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

[Tzoax]

Alex,

If known, and not too much trouble, can you list the manufacturer's name and the actual, or approximate, manufacture date of the boards?

Thanks!
James

Sure, i will post it tomorrow - for date manufactured is not a problem - it is in printed in code of almost any IC chip on board. I just have to figure out where (if it is anywhere) printed model of board....or manufacturer name. So far all printed markings i checked related just to components nearby... i will try to find it.

I think i found one model on one board (board from picture No. 8 )- based on markings i found - 123202-h it seems it is Digital cellular telecommunications system (Phase 2+);
Universal Mobile Telecommunications System (UMTS);
Circuit switched data bearer services
(3GPP TS 23.202 version 9.1.0 Release 9)

http://www.etsi.org/deliver/etsi_ts/123200_123299/123202/09.01.00_60/ts_123202v090100p.pdf

I also found on IC chip's code - they are manufactured at 2010.

Tomorrow i will find models / manufacturers / date manufactured for rest of the boards.

I will also note mlcc's weight from those boards when i depopulate them.

Alex

 

[Tzoax]

I haven't found manufacturer but i checked date made on IC's - they are from 2010/2011.
Today i depopulated some of the BGA IC chips - 257g without solder balls / with solder mask. There are also BGA IC's with heatsink and small BGA's - i will depopulate them also and weight them, like the rest of non BGA chips.

 

[g_axelsson]

"Solder mask" is the green (or other color) coating on top of the conductors on the outer layers on a PCB. What you are calling "solder mask" I would call "PCB", "Printed Circuit Board" or "substrate". It's also been called "green bottom" if I'm not totally wrong.

Göran

 

[Tzoax]

"Solder mask" is the green (or other color) coating on top of the conductors on the outer layers on a PCB. What you are calling "solder mask" I would call "PCB", "Printed Circuit Board" or "substrate". It's also been called "green bottom" if I'm not totally wrong.

Göran

Thank you for explanation Goran. I will remember that.

Alex

 

[Tzoax]

53.2g of small BGA's.



96.2g of BGA's with heatsink/metal plate.



I will separate black tops from green bottom/PCB substrate from all of 3 types of BGA's and i will note ratio between them.

 

[Tzoax]

I excluded this type of BGA's from the weight because they are flip chips - they don't contain gold:



Ratio between black tops / substrate:



small BGA: Black Tops: 33.2g - 62.4% Substrate: 20g - 37.6% Total: 53.2g
Metal Top BGA: Black Tops: 5.5g - 5.7% Substrate: 90.7g - 94.3% Total: 96.2g
small BGA: Black Tops: 118.3g - 55% Substrate: 97.4g - 45% Total: 215.7g

Total black tops from all of (gold containing) BGA's: 157g

 

[Tzoax]

There is total 13.7g of ceramic resonators:





I am planning to test those too...but i'm afraid that this is very small amount of material to test, unless the yields are pretty high. Does anyone have some ball park yields about ceramic resonators?

Alex

 

[Grelko]

I excluded this type of BGA's from the weight because they are flip chips - they don't contain gold:

I'm sure you just put the DSP chip there as a reference to show what the "flip chips" look like, but there's a chance that some people might think those are "flip chips" also, just because it's in the picture.

Very nice work by the way, thank you for taking the time to do this.

 

[Tzoax]

I excluded this type of BGA's from the weight because they are flip chips - they don't contain gold:

I'm sure you just put the DSP chip there as a reference to show what the "flip chips" look like, but there's a chance that some people might think those are "flip chips" also, just because it's in the picture.

Very nice work by the way, thank you for taking the time to do this.

Thank you Grelko. Those "DSP" chips really are flip chips - on the bottom of that picture is how that IC chip looks like when it is "opened" - both internal sides of it. At first i thought it is a regular BGA gold containing IC chip, but when i opened one - that "black top" is not IC chip but some sort of heatsink. It just look like gold containing BGA chip.

By the way, i am still depopulating boards - so far i depopulated 210g of MLCC's and 320g of IC chips (non BGA).

 

[Grelko]

Thank you Grelko. Those "DSP" chips really are flip chips - on the bottom of that picture is how that IC chip looks like when it is "opened" - both internal sides of it. At first i thought it is a regular BGA gold containing IC chip, but when i opened one - that "black top" is not IC chip but some sort of heatsink. It just look like gold containing BGA chip.

Now, that's interesting. I'll need to go through my chips again. I've seen that same DSP while I was depopulating. They sure looked like regular BGA's to me.

 

[Tzoax]

I depopulated the most of of MLCC's from the boards - 255.7g.



And there is 406.2g of mixed non-BGA IC chips (132.5g of small IC's and 273.7g is the rest)



Now i am collecting fully gold plated pins from these connectors. I never processed them - i have no idea about yields - i assume they have about 2g / kg.

 

[Tzoax]

There is total of 187.2g of those fully gold plated pins only from that type of connector. The rest of pins from boards (white connectors) are partial gold plated.

 

[Tzoax]

I incinerated some of the IC's...



And i remember i said i will test transistors (mixed transistors from motherboards) so i have incinerated 348.3g of them to make a test...i hope i will recover some visible gold from them.

 

[Tzoax]

Out of 348.3g of transistors - 204.5g are copper plates with wires, and 113.8g is powder. That means the weight of emission of smoke/burnt particles is 30g.



I made concentrate out of powder and this is a picture of gold bonding wires from mixed transistors out of all kind of motherboards.

 

[Tzoax]

Reaction of concentrate from transistors (with AR) was very vigorous and long - i did recover some of gold but it was too small amount to get some data.

Now i have some really interesting samples of most common IC chips for testing. I will begin with this type of BGA IC chips - NVidia GeForce BGA IC chips (with internal heatsinks) - they are most common on older AGP graphic cards.



There are several versions of NVidia IC chips - coming with different sizes and with or without internal heatsinks. This type is most common, the side length of square is about 2.85cm and having shiny circle on the chip means it have internal heatsink.



For this test i will use 36 peaces of this kind of IC's.




Total weight of 36 peaces is 116.4g - so average weight of one piese of ICchip is 3.23g (just a chip - without green bottom solder mask).



This type of BGA IC chips is very interesting for testing because they are heavy (3.23g) and they are very common in AGP graphic cards.... Knowing that average gold yield for BGA's are 10+ g/kg (black tops) i wonder how this kind of BGA's fits into that proportion.

 

[archeonist]

I'd love to see your result. I guess these will have less gold than the normal bga, but we'll see. You deserve a statue for your work

 

[cosmetal]

I'd love to see your result. I guess these will have less gold than the normal bga, but we'll see. You deserve a statue for your work

Naw . . .

More like the Nobel Prize in eWaste Recovery and Refining!

James

 

[Tzoax]

Thank you guys, i hope these tests will improve knowledge about gold content in specific types of IC's.

Internal heatsinks weight: 62.7g. That is 1.74g per chip...or 53.86% of total weight of chips.




Incineration is complete - today i will start making concentrate.

 

[Tzoax]

This is look of concentrate after first incineration / water rinsing. Gold bonding wires starts clumping together and becomes visible with naked eyes. I see there are some of unburned parts of chips so i will repeat incineration / crushing / water rinsing to make much better concentrate - to "get free" gold bonding wires still trapped inside those unburned / uncrushed particles and to rinse as much as i can carbon / ash material to avoid absorbing gold chloride in AR solution in later steps.

 

[Tzoax]

Final incineration...



After final water rinsing i added HCl in a beaker with concentrate and after little mixing around - a pile of gold bonding wires formed.



And it looks pretty promising to me... a group of gold bonding wires are about 2cm long.



It is time for KNO3 addition in small portions (making a poorman's AR).