Video Card Reflow
Lead free solder seems more vulnerable to thermal fatigue than lead based, and I ended up with several failed video cards in my collection that had video artifacts or that would not work at all.
One approach to repair such failures is to reflow the solder and some people do this in an oven, but the temperatures needed to melt the solder tends to destroy electrolytic capacitors on the board, so local heating with a heat gun seems a better alternative. Some people heat the GPU with a heat gun but without using soldering flux and the resulting fix can be short lived. The ideal solution is re-balling where the chip is removed and the solder balls replaced, sometimes with a leaded variety for longer life, but this was too complicated an option for me.
Instead, I decided on a flux-based compromise and purchased a medium sized heat gun (~$25) that was able to reach 480°F to 850°F (there are two settings) and since solder melts around 390°F (less for leaded, more for unleaded) it seemed suited to such work; best to avoid the ultra-cheap ($10) heat guns that do not get hot enough.
I decided on a compromise and purchase a medium sized heat gun (~$25) that was able to reach 480°F to 850°F (there are two settings) and since solder melts around 390°F (less for leaded, more for unleaded) it seemed suited to such work; best to avoid the ultra-cheap ($10) heat guns that do not get hot enough.
Exposing the GPU by removing the heatsink allows one to apply liquid flux that can then flow under the GPU (and memory chips if also reflowing these); I would then heat the chip from above till the board itself started smoking ever so slightly and would test the board after cooling (with the heatsink back in place), but before cleaning off the flux. There is special reflow flux, but I just used what I had for normal soldering which was rosin based and so a little more difficult to clean off; for this I use 91% isopropyl alcohol.
I had thought to use an Infra-Red thermometer I have but ended up just heating till the board started smoking slightly and no burn marks resulted; people using the oven method seem to see a lot of smoke from the board, perhaps from the destroyed electrolytic capacitors and one really does not want to use the same oven one cooks with.
Of the 4 failed PCIe video cards that I had, I managed to bring 2 back to full functionality; not bad for a $25 investment.
The real test is time, but my preference for fan-less cards means I will not be using these cards a lot. Instead, I performed stress testing using the free FurMark program.
This article was written by Anwar (Andy) Shiekh originally from London, England; he repairs things to help make an income go further and presently teaches Physics in Colorado, U.S.A.
Please give a support by clicking on the social buttons below. Your feedback on the post is welcome. Please leave it in the comments.
P.S-If you enjoyed reading this, click here to subscribe to my blog (free subscription). That way, you’ll never miss a post. You can also forward this website link to your friends and colleagues-thanks!
Note: You can check his previous repair article in the link below: