ram yeild

[goldenchild]

No kidding, It makes me chuckle when I see bidders throwing $20 for a pound of ram, and then another $10 in shipping. I can't imagine there being $30 worth of gold in a pound of RAM. Even if there was, you're looking at the time and additional money to refine it. Hopefully, the bubble will burst soon.

I think it's these types of transactions that make the people coming on the forum asking for a-z refining particularly annoying. They are the ones making it nearly impossible to win auctions at a good price. eBay used to be a decent place to get scrap. Now it's just madness.

 

[zarkava]

That's about 2 Lbs of RAM. He should be able to come out okay on it as long as he isn't looking to make a great profit. It would be an investment to learn with. :|

He will not come out OK.
1 kilogram of mixed RAM is anything from 40 to 100 pieces /they weight from 10 to 25gram each/
So we talk about 40 to 100g of close cut fingers. He may pull out 0.1 to 0.4g from fingers.
If he process chips too - even with crude method as we seen on youtube with guy grinding them with mortar and pestle and panning - he got 3g of not refined gold colored powder of questionable karat or composition from 800 chips - we can consider it as urban legend as nobody from here did the same experiment, but hey lets believe this for this moment.../ In his 40-100 pieces he may have 8-16 chips on each so it is from 320-640 or 800-1600. Let take average so we talk 480 to 1200 chips so he may actually get more gold out of chips than from fingers - it depend what type of RAM is he processing and if it all was really gold - on that youtube video. 1.5 to 4g

My conclusion is he is going to lose money on this one as return will not cover purchase, use of acids and in case of processing chips again work - you have to get them crushed, gas for incineration and chemicals for refining.

He will not lose a lot and certainly it will be gain if he will look at it as a learning as you go. It may be nice investment to find out what his ability is and at the end of the day he will know how much he can pay as it all depend on how much is he able to recover. Now he need to take few weeks off and spend them studying.

what about pgms ??? i also got 2% pgms aint ??

 

[Claudie]

patnor is right. You won't make any money on this lot, but you can gain some experience. If you're already out the money for the purchase, use it to learn with. You can get some gold and an education at the same time. Many people out there spend more than $30 on a useless "Get Gold From Computers" DVD or book and end up with nothing to show for their money. Use the information on this site and you will be far ahead. Experience may not always be the best teacher, but it is the most effective.... :|

 

[patnor1011]

That's about 2 Lbs of RAM. He should be able to come out okay on it as long as he isn't looking to make a great profit. It would be an investment to learn with. :|

He will not come out OK.
1 kilogram of mixed RAM is anything from 40 to 100 pieces /they weight from 10 to 25gram each/
So we talk about 40 to 100g of close cut fingers. He may pull out 0.1 to 0.4g from fingers.
If he process chips too - even with crude method as we seen on youtube with guy grinding them with mortar and pestle and panning - he got 3g of not refined gold colored powder of questionable karat or composition from 800 chips - we can consider it as urban legend as nobody from here did the same experiment, but hey lets believe this for this moment.../ In his 40-100 pieces he may have 8-16 chips on each so it is from 320-640 or 800-1600. Let take average so we talk 480 to 1200 chips so he may actually get more gold out of chips than from fingers - it depend what type of RAM is he processing and if it all was really gold - on that youtube video. 1.5 to 4g

My conclusion is he is going to lose money on this one as return will not cover purchase, use of acids and in case of processing chips again work - you have to get them crushed, gas for incineration and chemicals for refining.

He will not lose a lot and certainly it will be gain if he will look at it as a learning as you go. It may be nice investment to find out what his ability is and at the end of the day he will know how much he can pay as it all depend on how much is he able to recover. Now he need to take few weeks off and spend them studying.

what about pgms ??? i also got 2% pgms aint ??

No.
What you got is about 1 gram of monolithics on every ram stick. When you finish taking them off you weight them and you may have Pd - 1-2% of their weight.
So in theory you need 100 RAM sticks to have 100g of capacitors and you may have 1g Pd in them.

 

[dtectr]

And PGMs would need to be processed & refined, as well. Believe it or not, the gold is the easiest to recover, but additional steps, perhaps chemicals - it isn't something you can take a list of A-Z steps & process away.

This is why you'll see some, but not all, refiners here focus on a speciality - Gold, silver, the platinum group, etc.

If you just want to give somebody $30 [comment deleted before posting by author].

Good luck.

 

[arthur kierski]

i got by experiments some information on gold yeild in chips from memory sticks that i want to pass to the forum members----------
i took 2,5 kilos of memory sticks and from them i obtained exactly 1kilo of chips
from the chips obtained i did my usual extracting method(incineration and so on) and obtained--1gram of gold------and some silver(did not measured it)

conclusion: 1kilo of mixed memory sticks contains 400grams of chips and 0,4grams of gold and some silver------27grams of smd(aprox 0,3grams of pd)

 

[Marcel]

This yield is very hard tp predict.
The result depends on the type of chip, its date of production and the manufacturer.
Micron f.i. today uses Pd or Au bondwires..
The more pins a chip has, the more and the longer are the connection bond wires. Just today I researched one chip, because there is an auction on
18.000 ICs Motorola MC14053BCP. (Dont know if all can see that:
http://www.ebay.de/itm/200607574579?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
Well, this is no DRAM it is a analog switch and even in DIP but it also has to be interconnected.
The amount of gold in this 16 Pin DIP IC is 0.36 mg/piece, which results in 6480 mg or 6,48g for the whole batch of 18000 chips ! (source is mfr. datasheet). Would you have guessed it? No good deal...
Btw. one chip weights 1g so the whole shipment is 18Kg.
If you have mixed chips it is impossible to predict the outcome. It will vary from batch to batch. Flatter chips with many pins are more likely to yield better - if they contain any gold at all!
And no, there is no gold in those ICs apart from the bond wires.
The best gold containing chips in the PC area are the chipsets, FBGAs and other large ICs on motherboards VGA cards or controllers.
In a modern SDRAM chip I estimate the gold content around 0,1 mg/chip. Which brings us back to a rough 0,1% ratio of total weight vs gold(-wires) - if there is any gold at all.
To test I use a small, very cheap dremel tool (engraver) and open the chip in the main area and simply look inside if there is gold.

 

[patnor1011]

I have opened many thousands of them and I found a lot of them with gold plating on various places, not only bonding wires. 8)

 

[Marcel]

I have opened many thousands of them and I found a lot of them with gold plating on various places, not only bonding wires. 8)

Are we still talking about SDRAM? or in general? Of course there is gold in leadframes of many other chips.

 

[samuel-a]

This yield is very hard tp predict.
The result depends on the type of chip, its date of production and the manufacturer.
Micron f.i. today uses Pd or Au bondwires..
The more pins a chip has, the more and the longer are the connection bond wires. Just today I researched one chip, because there is an auction on
18.000 ICs Motorola MC14053BCP. (Dont know if all can see that: http://www.ebay.de/itm/200607574579?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
Well, this is no DRAM it is a analog switch and even in DIP but it also has to be interconnected.
The amount of gold in this 16 Pin DIP IC is 0.36 mg/piece, which results in 6480 mg or 6,48g for the whole batch of 18000 chips ! (source is mfr. datasheet). Would you have guessed it? No good deal...
Btw. one chip weights 1g so the whole shipment is 18Kg.
If you have mixed chips it is impossible to predict the outcome. It will vary from batch to batch. Flatter chips with many pins are more likely to yield better - if they contain any gold at all!
And no, there is no gold in those ICs apart from the bond wires.
The best gold containing chips in the PC area are the chipsets, FBGAs and other large ICs on motherboards VGA cards or controllers.
In a modern SDRAM chip I estimate the gold content around 0,1 mg/chip. Which brings us back to a rough 0,1% ratio of total weight vs gold(-wires) - if there is any gold at all.
To test I use a small, very cheap dremel tool (engraver) and open the chip in the main area and simply look inside if there is gold.

I like this guy :mrgreen:

 

[trashmaster]

Marcel welcome to the forum;;;

What part of Germany are you from ??? I lived in Giessen for about 17-18 years ..

I can speak more german than i can type.

 

[trashmaster]

Sam check his profile '' :lol: We need more of this type,,, He can teach us a thing or two or three..

 

[Marcel]

Marcel welcome to the forum;;;

What part of Germany are you from ??? I lived in Giessen for about 17-18 years ..

I can speak more german than i can type.

This is getting offtopic... I was born in Giessen and I went to university of applied science there.
It is only 15 minutes from where I live now, and another 15 from Friedberg.
Best times for theses places was when the US army was there. Now all facilities are empty or torn.
But Giessen depot is still there.
You may know this place also. This is how it looks today...



At least something on the topic: The guy using the "crude" method of recovering gold with pure water and mechanical force is a german dude.
I like this way because, there are no chemicals involved and it really is a poor man´s method. I doubt he yield very good, due to his technique but he has prooven it to be possible. Datasheets is one thing, but the ideas people with limited means get are amazing to me. If you asked an engineer on how to recover the gold, he will make it a complicated, costy task.
Btw. he is not saying that he uses RAM chips, just chips in general. So his result is supposingly a mix of various types of chips and I believe, that mostly are large flatpacks.
In teh youtube video he says, "I received them from a friend". Which may mean, he stole them from the junk yard
Here there is a big fight going on , about the e-waste. Most large garbage collectors, just disasamble the PCs then sell the components over ebay. There are only 3-4 real refiners in Germany, due to the very high standarts when working with chemicals. Also we are not allowed to buy H2O2 33% without large paperwork ( thanks to some matyr idiots who tried to blow up a US facilities with it (TATP) a few years ago).

Btw: I dont want to mess up this thread completly, so if it does not fit in here, please move it to another one:
You can easly crack open any IC at the very right area without burning by simply putting it into the microwave oven for 5-6 seconds.
The area around the die gets cracked and you can very easy remove the epoxy at that very area, without the need to process the leadframes, copper etc.
I do that sometime sto look inside a chip, but I am not certain wheter the bonding wires survive that. maybe they get melted and could be recovered using AR?
Have not tried that out, maybe never will.
Don´t be afraid on the sparkles and the sound, this will not harm your oven or installation. (You should however not always bake your chips in the same oven where wifey makes the popcorn..). Just get an old microwave for that purpos. Brom and other chemicals will leave the compund and they are very unhealthy! If you get the right point, no odur or fumes will come out. Got to play with the watts and time a bit. Also experiment placing them in water or not.
My camerea´s batteries are down right now, so maybe later I can post some pictures on that.
Its a very fast process to crack them open at the right spot with minimal BM involved.
And while I am at it: I rember an article from the US magazine "popular science" from the late 1990s. There a guy tried out to melt(smelt?) gold in a microwave oven.
And yes, it is possible.

 

[Claudie]

I would like to see pictures of that.

 

[lazersteve]

I've done a few experiments several years ago using a pressure cooker and water (superheated H2O) to degrade the plastic housings on integrated circuits and met with some success. I also inlcuded several other e-scrap items in the cooker to observe the effects.

Overall the test went well, until the cooker ran dry of water. At this point the steam ceased and the items inside begin to burn.

I have not pursued the experiment any further. I feel it is a moderately safe and feasible method to break the epoxy matrix in the plastic chips but you will most likely need a proper high pressure autoclave to get the job done.

I based the experiment on research I did on the effects of supercritical water on organic materials.

Welcome to the forum Marcel !

Steve

 

[Marcel]

I would like to see pictures of that.

Picture of what?

The metal melting microwave?

http://www.popsci.com/diy/article/2003-09/smelting-microwave
( That is part of the PopSCi article I was talking about. I love this mag, only too expensive for me now.)

and the "inventer of homemade metal microwave melting", David Reid. Maybe ypu shoould invite him here and he can explain it all.

http://home.c2i.net/metaphor/mvpage.html


and if you mean pictures of a microwaved chip - you can do it by youself in a minute. But I will make some and post them, but not today.

 

[Claudie]

Thank you for the links. That helps me understand more. The site shows him handling the insulation bricks, that are cool, with molten metal on the inside. There is something about running a microwave oven with metal inside that is going against what I was taught. I know that many things I was taught are wrong, but it still throws up a red flag in my mind. :|

 

[samuel-a]

Sam check his profile '' :lol: We need more of this type,,, He can teach us a thing or two or three..

YES i agree!
My comment was positive (but might be interpreted otherwise). I think very highly on Marcel and he seems to be very knowledgeable and experienced.
Definitely a welcomed addition to this house.

Welcome aboard Marcel !

 

[Marcel]

Thanks for the warm welcome golddiggers :lol:
As for me, my eyes open wide reading through this forum, because you have experience and knowledge of the how-to that nobody else has.
You can learn how to design semiconductors and circutry but noone will teach you how to take it apart and refine the PM from them.
In that sense I can only contribute very small, but maybe from a different perspective. The knowledge in this forum is unquie- worldwide I would say. Not even the engineers who work for refiners dig in that deep. Many just pulverize everything then let the material pass throughs everal stations they bought.
But the processes that you have worked out are unbelieveable.
Pd in resistors ? May I confess than I have thrown roughly a million of them away, because i thought they were worthless (they were not RoHS compliant)?
I can only recommend any new user to this forum to read, read, read and read again. There is so much precious information here at bare hand, this is just awesome!

 

[Marcel]

Sorry,
I just couldnt find one of the major DRAM mfr, who uses goldbonding. (Micron (Al, Pd), Samsung, etc)
Are you sure you find gold bondwires in DRAMs?
Some even use wireless bonding (die-toframe bonding, direct bonding etc)
Some info on bondwire material:

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

I crunched a handful of DRAM ICs tonight and inspected them, but could'nt find any gold in them, neither bondwires nor any other component with gold on it.
But who knows? Maybe some DRAM mfrs do use gold in their production.
Maybe next time dont blend all ICs and check by type to see, if there really is gold in them.
I at least will sell my DRAM ICs at ebay next week.. Maybe soone else has more luck with them.
For me RAM (chip) yield is quite unpredictable, but not high.