IBM ceramic CPUs and their boards

[ChemGeek]

It seems that these ceramic processors belong to ebay and collectors.
One can easily get 2-4 fold Au value by selling them there.
Refiners who try to recover Au from these should ask themselves, if they are not attempting to duplicate Great Leap policies of commander Mao of China.
Just a lot of work to reduce value of your feedstock.
It is easy to question competence of each other etc, but ebay is a safe bet uniformly delivering better results, so why not to use it?
Of course there are many damaged no longer merchantable processors out there which need to be recycled. Other are best left as they are.

I would only ask kjavanb123 to think well what could go wrong, eg was there an excess of nitric left and not enough of Na2S2O5 used to precipitate Au, or maybe not enough of nitric/hydrochloric, possibility of cementation on other metals etc before final conclusion is reached.

 

[kjavanb123]

I would only ask kjavanb123 to think well what could go wrong, eg was there an excess of nitric left and not enough of Na2S2O5 used to precipitate Au, or maybe not enough of nitric/hydrochloric, possibility of cementation on other metals etc before final conclusion is reached.

Hi

I can't really point at anything out of norm in the process. I even tried to do another AR on crushed pieces and got only 0.1g

Even at the 0.1g per CPU seems very low with that gold cap.

Best regards
KJ

 

[ChemGeek]

Kevin,
OK, you got what you got. That is the way how these are often processed. No obvious error.
Perhaps there are members here who can say more about different batches of these CPUs produced in different plants.
Maybe some substandard version, less realiable and also of lower Au content was sold to Iran and to some other nations unfavored by US?

 

[kjavanb123]

I lean toward difference in manufacturing caused my results. Which proved once again selling to collector would be the wise thing to do.

I am processing the boards, depopulation is finished.

 

[Owltech]

I was told that the yield also depends on the manufacturing plant (Malaysia, Singapore or Mexico)

 

[anachronism]

I was told that the yield also depends on the manufacturing plant (Malaysia, Singapore or Mexico)

It does, but it doesn't vary by that much.

 

[johnny309]

I was told that the yield also depends on the manufacturing plant (Malaysia, Singapore or Mexico)

Not true.
Why? Because is a blueprint...is quality control...and is processors(with a lot of rejected items...about 30%)...so you got to stick to specification of original project ...no matter in which country.
They used the same assembly line and the same machines.

 

[anachronism]

I lean toward difference in manufacturing caused my results. Which proved once again selling to collector would be the wise thing to do.

I am processing the boards, depopulation is finished.

Sorry old friend but they are never that low.

 

[solar_plasma]

Do you have a picture of the processed cpu's? According to the thread beneath, the first batch was not processed to the end.

https://goldrefiningforum.com/phpBB3/viewtopic.php?f=37&t=23462&p=247224&hilit=cpu+tungsten#p247224

 

[kjavanb123]

Do you have a picture of the processed cpu's? According to the thread beneath, the first batch was not processed to the end.

https://goldrefiningforum.com/phpBB3/viewtopic.php?f=37&t=23462&p=247224&hilit=cpu+tungsten#p247224

No, after my second AR leaching them I threw them away.

 

[kjavanb123]

Hi

I finally completed processing boards that came with those IBM ceramic CPUs. Here is a breakdown of numbers.

Number of boards: 14 pieces
Total weight: 7,172 grams (15.8 lbs)

Total weight of boards
after depopulation: 2,780 grams (6.1 lbs)

Total weight of ICs: 380 grams (0.8 lb)

Total weight of pins
and their plastics: 3120 grams (6.8 lbs)

ICs
They were first pyrolized then incinerated and crushed to fine powder as in the following photo,


Those are then sifted using kitchen sifter, oversized were digested in HCL, then AR. The fines that pass through the sifter, processed using concentrated lye solution and heat to concentrate to much smaller volume then dissolved in AR.


For more information on how to use concentrated lye solution to dissolve much of silica, tin or aluminum please watch this video.
https://youtu.be/Z47ZEyD5zSE

Gold bond wires visible,


That was dissolved in AR and yielded 0.8-0.7 of gold brown powder, the atual yield will be known after final melting.

Pins:
Those pins and anyother gold plated components was processing using 7000 ml hot tap water, 20 grams of sodium cyanide, and 10 ml of hydrogen peroxide.
It was quick, and components were rinsed in hot water 3 times, and aluminum foils were dropped in combined rinsed and cyanide solution over night.

Pins and gold plated parts after cyanide leaching completed


And here is solution with aluminum foils after 24 hrs, filtered aluminum foils washed 3 times with hot water, HCL added, residue with some junk remained


Yield from pins roughly 0.1-0.15 grams of gold as it has not being melted.

Thanks and regards
KJ

 

[anachronism]

Kevin please confirm something for me that I may have read incorrectly.

Did you find 0.1 to 0.15g of gold from 3.16Kg of pins in their plastics?

 

[kjavanb123]

Kevin please confirm something for me that I may have read incorrectly.

Did you find 0.1 to 0.15g of gold from 3.16Kg of pins in their plastics?

Since recovered gold was small amount there might be off in my scale. I would know for sure when gold powder is melted.

Best
KJ

 

[kernels]

Kevin please confirm something for me that I may have read incorrectly.

Did you find 0.1 to 0.15g of gold from 3.16Kg of pins in their plastics?

That's the way I read it based on his picture of the stripped connectors. I'm not sure whether I think that is good or bad, but certainly very easy if you are able to acquire Sodium Cyanide!

 

[anachronism]

Kernels

It's not enough yield.

 

[niks neims]

Total weight: 7,172 grams (15.8 lbs)
yielded 0.8-0.7 of gold
Yield from pins roughly 0.1-0.15 grams of gold
KJ

That is some low numbers KJ :/, it comes out to only about 110-130 PPM Au, or about 30 % lower than I would expect from material like that... Not sure what caused it - but since your yield on the CPU's also was lower than expected - maybe there is a common answer there....

Or maybe gospel according to Jon is correct, and we're in for a rude awakening due to diminishing PCB yields

Did you process your batch for Pd, Ag & Cu also?

 

[kjavanb123]

Total weight: 7,172 grams (15.8 lbs)
yielded 0.8-0.7 of gold
Yield from pins roughly 0.1-0.15 grams of gold
KJ

That is some low numbers KJ :/, it comes out to only about 110-130 PPM Au, or about 30 % lower than I would expect from material like that... Not sure what caused it - but since your yield on the CPU's also was lower than expected - maybe there is a common answer there....

Or maybe gospel according to Jon is correct, and we're in for a rude awakening due to diminishing PCB yields

Did you process your batch for Pd, Ag & Cu also?

Niks

Yes that seems low since these are old boards and I had recovered 150 ppm Au from such boards using lead smelting.

No I haven't done silver/palladium recovery on them yet or copper.

They are going to be done soon. I am also collecting small socket PC boards for processing

 

[kjavanb123]

Kevin please confirm something for me that I may have read incorrectly.

Did you find 0.1 to 0.15g of gold from 3.16Kg of pins in their plastics?

That's the way I read it based on his picture of the stripped connectors. I'm not sure whether I think that is good or bad, but certainly very easy if you are able to acquire Sodium Cyanide!

Kernel

Yes that is how much gold powder was recovered from pins but my scale is not that accurate so I will take powders to jewler melt them and weigh them on his scale.

Cyanide is so fast and great for gold plated or even silver scraps.

 

[anachronism]

Cyanide is so fast and great for gold plated or even silver scraps.

It certainly is, however how do you check to make sure that you have no values left in the solution before calculating your yields?

 

[niks neims]

Niks

Yes that seems low since these are old boards and I had recovered 150 ppm Au from such boards using lead smelting.

No I haven't done silver/palladium recovery on them yet or copper.

They are going to be done soon. I am also collecting small socket PC boards for processing

Glad to hear it, still for large socket MB (& slot CPU MB, pre-2001) a number of ~170 PPM Au would be my guess, Ag & Pd yields there should be rather high, so that too needs to be accounted for, for full value; also Cu...

As I understand, lead smelting is your "industrial" way of processing MB, and HCL+AR is analytical, laboratory method? If so, it's a little bit strange that the latter yields lower than former - one would think it should be the other way around, it`s obvious that streamlined industrial process could leave some fringe values behind, analytical process on the other hand should be as precise as possible... Maybe go over your designed procedure once more, where could the losses come from? On a hunch, mechanical processes (like grinding, milling) always make me nervous when it comes to soft, malleable gold......
In any case I am eagerly looking forward your results, as always