My results of specific types of IC chips, flatpacks and BGA

[kurtak]

Alex

Just want to say that I always enjoy seeing when you add something new to this thread 8) :mrgreen: :G

Kurt

 

[Tzoax]

Kurt, thank you very much - in a meanwhile i collected a bunch of all kind of BGA's - so now i can test them separately - soon we will have some comparison related to gold content between Intel, ATI, VIA and nVidia BGA's - and i am sure there will be many surprises.

After all of the gold dissolved, solution was cooled and filtered...



After adding little water and a few grams of SMB...



Gold powder formation...



I will leave it over night to completely settle down.

 

[Tzoax]

Gold powder completely settled down...



Gold powder after washing (3xwater - 3xHCl - 3xwater)...





And finally results...



0.39g of gold out of 116.4g (36 pieces) of this kind of BGA's...

That is 3.35g of gold for 1kg of material... or 0.0108g of gold per chip.

I'm surprised...even small types of BGA's have better yields...i have considered those like high yield BGA's based on their size but...it seems they are not.

I'd love to see your result. I guess these will have less gold than the normal bga, but we'll see.

Yes, you have right.... i expected better yields for this kind of BGA's....but on the other side - that also means there are some other type of BGA's that have more than 10-11g of gold per kg - this kind of BGA's only degraded yields - and still ... when processing mixed lot (including this type and other types of BGA's with internal heatsinks) results are about 10g/kg - so there must be some type of BGA's that bring yields up....and if i had to guess i would say it is Intel BGA's ...but we will see..

 

[LT_golden]

This is my little contribution to the site, I have tested various types of chips for gold content using these steps:
...

Hi Tzoax,

This is awesome beyond words. I believe I speak for both myself and many others when I say that your "little contribution" has turned into a de facto end-all go-to page for all things on what types of materials produces what amount of gold.

This is awesome. You're awesome. Thank you. Can I PayPal you some money? :lol:

 

[kernels]

Hi Alex, good to see you still going strong! Your meticulous work is fantastic.

Yes, your yield on those BGAs seems a bit low, but I have never tried processing only the ones with integrated heatsinks, might be the cause of the low yield. I would still have expected 5+g per kg.

 

[Tzoax]

This is my little contribution to the site, I have tested various types of chips for gold content using these steps:
...

Hi Tzoax,

This is awesome beyond words. I believe I speak for both myself and many others when I say that your "little contribution" has turned into a de facto end-all go-to page for all things on what types of materials produces what amount of gold.

This is awesome. You're awesome. Thank you. Can I PayPal you some money? :lol:

Thank you very much LT_golden, you reminded me of this saying:
"Give a man a fish and you feed him for a day; teach a man to fish and you feed him for a lifetime."

In other words - instead of sending me some money you could share some data from processing specific material - it would bring much more pleasure and benefit to me and to all other members who is interested in gold content...so - when/if you are processing some types of IC's - you are welcome to share your results here - it is very rare that people are willing to share their data but....if you are interested in contribution of knowledge - that is a best way to thank me - to become a part of it.

Hi Alex, good to see you still going strong! Your meticulous work is fantastic.

Yes, your yield on those BGAs seems a bit low, but I have never tried processing only the ones with integrated heatsinks, might be the cause of the low yield. I would still have expected 5+g per kg.

Hi Hein! Thank you, nice to see you. I was very surprised with low yield of those but...now i am preparing some INTEL BGA's (which does not have internal heatsinks) for testing - and i really expect much more gold than 10g/kg. This 4 types of INTEL IC's are most common in PC and laptop motherboards:
1. Weight of IC chip: 1.91g /// side length: 3cm
2. Weight of IC chip: 1.5g /// side length: 2.6cm
3. Weight of IC chip: 1.19g /// side length: 2.4cm
4. Weight of IC chip: 0.8g /// side length: 1.9cm



I am going to test all of those 4 types separately - and with that i will cover most common INTEL types of BGA's.

I will start from the left - with biggest ones - (1.91g, 3cm).

For this test i will use 23 Intel BGA IC chips of this kind - total weight is 44g.

 

[Palladium]

Very concise thread and great work!

 

[Tzoax]

Very concise thread and great work!

Thanks Palladium. A cold weather is slowing me down a bit (it is about -7C) so a incineration is going slow...especially because i am using electric hotplate for that - eventually all of the BGA's are going to become white, fluffy - incinerated properly...and ready for making concentrate..

 

[Tzoax]

I removed as much as i can silicon dies (trying to keep them in one piece - to avoid crushing them - i will add them to concentrate at AR stage). I broke IC's to smaller pieces and continue to incinerate. This is how they look after incineration.



I crushed the IC's with spoon - this is picture of crushed IC's and silicon dies by side.



After several water rinsing/crushing - this is top look of concentrate inside of beaker. Gold bonding wires are forming and could be seen in concentrate.



This is the same - dried concentrate. Weight of this concentrate is 12.2g.

So, after first round of incineration/water rinsing - this concentrate contains 27.72% out of starting material (44g).... or - the volume is reduced by 72.27% out of starting material.

On a top right side of the picture - i separated a few parts of broken silicon dies and a group of gold bonding wires.



And i separated this group - just to see this beauty... 8)



I see there are still some epoxy ash particles that are not incinerated properly...so i will make one more round of incineration/crushing/water rinsing.

 

[FrugalRefiner]

And i separated this group - just to see this beauty... 8)

Angel's Hair. 8)

Dave

 

[Tzoax]

Incinerating final concentrate....



After several crushing/water rinsing - i added 50ml of HCl to try to clean the concentrate even further.

Immediately after adding HCl to concentrate - solution turns to yellow/green color like in the pictures.





I am considering to add this step (washing IC''s concentrate with HCl) in future processing because i noticed that after adding HCl - gold bonding wires are much more tending to clump together, and obvious color change of solution meaning some of base metals went to solution, so adding this step will make concentrate even more cleaner.

I am not sure what dissolved base metal gave this yellow/green color to solution. I think it is from copper (a little of incinerated copper (copper oxide) - which would explain how quick it went to solution).

 

[Tzoax]

After all of gold dissolved - this is look of neutralized and decanted AR solution... even it is not filtered it looks pretty clear - so making good concentrate have multiple benefits.



This is look of remaining material (silicon dies and crushed material) after removing AR solution.
Note that it looks like there is more of gold bonding wires in the center of a beaker - but it is just a sand from outside of the beaker...and bigger particles that looks like not crushed/not incinerated epoxy is really a small parts of crushed silicon dies.



After cooling down and filtering - solution is ready for adding SMB... This is look of first 5 minutes of dropping gold - at speed x5...



Currently - gold powder is settling down...

 

[Tzoax]

Washing gold powder...



Drying gold powder...



Results...



0.65g of gold out of 44g (23 pieces) Intel (type 1) BGA chips.

44 / 0.65 = 1000 / x
x = 1000 * 0.65 / 44
x = 14.77

Yield: 14.77g/kg

23 / 0.65 = 1 / x
x = 1 * 0.65 / 23
x = 0.028

1 piece of IC chip: 0.0282g of gold

By today's price of gold that is: 1.125 US Dollars or 0.988 Euros ... for 1 chip.

Not so bad info for cherry pickers... :shock: 8)

I will take a few days off...and then i will continue with other 3 types of Intel BGA's.

 

[cosmetal]

"I will take a few days off...and then i will continue with other 3 types of Intel BGA's."

You deserve some time off.

Thank you for your continued hard work and shared information!

James

 

[Tzoax]

You deserve some time off.

Thank you for your continued hard work and shared information!

James

James, i can't wait to continue... i will rest for this weekend - and i will continue in monday. And thank you again for resolving mystery of "little elf shoe" here... https://goldrefiningforum.com/phpBB3/viewtopic.php?f=49&t=27544#p290938

I forgot to add some more data about this last test i made (Intel BGA - type 1). Actually it is 492-pin BGA chip - so naming this class of BGA's like that is more precise terminology and easier to remember/recognize it.
They are found on older motherboards - i saw many Socket 370 (and older than that) motherboards having 492-pin BGA chip.



And again, 14.77g of gold /kg and 0.0282g of gold per chip for this type of BGA's - brings this BGA type of chip at first place for gold content based on my tests (so far).

Alex

 

[kernels]

Wow, that is a great result, probably the most meticulous work I've seen. I average somewhere between 10 and 12g per kg for the "black tops", but I usually mix the ones with internal heatsinks with the others which will bring the yield down a bit. Lovely to watch you work mate.

 

[Tzoax]

Wow, that is a great result, probably the most meticulous work I've seen. I average somewhere between 10 and 12g per kg for the "black tops", but I usually mix the ones with internal heatsinks with the others which will bring the yield down a bit. Lovely to watch you work mate.

Hein, thank you...i am sure there will be many more of surprises - there are many more types of BGA's to test.
I have a next candidate for testing. It is a type that in most cases comes together with previous one i tested - 492-pin BGA. More precisely - on Pentium III and older motherboards Intel 492-pin BGA is the Northbridge...and our next candidate is Intel 324-pin BGA - Southbridge.
Weight of IC chip: 1.19g /// side length: 2.4cm



I don't have many of them - only 12 pieces...but it will be enough for test.







After this test - all motherboard BGA's yields - especially Pentium III and older, that contains Intel (Southbridge - 324-pin BGA and Northbridge 492-pin BGA) will be known, of course "ballpark" known.

I don't know is this the case in other countries - but here - there are two classes of motherboards - first class is Socket 370 motherboards and older (higher selling price) and a second class is Socket 462 and newer (lower selling price of motherboards). I am convinced that it is related to higher yield of among all other components --- especially BGAchips... and after this test i will be able to prove it - if gold content of Intel Southbridge - 324-pin BGA is anywhere similar like gold content of Intel Northbridge - 492-pin BGA (14.77g of gold /kg and 0.0282g of gold per chip) - thats it - the main reason for "higher yield" of motherboards starting from Socket 370 and older ones.

Another thing i would like to mention is that i started to make more details and pictures about processing IC's...and after some time - it would be hard to read/find some results through pages and pages of this thread...so if anyone think it is excessive and i should post only my results (not including pictures and details of processing) just let me know. Also - if sending many pictures and gifs are spending too much of server's resources please let me know.

Alex

 

[Tzoax]

First incineration...





After separating (as much as i could) silicon dies and first crushing...



After several water rinsing... (gold bonding wires are grouping in top left corner of concentrate)



After drying concentrate...



I see there are still some unburned/ not properly crushed particles - so i will do another round of incineration/crushing/washing concentrate.

 

[kurtak]

Another thing i would like to mention is that i started to make more details and pictures about processing IC's...and after some time - it would be hard to read/find some results through pages and pages of this thread...so if anyone think it is excessive and i should post only my results (not including pictures and details of processing) just let me know. Also - if sending many pictures and gifs are spending too much of server's resources please let me know.

Alex

Alex

You are doing an "awesome" job :!: :!: :!:

Posting details & pictures of you processing is ALWAYS a good thing

It is threads like this one (& the many other "such" threads write by other "dedicated" members of which there are to many to name) that makes this forum so great

THANK YOU for continuing with this thread

Your work is :G

Kurt

 

[Tzoax]

On a moment i felt like there there are not much interested members about this subject... but then again - when i look back - i see many positive comments. Thank you Kurt, and thanks to all other members supporting my work - for seeing a point of my tests and for keep adding wind at my back.

After final incineration/crushing/washing - this is a look of final concentrate inside of beaker... gold bonding wires looks long and thick - even with naked eyes.



Dissolving gold in AR...





After all of gold dissolved i decanted AR solution in another beaker...then i added a fresh HCl and a little more of KNO3 (making poorman's AR) into concentrate - to get in solution as much of possible gold ions trapped inside of concentrate.. then i decanted solution while still very hot to another beaker. Finally - i rinsed the concentrate with fresh water and again - add a solution in another beaker... Then i heat this solution and neutralize NOx with little of urea (until it stops foaming). I ended with 100ml of AR solution...



After cooling - it is ready for filtering... I am using "charmin plug" filtering with wet tissue wipes (previously washed with water) - inserting it tight into funnel's tube like in the picture.