X51 Rebooting Problem & Best GPU Replacement

[Parmageddon]

Hello,

Bought an X51 a couple years back, didnt really use it until recently, unfortunately its suffering from the intermittent rebooting issue that seems to quite a well known issue. Windows event logger identified this as a Kernel Power Cricitcal issue, which more than likely means its shutting down due to insufficient power. (Could be drawing too much power? Could be faulty PSU, Could be faulty GPU too apparently) Spec is as follows,

X51 R2
I7 4770K
GTX 760 OEM 192bit
8GB (2x 4gb) RAM
1TB HDD
330W Brick

So far on the hardware side, i've removed the CD Drive, reseated all the components inside, made sure all the wires and cables are properly connected.
Software side, i've limited the CPU to 90% capacity, same goes for the GPU.

One thing I wanted to change out was the GPU, The 760 OEM was OK 5years ago, not so much now. A natural card that'd be a good replacement would be the 1060 6GB. I just wanted to ask you guys because of the awful heat temps in this tiny case, is there a specific version of the 1060 that would work best in the X51? I read something about this on one of the posts, but cant find it....something about one that blows the air in a certain way.


Finally, if there's anything else you guys can think of that may be causing the restarts, i'd obviously be very grateful.

thanks for all your help.

ParmaG

 

[Parmageddon]

Update :- I can't be absolutely certain that this is the same issue. But on the latest 3d Mark, I managed to repro an issue 90% of the time.

Basically when running one 3d Mark simulations, all of a sudden the screen would go black saying "no input can be found" as if the HDMI wasnt plugged in. and the GPU would speed up to 100% fan capacity. It would stay in this state, until I shut the PC down.

Frustratingly, Windows Event Logger, did not kick out an error for this....and there were no memdumps to be found either. I also checked my Openhardware Monitor logs to find, all temps and fan speeds were not crazy, temps for CPU and GPU around 70'c and fan speeds also around 60%...

Now at the scratching my noggin phase, I removed the GPU, and plugged my HDMI into the Intel onboard graphics slot this time and tried running the 3d mark simulation again. It ran like shit ofcourse, but no crashes, no odd behaviour, completed the simulation, albeit with a terrible score.

This still did not rule out faulty PSU, it also didnt rule out that my system could be drawing too much power for the PSU as i'd took the GPU out...... So I tried putting an older 660Ti in this time. Bearing in mind the power output for this is 30w HIGHER than the 760 OEM, and it requires 2 x 6pin rather than the single 6pin.... I expected this to give me all sorts of problems, but it ran the 3d Mark simulation absolutely fine, and actually logged my highest score ever on this machine.

I'm now lead to believe the problem is the 760 OEM GPU, but i'll continue to monitor and update here accordingly.

 

[MURRARI]

Update :- I can't be absolutely certain that this is the same issue. But on the latest 3d Mark, I managed to repro an issue 90% of the time.

Basically when running one 3d Mark simulations, all of a sudden the screen would go black saying "no input can be found" as if the HDMI wasnt plugged in. and the GPU would speed up to 100% fan capacity. It would stay in this state, until I shut the PC down.

Frustratingly, Windows Event Logger, did not kick out an error for this....and there were no memdumps to be found either. I also checked my Openhardware Monitor logs to find, all temps and fan speeds were not crazy, temps for CPU and GPU around 70'c and fan speeds also around 60%...

Now at the scratching my noggin phase, I removed the GPU, and plugged my HDMI into the Intel onboard graphics slot this time and tried running the 3d mark simulation again. It ran like shit ofcourse, but no crashes, no odd behaviour, completed the simulation, albeit with a terrible score.

This still did not rule out faulty PSU, it also didnt rule out that my system could be drawing too much power for the PSU as i'd took the GPU out...... So I tried putting an older 660Ti in this time. Bearing in mind the power output for this is 30w HIGHER than the 760 OEM, and it requires 2 x 6pin rather than the single 6pin.... I expected this to give me all sorts of problems, but it ran the 3d Mark simulation absolutely fine, and actually logged my highest score ever on this machine.

I'm now lead to believe the problem is the 760 OEM GPU, but i'll continue to monitor and update here accordingly.

Sounds like the 760 is faulty for sure.
If it’s an issue of it overheating at some time to the point that the solder was melted you could do a ghetto solder reflow in your oven.

It’s a temporary fix but it is the only suggestion I have. Could be other issues.

 

[MURRARI]

Hello,

Bought an X51 a couple years back, didnt really use it until recently, unfortunately its suffering from the intermittent rebooting issue that seems to quite a well known issue. Windows event logger identified this as a Kernel Power Cricitcal issue, which more than likely means its shutting down due to insufficient power. (Could be drawing too much power? Could be faulty PSU, Could be faulty GPU too apparently) Spec is as follows,

X51 R2
I7 4770K
GTX 760 OEM 192bit
8GB (2x 4gb) RAM
1TB HDD
330W Brick

So far on the hardware side, i've removed the CD Drive, reseated all the components inside, made sure all the wires and cables are properly connected.
Software side, i've limited the CPU to 90% capacity, same goes for the GPU.

One thing I wanted to change out was the GPU, The 760 OEM was OK 5years ago, not so much now. A natural card that'd be a good replacement would be the 1060 6GB. I just wanted to ask you guys because of the awful heat temps in this tiny case, is there a specific version of the 1060 that would work best in the X51? I read something about this on one of the posts, but cant find it....something about one that blows the air in a certain way.


Finally, if there's anything else you guys can think of that may be causing the restarts, i'd obviously be very grateful.

thanks for all your help.

ParmaG

You want the bottom style GPU. It’s called a blower style. The purpose is to blow the heat out of the case instead of just blowing it around inside the case.


https://goo.gl/images/wfGkkw

 

[Parmageddon]

Ah thats really helpful @MURRARI thankyou.

- Do the mini versions qualify as blowers too? Looking at this one in the 360 view, it does appear to have an open back to exhaust the hot air out? I would've thought the smaller the better in such a tiny case...?

https://www.newegg.com/Product/Product.aspx?Item=N82E16814487263

 

[Ravgov]

Ah thats really helpful @MURRARI thankyou.

- Do the mini versions qualify as blowers too? Looking at this one in the 360 view, it does appear to have an open back to exhaust the hot air out? I would've thought the smaller the better in such a tiny case...?

https://www.newegg.com/Product/Product.aspx?Item=N82E16814487263

no the mini versions do not qualify as blower gpus. However I have installed a mini evga 6gb without any problems up to now.
Maximum heat up to now with ultra settings ROTR 83c/ DOOM 73c./ CIV 6 71c. / WOLFENSTEIN 73c. Hence all temps aceeptable according to EVGA.

Regarding reboots I d say u check the PSU. I had same problem with a new 330W I had installed, and when i changed it with the original 220W everything returned to normal. BTW the MINI EVGA 10606GB works fine with the 220WPSU.

 

[Gromz]

Im getting a event 41 kernel power failure but only 1 time in about 2 weeks. my pc is only 1 year old and my 1080ti SLI's are less then that. Nothings changed in pc recently to cause this except windows and drivers updates. it only started about 2 months ago if that and is complete random and happens when pc is idle. Its just random.. when I google event 41 its stated that its almost always the psu..

I tried all obvious things like scans and tests. I do not believe these cards are bad.. I do not think its anything but the psu. But us Alienware owners are screwed. I do not want to put in the same 850w dell crappy psu. I want evga or corsair. It will fit but the cables are stupid in this thing.. Its not normal cables. area 51 R2.. Im done with Alienware just because of the silly components.. I guess I will try to reseat stuff but I find this very unlikely.. Its been running fine for like 10-11 months cause I got this pc and realized I put the weakest gpu in ever and immediately sold them on ebay and bought dual 1080ti FE cards..

Ran fine for very long time then got new windows update and stuff and it started.. Its said that event 41 is never windows or software though... PS nothing is over heating.. I have command center temps up on my other monitor all the time and watch constant and nothing hot.. as I said it does it when idle.. Funny thing is I have never seen event 41 forums say that its a gpu problem.. this is first time I have heard it might be faulty gpu.. I guess it could be anything hardware but like I said its said its usually psu but for it to happen just idle and like 2 times a month is just weird..