Gold content of Pentium Pros

[glondor]

New people have an interest in old topics.

 

[Palladium]

that 5 years from one post to the next ???

Yep!

 

[goldsilverpro]

There was lots of good stuff posted in the early days. Except for stuff like auctions, nothing on here is timely. I like seeing the old stuff dredged up. Shows people are searching.

 

[Anonymous]

I like seeing the old stuff dredged up. Shows people are searching.

Thumbsup.

 

[ΩPhoenix]

Hello Skyline,
Making a list with all the CPUs would take too long.
But if someone makes a list with common CPUs, I think everyone would be interested.

Here is what I know:
486 CPUs has 3$ worth of gold
Pentiums Pro has 7$ worth of gold.

Thank you Noxx,

That is the figure that I came up with also for the 486's.
I didn't know that the Pentium Pro's were running $7 though.

Great information to know.

Steve

 

[glondor]

I believe those are very old numbers.

 

[Anonymous]

I think what Glondor means is, those are very,very old numbers,and Canadian dollars at that time.
THe date on that post was late 2007.The price of gold then was hovering in the neighborhood of $800(U.S.)/ozT

 

[goldenye]

can you give me this guys contact info? my email is [email protected]

There were two places that I found the information. I bookmark websites in the toolbar of Firefox, so I have hundreds to look through to find the same site, but it was a PDF file, I thought I had found on Scribd, but I can't find it again, yet. When I do, I will post it. When I first started learning about recovering gold, I scoured the internet for information about yields and availability of scrap. I thought I could easily make a fortune by finding a good cheap source, that is impossible. I ran across a guy on Alibaba that sells e-scrap, that's all he does. He is selling 1lb of ceramic CPU's for $107 usd per lb, he has over 800 lbs. All this guy is doing now is selling e-scrap, he no longer refines e-scrap, but did for a long time. He told me that when he was really actively recovering PM from E-Scrap, he contacted Intel and got the real numbers for the amount of gold contained in each CPU. According to him, someone at Intel gave him the information, it was .9 grams per CPU. However, he also told me that there is some variance dependent upon which variation of the CPU you are recovering from, and even where it was produced. I don't want to post his number in the forum, but I would be more than happy to give up his information to someone who would call, and ask about the ceramic CPU scrap and then ask about Pentium Pro's and how much gold content they have.

There were two places that I found the information. I bookmark websites in the toolbar of Firefox, so I have hundreds to look through to find the same site, but it was a PDF file, I thought I had found on Scribd, but I can't find it again, yet. When I do, I will post it. When I first started learning about recovering gold, I scoured the internet for information about yields and availability of scrap. I thought I could easily make a fortune by finding a good cheap source, that is impossible. I ran across a guy on Alibaba that sells e-scrap, that's all he does. He is selling 1lb of ceramic CPU's for $107 usd per lb, he has over 800 lbs. All this guy is doing now is selling e-scrap, he no longer refines e-scrap, but did for a long time. He told me that when he was really actively recovering PM from E-Scrap, he contacted Intel and got the real numbers for the amount of gold contained in each CPU. According to him, someone at Intel gave him the information, it was .9 grams per CPU. However, he also told me that there is some variance dependent upon which variation of the CPU you are recovering from, and even where it was produced. I don't want to post his number in the forum, but I would be more than happy to give up his information to someone who would call, and ask about the ceramic CPU scrap and then ask about Pentium Pro's and how much gold content they have.

In the meantime, let me clarify what I have stated already.

My understanding is that the Pentium Pro's used 22k gold, and that there is, according to Intel, .9 grams per each.

I processed these using a hammer first, and broke them up into small pieces. I just received a ball mill I bought off eBay, and hollow titanium balls just over an inch. I am planning on throwing the rest of what I have in ceramic into the ball mill, then processing the large ceramic parts separate from the fine powder (I just did 303 EProms and it makes a very fine dust). I am very curious now to see if I am able to recover anything more. I am planning on sifting the fine material through a fine metal screen to see if I find any gold wires, but I plan on processing everything regardless.

Because I was not expecting there to be other metals present that could be consumed by HCL before AR, I did not bother giving the substrate a HCL bath, I just washed everything with DI water, decanted and filtered, then mixed my AR. I mixed 1 part DI water to 4 parts HCL then added Sodium Nitrate to make Nitric Acid in solution. Over the next 30 min, while on the hotplate, I slowly added small quantities of Sodium Nitrate, waiting each time for the SN to dissolve before adding more, to each beaker I had substrate in and was processing. This takes a lot of time but I did not want to create too much excess nitric acid, so I took it slow and added SN slowly. When I added the last bit, and got no reaction I stopped adding SN. Then I let the solution evolve on the hotplate for the next hour until I could plainly see by eye, that the gold was stripped from the substrate. I don't like playing with the liquid form of Nitric Acid, I have it, but I hope to avoid it as much as possible, I like using SN because the reaction is slow, I can control it much better and the outgas, and it's much safer to handle.

After I took the AR solution off the heat, I waited until it cooled down, several hours, to room temperature. I am using 2000ml and 3000ml beakers, I decant from 2000ml into 3000ml with DI Ice. I had several going so that when I pulled one off the hotplate, I would put another one on. I have 5 gallon buckets, but it just seems to me that in a large bucket I would loose more AU. Plus I cannot put a 5 gallon bucket on a hotplate. I am using a Buchner vacuum filter with "nerd" filters, to decant the AR. I then add more DI water to the Buchner and vacuum filter that through just to make sure I get all the AU in solution.

I decanted as many of the 2000ml solutions into the 3000ml beaker as possible before I dropped the AU. I am using a combination of SMB and Ferrous Sulfate, I have been playing around with the ration, currently I am using 1/2 ration of FS to SMB. I don't like using Urea if I do not have to, so I first test the AR solution with SMB to see if I get NOx offgas, before I go dumping Urea into the solution. I am not really keen on creating Urea Nitrate, which I understand is the reaction result of mixing Urea with HCL, if I am wrong about this please correct me. I realize the amounts would be small, but still. I also think adding SMB until you stop getting a reaction may leave AU in solution, SMB will only react if there is free nitric acid available is my understanding, so when you no longer get the brown rusty NOx offgas I feel that there is still AU in solution that didn't drop because the SMB was burning Nitric Acid. I go by color. I add FS and SMB until the solution turns black and opaque. This seems to work the best for me and I realize other people may have better methods, if so please tell me.

Then I wait over night while the solution settles. I put the beakers on a small vibrating pad that I have, this might not do anything but I wanted to see if it made the gold precipitating out of solution solidify, fall to the bottom and clump together tighter. I am not sure it does anything at all but I was curious non the less.

Next day I decanted the excess AR through the Buchner Filter and pumped it through into the flask. I collected both the decanted fluid, and the black mud into two different beakers, washing everything down into the beaker that had the mud in it with a snifter and spray bottle, to ensure I was able to claim all the black mud. I set the AR solution aside to add more SMB to later, to see if I got anymore black mud.

I took the final beaker, with the concentrated black mud, and added DI water then put it on the hotplate to evaporate the remaining AR/Water.

After the mud became clumpy, and dry, I added more water, decanted into the Buchner filter, and pumped it throw, I took this water and added it to my SMB solution. Adding mass quantities of water to AU solutions helps, or even can precipitate gold. My thinking is that if there is anymore gold in the SMB solution, this method will help force it out. I did this three times and was left with a brownish, greyish clumpy dry mud. I then collected this and set it to the side, then used a coffee filter, and wiped the entire surface area of the inside of the beaker down really good.

I took all the filters I used, and saved, plus the coffee filter that I wiped the inside of the beaker with, and burned them in a cuppel until they were ash, then I added the brownish/greyish mud to the cuppel. I know what you are thinking, and what you are probably going to comment on, and I figure I need to explain the part of the process I didn't follow so that it is understood how I came up with the figures I have, and why I did not give my mud a HCL bath, three times, then wash three times, before melting it. I wanted to see how much total weight I had after AR, not in powdered form but in solid form because the weights are a little different. I am melting in a cuppel with a propane torch, not really sophisticated, but I don't want to deal with a torch and the long hoses, tripping over hoses while working with acid just didn't seem like a great idea to me. I have already ordered, and I am waiting for my melting furnace to arrive, so anything else I melt will be with that. It took about 45 min just to get things hot enough to start really melting the gold down. I used soda ash and borax in a 1/1 ration, adding it in very small amounts. I would move the torch and add borax and watch it grow on the gold, then apply the torch again, then added soda ash, then borax, soda ash, etc.

Once the gold was in a molten state, I kept the heat on it for 5 more min, then poured it off into an old bear keg I could the inside top off and turned into a water receptacle for cooling molten gold. I do have to admit that it's not the super shiny looking gold buttons I made before, I can tell there are metal impurities in the nugget, and tiny pieces of glass fused with the gold.

I added more FS and SMB to my SMB solution then waited overnight to see if anything more dropped. I did get a little bit of dust on the bottom of the beaker, so I know I didn't drop all the gold the first time. I didn't put a 1/1 ration of SMB/FS to gold expected because I was watching color instead, and I had no idea how much gold I would yield. I am also not sure of the ratio of FS to Gold expected to drop. So, in my excitement I just weighed my impure nugget, expecting to make up any loss in reprocessing it by adding the find layer of black sand that precipitated from my SMB solution after adding more precipitant.

I got less than .1 g from the second drop, so I was very wrong about that I think. I just processed my 22.3 nugget with HCL to burn off any base metals, and now it's bathing in AR, then I will decant, wait, filter, dry, wash with HCL then rinse X 3. Since I was planning on reprocessing to get as fine gold as possible, I thought that the HCL after being melted, may be able to force the other base metals into HCL.

I am dissolving the nugget in AR right now. I will post my new weights after it's finished and melting is completed. I am sure the amount will go down, but I don't think it will be by too much.

Anyway, I appreciate the observations, and will post the Pentium information when as soon as I find it again, may be a couple days, I am setting up my new hood tomorrow, and starting on the scrubber tomorrow. By the way, if anyone is looking for metal lab cabinets, I know a property management company near San Francisco that is selling what looks like brand new lab cabinets for $100 each. I already purchased the one hood, an acid, flammable, two drawer cabinets and one cabinet with glass sliding doors in front. But they have 40+ cabinets left at $100 each. All are blue. I will take pictures Wed and post them with the number and contact information Wed/Thur in the correct part of the forum, I just thought I might mention it here now.

Anyway, thanks again everyone, and so long as the criticism is constructive, I would love to learn more, and improve my process. Please do critique and comment if you see something I need to correct. I want to improve, not argue if that makes sense.

 

[pugle1]

There was lots of good stuff posted in the early days. Except for stuff like auctions, nothing on here is timely. I like seeing the old stuff dredged up. Shows people are searching.

I'm new, so I need the old stuff

 

[Alentia]

I have recovered fully from 4 Pentium Pros (512K Cache) 2.1gr of gold exactly (cleaned and washed).

It is 0.525gr per CPU.

It seems 256K Cache CPUs may have bit more gold. Will post results when completed.

 

[Alentia]

Somewhat interesting... 256K Cache P-Pro CPU yielded 0.25gr per CPU with SMB.

It could be that I still have gold in the solution. Dropped copper wire there to see if some more will be cemented.

 

[lazersteve]

Did the tungsten heat spreaders separate from the ceramic housing in both test batches?

Or did you crush everything?

Steve

 

[Alentia]

Yes and Yes, I crashed it and they did separate. As a matter of fact I have had 90% of it (bigger pieces) removed before AR treatment, so I do not waste time and acids for it dissolve.

Now I can confirm:

1. Pentium Pro 512K Cache - 0.525 gram of gold per CPU
2. Pentium Pro 256K Cache - 0.250 gram of gold per CPU
3. Pentium Pro 1Mb Cache - X.XXX gram - I do not have it to test.

The above may explain all those different results people getting per CPU. It may as well explain why eBay guide claiming 1gr of gold in CPU as it theoretically could be the case for 1Mb Cache chips, which are somewhat rare.

 

[necromancer]

Pentium Pro 1Mb Cache

is there such a thing ??

 

[rucito]

http://en.wikipedia.org/wiki/List_of_Intel_Pentium_Pro_microprocessors

 

[Alentia]

More CPU recovery data approximate as I have processed them all together:

CPU Type                 Grams Quantity    CPU Weight          Output per piece              Output per lb 
DX486                    444       19           23.5               0.18                            3.47 
IBM 6x86                 111	     3	        34.6		          0.2	                         2.62	
Pentium 100 Ceramic	   200	     7	        28.4		         0.07                            1.12	
Pentium MMX Ceramic       88	     3	        30.4		         0.09	                         1.34
K6 (caps removed)	     195	    12	        16.2		         0.08	                         2.24
Athlon Ceramic	        192	    11	        17  		         0.05                        	 1.33

 

[necromancer]

http://en.wikipedia.org/wiki/List_of_Intel_Pentium_Pro_microprocessors

thank you, (two 512 KiB dies)
looks like i need to open my eyes wider

(was originally wondering if it was a single 1024 cashe)

 

[Alentia]

I have processed AMD Fiber Athlon and P3/Celeron following are the results:

CPU gram Quantity Piece gram Output piece Output lb

Athlon Fiber 407 37 11 0.006 0.25 0.222
Celeron/P3 Fiber 504 56 9 0.004 0.20 0.224

 

[g_axelsson]

I have processed AMD Fiber Athlon and P3/Celeron following are the results:

CPU gram Quantity Piece gram Output piece Output Sample

Athlon Fiber 407 37 11 0.006 0.25 0.222
Celeron/P3 Fiber 504 56 9 0.004 0.20 0.224

I have processed AMD Fiber Athlon and P3/Celeron following are the results:

CPU                gram   Quantity  Piece gram  Per Piece Per Pound  Output Sample
Athlon Fiber       407    37        11            0.006   0.25       0.222
Celeron/P3 Fiber   504    56        9             0.004   0.20       0.224

I tried to format the table a bit more readable for the yield data.

Have I understood you right, you processed 37 Athlon Fiber CPU:s, weighing 407g or 11g/piece and you recovered 0.006g per CPU of 0.222g gold per pound.
What I don't get is the number 0.25, what does that mean?

Göran

 

[rucito]

I have processed AMD Fiber Athlon and P3/Celeron following are the results:

CPU gram Quantity Piece gram Output piece Output lb

Athlon Fiber 407 37 11 0.006 0.25 0.222
Celeron/P3 Fiber 504 56 9 0.004 0.20 0.224

I have processed AMD Fiber Athlon and P3/Celeron following are the results:

CPU                gram   Quantity  Piece gram    Output  piece  Output lb
Athlon Fiber       407    37        11            0.006   0.25       0.222
Celeron/P3 Fiber   504    56        9             0.004   0.20       0.224

I tried to format the table a bit more readable for the yield data.

Have I understood you right, you processed 37 Athlon Fiber CPU:s, weighing 407g or 11g/piece and you recovered 0.006g per CPU of 0.222g gold per pound.
What I don't get is the number 0.25, what does that mean?

Göran

Hello
0.25/lb
0.222 from 407gr