Decaping an Ic

[pwarrow88]

Hi I was wondering if there is a way beside using acid and high temperature to dissolve epoxy-cresol-novolak. I wish to do this as to be able to expose the Silicon die within the package.

[Guest]

Expose what Silicon die inside what package? The CPU? 

[pwarrow88]

2 minutes ago, ShrimpBrime said:

Expose what Silicon die inside what package? The CPU? 

Yes but no. What I meant was to remove the outer epoxy layer on an IC

[pwarrow88]

6 minutes ago, ShrimpBrime said:

Expose what Silicon die inside what package? The CPU? 

Although I am also working on removing CPU die as well. I have done it successfully multiple times however the results are not as consistent as I want, so I am doing research and testing to find a better method that is easy enough for everyone to do. Here is one of the CPU I have done.

 

 

 

[CommanderAlex]

Perhaps take the surface against a belt sander and sand it away and be careful to not go too far?

[pwarrow88]

4 minutes ago, SpiderMan said:

Perhaps take the surface against a belt sander and sand it away and be careful to not go too far?

This although possible will yield inconsistent results because we often don't know how deep the die is located in the epoxy. I am currently in search of a fast and consistent process and trying to move to a more chemical method as that leads to a more reproduceable result.

[CommanderAlex]

25 minutes ago, pwarrow88 said:

This although possible will yield inconsistent results because we often don't know how deep the die is located in the epoxy. I am currently in search of a fast and consistent process and trying to move to a more chemical method as that leads to a more reproduceable result.

I'd think if it was a chemical process, then this can lead to other issues affecting the IC. 

[Guest]

26 minutes ago, pwarrow88 said:

This although possible will yield inconsistent results because we often don't know how deep the die is located in the epoxy. I am currently in search of a fast and consistent process and trying to move to a more chemical method as that leads to a more reproduceable result.

Usually the cpu core is located in the PCB. Near the center towards the bottom. The silicon on top is an insulator, but good for heat transfer up and out, then the use of a heat sink or IHS plate. 

 

Chemical separation is an interesting approach though. Nice picture above too btw.

 

Not sure about other ICs. Pretty sure the one depicted would come apart in 2 pieces being epoxied together. 

[pwarrow88]

34 minutes ago, ShrimpBrime said:

Usually the cpu core is located in the PCB. Near the center towards the bottom. The silicon on top is an insulator, but good for heat transfer up and out, then the use of a heat sink or IHS plate. 

 

Chemical separation is an interesting approach though. Nice picture above too btw.

 

Not sure about other ICs. Pretty sure the one depicted would come apart in 2 pieces being epoxied together. 

If I am not mistaken for ceramic IC they can come apart as 2 pieces reletively hasle free with just some heat but that still leaves the question for how to get the chip out of a full epoxy one. Also glad to hear you like the pic hopefully I can capture more in the future and post them. I am currently working on a Intel xeon 3430 die

[LogicalDrm]

-> Moved to Hobby Electronics