I have a Galaxy 6600GT AGP graphics card like this one:
http://www.guru3d.com/articles_pages..._review,6.html
The caps are all the same as shown in those pictures. It's 8 and half years old so it's not like I'm fussed about it financially but it is a nice little card.
Recently I did a cooler swap to try and bring down the temperatures and noise. I'm very familiar with cooler swaps and have done a number of customised heatsinks. Under testing in Furmark the card was running low 70s and it froze. Now it won't boot and cross testing shows it is definitely the card. The temps are fine so it definitely isn't that the GPU itself is fried, more likely something else on the board.
It was suggested elsewhere that it could be the caps and having not fixed a card at the component level I am keen to try and learn something from this failure, diagnose a fault and fix if I can. All of these components are old so it could be a combination of age as well as stress (Furmark is unrealistic load) that has made something go.
Nothing is visibly wrong with the caps (they are mostly SMD) and I don't have any capacitor tester but something I did think of is that if the caps were to be damaged, they might end up shorting. Testing resistance obviously doesn't give much useful with most capacitors but 2 of them give exactly 4.8 ohms constantly. This got me thinking that maybe these 2 have been damaged?
Is damage under stress testing a possibility? Obviously I'm not definitely saying its the caps as it could easily be something else but if anyone has any ideas on how I could prove the caps are damaged or diagnose any other fault that might be fixable I'd be willing to hear your ideas.
Thanks.
http://www.guru3d.com/articles_pages..._review,6.html
The caps are all the same as shown in those pictures. It's 8 and half years old so it's not like I'm fussed about it financially but it is a nice little card.
Recently I did a cooler swap to try and bring down the temperatures and noise. I'm very familiar with cooler swaps and have done a number of customised heatsinks. Under testing in Furmark the card was running low 70s and it froze. Now it won't boot and cross testing shows it is definitely the card. The temps are fine so it definitely isn't that the GPU itself is fried, more likely something else on the board.
It was suggested elsewhere that it could be the caps and having not fixed a card at the component level I am keen to try and learn something from this failure, diagnose a fault and fix if I can. All of these components are old so it could be a combination of age as well as stress (Furmark is unrealistic load) that has made something go.
Nothing is visibly wrong with the caps (they are mostly SMD) and I don't have any capacitor tester but something I did think of is that if the caps were to be damaged, they might end up shorting. Testing resistance obviously doesn't give much useful with most capacitors but 2 of them give exactly 4.8 ohms constantly. This got me thinking that maybe these 2 have been damaged?
Is damage under stress testing a possibility? Obviously I'm not definitely saying its the caps as it could easily be something else but if anyone has any ideas on how I could prove the caps are damaged or diagnose any other fault that might be fixable I'd be willing to hear your ideas.
Thanks.
Comment