RAM modules and memory analysis report

xelexo
Advertise with us
shop deals on amazon
Shop deals on ebay
Gold Refining Forum

Help Support Gold Refining Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
kjavanb123

kjavanb123

Well-known member
Joined
Apr 1, 2009
Messages
1,746
Location
USA
All,

I also did 190 grams of mixed shredded RAM and memory sticks, and that so far produced 22.5 grams of alloy metal, which analysis will be posted in this post.

Here is the alloy picture,
image.jpg

Regards,
Kevin
 
I didn't count them as these are traded per lbs or kg so I wanted to get an estimate per kg. Here is a photo of shredded ones and originals before shredding
image.jpg

And
image.jpg

Regards,
Kevin
 
someone correct me if i am wrong, but those smaller RAM sticks look like older printer memory which might skew your results if there was a bunch in your sample batch.
 
No those sticks have a black dot in the middle, all of these are RAM or memory, some of them tin plated.

Regards,
Kevin
 
kjavanb123 said:
No those sticks have a black dot in the middle, all of these are RAM or memory, some of them tin plated.

Regards,
Kevin
Click to expand...


ahhh, okay. i was thinking that with fewer IC chips on the small ram modules could skew the results to the low side... it would be interesting to see how a batch of modern RAM (the bigger sticks with more chips on a stick) compares to the batch you are doing now.

I am looking forward to seeing the results. :)
 
mls26cwru said:
kjavanb123 said:
No those sticks have a black dot in the middle, all of these are RAM or memory, some of them tin plated.

Regards,
Kevin
Click to expand...


ahhh, okay. i was thinking that with fewer IC chips on the small ram modules could skew the results to the low side... it would be interesting to see how a batch of modern RAM (the bigger sticks with more chips on a stick) compares to the batch you are doing now.

I am looking forward to seeing the results. :)
Click to expand...
Those in pic 2 look like 30 pin SIMM's there is still a lot being sold and I think they actually yield higher than the newer DIMM's.
here is a link that shows the different types of memory;
http://www.google.com/imgres?client=firefox-a&hs=vH4&sa=X&rls=org.mozilla:en-US:eek:fficial&biw=964&bih=449&tbm=isch&tbnid=kNd04sjYW3R2mM:&imgrefurl=http://www.canadaram.com/glossaryofterms.htm&docid=zKBXSRInBSeN6M&imgurl=http://www.canadaram.com/images/RAMtypes2.jpg&w=680&h=375&ei=DiaFUv3dEOj3iwKwyYHYBA&zoom=1&ved=1t:3588,r:4,s:0,i:102&iact=rc&page=2&tbnh=167&tbnw=302&start=3&ndsp=11&tx=142&ty=62
 
There has been talk lately of having to sort RAM in the near future when selling to the board buyers. The newer DDR2 and DDR3 have very thin chips and (possibly) less PM's.
 
gold4mike said:
There has been talk lately of having to sort RAM in the near future when selling to the board buyers. The newer DDR2 and DDR3 have very thin chips and (possibly) less PM's.
Click to expand...

Or more, thin chips mean more units in weighted sample. More chips mean more wires.
 
patnor1011 said:
gold4mike said:
There has been talk lately of having to sort RAM in the near future when selling to the board buyers. The newer DDR2 and DDR3 have very thin chips and (possibly) less PM's.
Click to expand...

Or more, thin chips mean more units in weighted sample. More chips mean more wires.
Click to expand...
I just broke a DDR2 chip into tiny pieces. Finally I found the bond wires, they were tiny! The memory IC was a BGA type of package with two groups of solder balls along the long sides sitting on a thin glass fiber board, the die was mounted upside down on top of this board. The bond wires were bonded from the fiber board down to the die along the center line, resulting in a minimal length bond wire.

I think that the gold content is lower per memory stick just because the chips are so small. The fiber board will probably make recovery a bit trickier. The only base metals I could see were tin-solder in the balls and copper in the traces on the board.

I would probably treat the chips with HCl to remove the tin before incineration, then incinerating and crushing the ash where I would only remove the fine free flowing ash before dissolving copper in nitric acid to get rid of copper, and then finally a leach in AR to extract the gold.

Göran
 
All,

Here is the icp result for mixed memory and ram cards I smelted,

image.jpg

So 190 grams of mixed ram sticks and memory, produced 22.5 grams of alloy metal, so 1000 kg would produce 118.421 kg of metal alloy, which 1 metric tonne of it according to analysis would have 1000 grams of gold, therefore 118.421 kg of metal alloy would have 118.4 grams of gold at 100% recovery.

Please advise.

Kevin
 
Again Kevin, I would strongly recommend that you don't base a business decision on the assay you have had from this small batch. I can assure you that the results you have are glaringly inaccurate, and are based upon the small batch size, and the types of RAM shown in the pictures.

RAM memory yields far higher than 118.4g per tonne as an average across the board. It's actually far higher than the calculations presented for the mobile phone boards you posted on another thread.
 
did they remelt all the little beed together,or only took one for sampling, they probably not all have the same yield, did you tryed your thing on fiber cpu because i would love to see the result yield
 
Spaceship,

I know what you are talking about, because RAMs are more expensive that cell phone boards hre in Dubai, so they must have more PMs.
Next time I will group them base on models and year and type, then incinerate, ball mill then smelt.

Regards
Kevin
 
Depending on melting or smelting (with flux), base metals can be oxidized or reduced, this would change the amount of base metal, as end result, more or less depending on procedure used, and or flux used.

As ericrm, points out unless the metals were all melted together in a good mix and representative sample of the mixed metal bar was taken (usually by drilling the bar), the assay may not be a true representation of what you have.

When I add up the percentage's of every metal in one of the assay results posted, I only see about 50%, this is confusing to me, if this was an assay on the melted metal what was the other 50% not noted?
 
ericrm said:
did they remelt all the little beed together,or only took one for sampling, they probably not all have the same yield, did you tryed your thing on fiber cpu because i would love to see the result yield
Click to expand...

That is what puzzling to me, last time I melted the beads into one bar then send that to lab, but this time I sent the as beads so Might be possible they made a homogenuis sample and worked on that.
I have incinerated one Chinese and one non-Chinese PC boards, and will ball mill them, then try it again.

Sample incinerated PC boards;
image.jpg

Regards,
Kevin
 
Platdiger,

Please note 1000 grams of gold per 1 metric tonne of alloy metal, not 1 metric tonne of RAM modules.

Butcher,
Please advise if the following procedure is assuring to get me a good estimation on assay,
1- Take one kilogram of same model and type RAM module, log every one of their models or ID number,
2- incinerate them
3- Ball mill the result
4- Pass it through a sieve of 10 mesh
5- Melt the metals on top of the seive into a bar
6- Drill the bar.
7- Send the drilled piece, and ashes that pass through the seieve to lab
8- fire assay both samples for gold and palladium, ICP them for the rest of elements.

Best regards,
Kevin
 

Latest posts

Back
Top