads).

you can read more in this file:
Au/Sn Solder Alloy and Its Applications in Electronics Packaging

AuSn solder usually runs at 80% + gold content.

 

[NobleMetalsRecovery]

I read most of the Au/Sn article, and skimmed the rest of the article. Very good info, thanks. I question the braze on the CPU in question since there is a lot of solder. The Au/Sn preforms used to attach a lid can be a thin as 25 um. That's extremely thin. Any other opinions on this item?

 

[qst42know]

Are you certain it's not an epoxy?

Acid test?

 

[NobleMetalsRecovery]

I'll test it with Nitric Acid. Could be epoxy, it's really bonded together very strongly.

 

[qst42know]

Or a torch flame. If it burns it's epoxy. :mrgreen:

 

[NobleMetalsRecovery]

That sounds even better. I have a heat gun that gets hot enough to test it.

 

[NobleMetalsRecovery]

I hit it full blast with the heat gun . Nothing happened. No epoxy smell. So I scraped it with a knife. The gray color solder turned silver colored. Acid will be the next test.

 

[NobleMetalsRecovery]

Here's a pic of the ceramic CPU with the solder burnished with a knife blade.

 

[Barren Realms 007]

Or a torch flame. If it burns it's epoxy. :mrgreen:

That type of brazing is not epoxy and it is hard to get the chip hot enough to remove the lid. The fastest way to remove the lid it break the chip in half and roll the lid back and peel it off of the chip like the pictures shows.

 

[NobleMetalsRecovery]

Sorry, my issue was not how to get the lid off (but the way you described is best), but rather what is the material holding the lid to the ceramic? Does it contain, gold? silver? Indium?

 

[Barren Realms 007]

Sorry, my issue was not how to get the lid off (but the way you described is best), but rather what is the material holding the lid to the ceramic? Does it contain, gold? silver? Indium?

More than likely just gold solder. Run the lids by themselves they will wreck a solution and give you nightmares.

 

[NobleMetalsRecovery]

Just GOLD solder? at 80% gold, that sounds good to me. Yes, I agree, the iron in the lids causes big trouble. May be best to dissolve away the iron from the solder?

 

[dtectr]

Buy GSP's book - $35 & ALL your questions are answered.

OR - follow the links on lazersteve's signature line re: CPU's in Poor Man's AR

Lids in sulfuric cell, everything else in Poor Man's AR. Remove with Mapp.

 

[NobleMetalsRecovery]

Got GSP's book over a year ago. He does mention that the gold braze does not look like gold.
I'll just have to do the acid test. Was hoping that someone had done these before and could share some first hand insight. Thanks.

 

[goldsilverpro]

Hey, Steve,

Long time no hear from.

For the older CPU packages I worked with, Au/Sn solder was not used unless the lid was gold plated to start with. For non-gold lids, some sort of Ni based braze was used. If you think about this, it makes sense. I don't know about the more modern stuff, but I have a feeling it would be the same. File or scrape some off, dissolve it in AR, and test it with stannous chloride.

In photos 843 and 848, it looks like you removed the lid mounting ring along with the lid. In that case, what appears to be braze may be something like moly-manganese thick film.

Maybe lazersteve can chime in. He has a better grasp of the newer stuff than I do.

Chris

 

[dtectr]

NMR,
Good photos, too, BTW. Will you keep some records & keep us posted on what works best for you?

 

[NobleMetalsRecovery]

Yes, I'll let you know the results.

Hi Chris, it's good to be back. $1,500 gold drives even the recluses out of the woods.

These CPU's should be good even without gold braze. I have them in two different sizes. The smaller ones ( a little smaller than a 486 CPU) have gold wires under the lid, and the wires are about twice the length of the wires found on a 486 CPU. Only thing is, the legs are not gold plated.