Sign in with
Sign up | Sign in
Your question

Can two slightly bent socket pins cause overclocking instability?

Tags:
  • Overclocking
  • Socket
  • Intel i5
Last response: in Motherboards
Share
July 17, 2014 3:33:20 PM

While putting together my system, I nudged the socket, displacing two pins so that they have just a slightly different orientation to the others, facing a similar direction. The pins are located on the lower middle of the socket, supposedly where the pins responsible for power are located.
Overclocking my i5-2500K with my Z77 Extreme4 appears to be causing hard crashes (Event ID 41: system has rebooted without cleanly shutting down first). I am currently trying a low overclock (4.1Ghz, auto voltage) and seeing if it'll crash. At stock clocks it does not appear to crash, but due to the highly random nature of the crashes, I cannot be sure (I've tested for two days straight with no crashes at stock, a little less when overclocked before crashing, could just be an outlier).
Other variables might be the cause, however: I have a 520W PSU that has previously had no troubles (and the system hasn't frozen under load, only in low-load scenarios), and two new sticks of RAM (HyperX Fury 1333Mhz, tested with Memtest to be fine, each stick boots when inserted individually). Finally, perhaps I just have a bad chip or board. When stress-tested, temperatures are normal and no abnormal voltage (1.3V) is required to get it to run at 4.5Ghz. I've reinstalled Windows to no effect. I'm just stumped.

More about : slightly bent socket pins overclocking instability

July 17, 2014 3:40:40 PM

Did you straighten out the CPU pins? Leaving them bent can cause issues.

My friend had a CPU with 400 bent pins, and he straightened out every single one. It worked flawlessly after that. Bending the pins is okay, just be very very careful.
m
0
l
July 17, 2014 3:47:22 PM

Jak Atackka said:
Did you straighten out the CPU pins? Leaving them bent can cause issues.

My friend had a CPU with 400 bent pins, and he straightened out every single one. It worked flawlessly after that. Bending the pins is okay, just be very very careful.


I did not realign any of the pins (doing so would entail the removal of my Hyper 212 EVO, which is a pain). I am not steady-handed, either, and I've broken a motherboard before with my clumsiness (in fact, this one was a replacement for that one). I might try it, but I lack the confidence to go through with it at the moment, and probably proper tools (I don't have a mechanical pencil or hobby knife handy).
m
0
l
Related resources
July 17, 2014 3:50:57 PM

You just need a small pair of tweezers. I own that heatsink and agree that removing it would be a pain, but if you want to overclock it at all, you may need to straighten the pins out.

You could also look up the voltages necessary for a 4.1GHz OC on that CPU. I have a 3570k in my Extreme4, and I don't remember there being a 4.1GHz OC profile.
m
0
l
July 17, 2014 4:11:41 PM

Jak Atackka said:
You just need a small pair of tweezers. I own that heatsink and agree that removing it would be a pain, but if you want to overclock it at all, you may need to straighten the pins out.


Tweezers, I think I can find. I'll see if I can fix the pins. Thanks.
m
0
l
July 17, 2014 4:57:12 PM

Well, I'm back and have reseated the cooler and CPU. As it turns out, I was unable to find any irregularities in the socket pins at all. It looks as though the difference corrected itself. Perhaps a reseat was all that was really needed, though I can't be sure for the moment. I've increased my clocks (4.5) and voltage (1.325V) to see if problems persist.
m
0
l
July 18, 2014 1:39:08 PM

It crashed again this afternoon. I can't think of anything else I can test for. It could be that my PSU provides insufficient power, which I doubt is the case, as nothing irregular happens under load, and I tested its output voltages not too long ago. Now that I think about it, the thing only has one 4-pin CPU connector... That might just be the problem; the supply isn't of reputable model (in fact, I can't find documentation about it online at all) and is several years old, and I'd thought up to now that one 4-pin was acceptable for a 2500K. There's also the possibility of a faulty board. In my research into this problem, something similar appears to have happened to someone else with the same board. His situation differs in that it's unstable no matter what he does, when my system only appears to lose stability under an overclock.
m
0
l
July 18, 2014 2:33:03 PM

Alright, thanks. I'll try overclocking again once I acquire a new power supply. Was planning to, along with a new GPU, sometime in Q4 of this year.
m
0
l
!