[johjoh]

I have a suspicion that my PSU is faulty after having quite a few BSODs.

Please look at
http://www.techsupportforum.com/forums/f299/bsod-stop-0x4e-623372.html
regarding my specific BSOD problems, for reference.

I have used the eXtreme Power Supply Calculator to calculate my PC's consumption (323W [420W with 30% for Capacitor aging, peak efficiency, and future upgrades]) - it's well below my PSU's 520W

Custom-built
· CPU
AMD 1090T Running stock + ASUS Silent Square EVO
· Video Card
GIGABYTE HD4770 1GB
· MotherBoard
GIGABYTE MA78G-DS3H
· Power Supply - brand & wattage
Antec 520W NeoEco
([email protected], [email protected], [email protected], [email protected], [email protected])
· RAM
Kingston HYPERX KHX8500D2K2/4G

 

[efx8]

I would say it's more likely your RAM. Reseat your GPU and RAM and try using 1 stick at a time. You can burn memtest to a disk and boot from it to test if your RAM is faulty Memtest86+ - Advanced Memory Diagnostic Tool

 

[Tyree]

We recommend a minimum 550W good quality PSU for any PCI-E PC.
You noted in the linked post that your hardware is 1-2 yrs. old and 2 yrs on a lower quality PSU, as well as other hardware, could be a problem.
0x4E is a RAM related Error Code.
Try using one stick of RAM at a time in each slot. If no joy run MemTest one one stick at a time and let it make several passes.

 

[johjoh]

Just to update about my current situation:

Before your posts I was on the path to change my PSU. Whilst the '2nd' PSU (borrowed) was in, I had not received any BSODs at all. Although I wanted to test it just to make sure, so I tried using OCCT's power supply test. No test would last very long until the system would crash, with no BSOD, and restart. I took out my graphics card - no luck - then tried each stick of ram in different slots (testing 1 stick each time) - no luck - and also a different sticks of ram - no luck! So just to make sure, I'm hopefully going to do a memtest on my original 2 sticks of ram for 6-7 hours and see how that goes.

Note: I'm having a feeling that perhaps the power supply crashing was due to overheating? Not sure.

So at this time, if the memtest is still running after 6-7 hours, then I'm going to be pointing my finger mainly at my old PSU, although I will keep this 2nd PSU in for a while to see if the system does anything suspicious again.

Thanks!

 

[johjoh]

Here's the results of the memtest:

How does it look?

 

[efx8]

Is that with 1 stick of RAM in? If so have you tested the other and in the same slot?

 

[johjoh]

That's with 2 sticks of ram at once, in adjacent slots

 

[johjoh]

Okay, just got another restart with no BSOD. This is really getting on my nerves! cmad:

I'll stress my GPU and CPU to see if they're unstable.

 

[johjoh]

Thought this was pretty noteworthy:

Using OCCT's ("CPU : OCCT") CPU testing,

With test mode set to "Small Data Set" and "Use all Logical Cores" ticked, the PC restarted with no BSOD usually around 62C.

With test mode set to "Large Data Set" and "Use all Logical Cores" ticked, the test stopped right when it reached 62C and said "Stopped: Error Detected". I'm going to post my graphs, that were produced from the test, in hope that it will aid whoever is trying to help me fix my problem. :beerchug:

According to the AMD spec sheet for my processor, the max safe operating temperature is 62C!

Please find the graphs in the link below:

ImageShack(TM) slideshow

 

[greenbrucelee]

check the clips on the cooler, apply new paste. You never want to be anywhere near the max temp.

also clear out any dust and get a better psu

 

[johjoh]

Okay, I'll give that all a shot and do another test when I'm done.

Just on the note of temperatures: my GPU at max stress runs at around 72C.

I would really like to leave purchasing anything to the very last when the fault is completely identified - would getting a new PSU fix these problems (or the majority)? -> And if so, would that basically mean that my pc isn't getting enough power? Because the PSU I have in the PC atm is the borrowed one and it works fine

:ermm:

 

[greenbrucelee]

its not just about enough power its about quality whitch the antec isn't. what is the other psu make?

gpu should have lower temps than that.

 

[johjoh]

It's a coolermaster (RS-500-PCAR-A3)
And just to add, I live in Brisbane, Australia, and the ambient temp is around 27C

 

[Tyree]

The CoolerMaster RS-500-PCAR-A3 is a very poor quality PSU.

 

[johjoh]

Oh... Are brands like CoolerMaster and Antec bad brands in the PSU market? or are their products varying in different qualities with respects to price? I thought all their products would be above-average?

Also, I've cleaned some dust around the case and the cpu + cooler and cleaned then applied new thermal paste (Arctic Silver). The CPU stress test (Large Data Set) of OCCT stopped at around 52C (average of 2 tests). Graphs are attached in links below if you desire them or not.

Test 1:ImageShack(TM) slideshow

Test 2:ImageShack(TM) slideshow

 

[Tyree]

To insure top quality we recommend SeaSonic-XFX-Corsair (not the CX-GS-M Series).
http://www.techsupportforum.com/forums/f210/power-supply-information-and-selection-192217.html

 

[efx8]

That's with 2 sticks of ram at once, in adjacent slots

You should test RAM sticks individually, the fact that there are errors coming up is pointing towards the culprit anyway. Test them individually and see if one get errors and the other doesn't.

 

[johjoh]

Okay, I'll try that! I'll post back with pics of the test when I'm done

Thanks everyone