Is it really overheating, bad sensor, or another problem?

[LeonDroid]

I received an iBUYPOWER NE690FX for Christmas of 2013 which has had the following modifications...

Stock 300 Watt PSU replaced with a Lepa N500
8GB OEM RAM (ADATA) upgraded to 16GB Kingston HyperX (4GB x 4)
Replaced NZXT case with Thermaltake V4 Black
Added Rosewill internal card reader & USB 2.0/3.0 Hub
120mm intake fans 1 front, 1 bottom, & 1 side
120mm exhaust 1 top, 1 rear

Other Specs:

Gigabyte GA-78LMT-USB3
AMD FX-4130 w/OEM Thermaltake H/S Fan
eVGA GeForce GT610
W/D 1TB HDD
DVD Burner

F/F to January this year I wiped the drive with DBAN, repartitioned the drive & reinstalled Win7 HP. While reinstalling some of my other software there was an awful noise which turned out to be the CPU overheating (I had to hold the power button to shut it down).
I replaced the OEM H/S Fan with the stock H/S Fan that came with an AMD FX-8350 using Arctic MX-4 thermal paste. Then used OpenHardwareMonitor & an earlier version of Core Temp to monitor temps. I began getting overheat alerts from Core Temp and it was putting the computer to sleep. So, I bought an Arctic Freezer 13 and installed it. For over 3 weeks it ran almost constantly never shut down for more than a couple of minutes while I used it to run everything it would ever need to do except internet (I actually tried to overwork it). Then one morning I noticed it had been put to sleep by Core Temp for overheating again. There was a notification in the taskbar telling me that core 3 was at 256 C (493 F), Core Temp had crashed ("stopped working and needs to close"). Since then it has happened several more times like that. The 2 times it happened while I was in the room, there was no audible warning. There's no discoloration indicating temps that high, the back of the motherboard was only warm to the touch yet it reported core 3 running at 256 C. That was with an earlier version of CoreTemp (v. 0.99.*). That computer now has the following: Core Temp 1.0 RC6, Open Hardware Monitor, Speccy & CPUID HWMonitor. I took several screenshots of 3 of these at bootup, while working, (compressing video files) and when finished.

Could the sensor have been damaged the only time I know for sure it overheated? Is there an issue with that early version of CoreTemp? Should I try another AM3+ motherboard (I have an ASUS M5A78L-M LX Plus). I'm out of ideas. I can give you just about any additional information that doesn't require me to go on the internet with that PC. Both side panels have been off since I replaced the OEM H/S Fan.

Final question...where do I go for help with a Windows XP problem?

 

[ignatzatsonic]

Then one morning I noticed it had been put to sleep by Core Temp for overheating again.........................Could the sensor have been damaged the only time I know for sure it overheated....................Is there an issue with that early version of CoreTemp?

I have very little experience with Core Temp.

Can it, on its own, shut down a PC that it thinks is too hot, even if it isn't true?

In normal operation, when things are "OK", do the temps reported by Core Temp match the temps reported by at least one other temp monitoring application?

If you disable or uninstall Core Temp completely, do you still have issues?

The temps of plus 400 Fahrenheit surely must be wrong if you see no sign of damage.

You describe "an awful noise which turned out to be the CPU overheating". I don't know that overheating by itself would necessarily make any particular noise. I'm wondering if there was some as yet undetected hardware failure at that time, which resulted in overheating.

I wouldn't expect the cooler swapping to make any significant difference, assuming they were properly mounted and the fan was spinning, as long as the PC is under a reasonable load.

Does your motherboard look perfectly normal, including capacitors?

Can you swap in another power supply?

 

[LeonDroid]

First, in the options menu of Core Temp, there are overheat protection settings which can be set to put the computer to sleep, hibernate, or shut down at a specified temp but it has to be enabled.

I forgot to post screenshots of the 3 system monitoring programs earlier. I'll do that now. They are all within a degree or two of each other.

Core Temp is the only one that shuts everything down in case of overheating so I hate to uninstall it. However, I removed the earlier version and then installed the newest release in case that was the issue.

I'm sure the noise I heard was an alert from the motherboard. The motherboard looks normal. However the capacitors to the left of the CPU are mostly obstructed by some kind of passive cooler and the Arctic Freezer fins. I gave it a once over when I swapped out the cooler because I needed to replace the brass standoffs with non conductive alternatives.
That actually reminds me that there was an issue with ESD. The computer would freeze or restart any time there was a shock when I touched it. Gigabyte tech support suggested using non conductive standoffs, fiber washers, and/or grounding the case directly to an outlet.

Swapping out the PSU at this time is not possible because I don't have a spare one that will work yet.

 

[Solarstarshines]

Well for one.......
Why is your CPU Vcore so high ?

 

[Layback Bear]

Also why is th 12V+ so low along with the VBat being very low.

Please put what power supply your system is using in your specs.

 

[LeonDroid]

Honestly, I've had tunnel vision on the temperature problem, and never thought to check the voltages, so I never even noticed.

Layback Bear: The power supply is the first on my list of upgrades...Lepa N500.
This one: http://www.newegg.com/Product/Produ...ET7180&cm_re=lepa_n500-_-17-494-002-_-Product
It's still under warranty if I need to replace it.

The computer was asleep again when I got up this morning. Attaching 2 more screenshots.

 

[Layback Bear]

Using more than one temp program at the same time can cause misreadings.
A poor power supply can cause all kinds of goofy things.

I'm not a AMD user so I really can't tell which temp program is best for you.

My personal choice for power supply is Corsair AX but they do cost a little more.

 

[LeonDroid]

The multiple hardware monitoring programs are suggested in this and other threads. They only vary by a degree or 2 and I attribute that to differing sample rates and times. Core Temp is the only one I usually run & the one that puts my computer to sleep. The only noticeable difference now is that it reports core 1 as the one overheating instead of core 3.

To be clear, are you suggesting I try a new power supply?
Newegg has this one on sale with a $20 rebate: CORSAIR CX series CX500 500W ATX12V v2.3 80 PLUS BRONZE Certified Active PFC Power Supply - Newegg.com

Or, I can save $10 with this one: CORSAIR CX series CX430 430W ATX12V v2.3 80 PLUS BRONZE Certified Active PFC Power Supply - Newegg.com I don't think the 5 Amps on 3.3V or the 6 Amps on the +12 will make much difference on this build.

I can't afford the AX series. A little more is ok, but they're 4x more.

 

[stormy13]

First ditch CoreTemp, there is no way that any CPU should be throttling at 77°F: 77°C yes (actually a few degrees lower than that). That or only run it by itself with no other monitoring programs running.

I would also seriously look at replacing that power supply if that 12V reading is in anyway accurate. Better to check with a voltmeter to be sure.

 

[LeonDroid]

Normally Core Temp is the only monitoring program I run, but I'll uninstall it and run just one of the other 2. there should be something for me to put a load on the CPU with tomorrow. I don't do any gaming except the downloadable casual games from sites like BigFish. Plus, a little audio/video conversion and/or compression so I'm not putting that much of a load on the processor at any time. The highest temp I've ever seen while doing that was 29 C (85 F).

I bought a new power supply. I should have it here, installed and tested by this time next week. I'll post then with the results.
The other will be RMA'd and the replacement saved for a spare.

 

[Layback Bear]

Great news. Let us know when you get your new power supply installed.

Just a reminder; always post your temps in (C).

 

[AddRAM]

The best way to see true temps and voltages is in the bios.

 

[LeonDroid]

It's been suggested that I pick one monitoring program instead of 3. So here are the results.
The first one is the computer working on video compression.
The second is after it finished.
The third is only included because of the huge jump in temperature that I don't understand because it was idling.
A few minutes later, while erasing the files I had just done, the cat jumped on the table, there was a static discharge when he touched the computer case and the computer went down. I'm thinking the power supply is not properly grounded. I wont be using it again until the new power supply is installed.
I welcome any other ideas.

I know there's a way to line these up horizontally, but I don't know how to do it...

 

[Layback Bear]

Temps look okay but the voltages are just horrible.
Do as AddRAM has suggested in post #12 and compare.
That will give you some idea if HWMonitor is working properly with you system.

 

[Solarstarshines]

Temps look okay but the voltages are just horrible.
Do as AddRAM has suggested in post #12 and compare.
That will give you some idea if HWMonitor is working properly with you system.

I think the main issue here is the PSU is underpowering everything 12+ is way below 10+volts no way should you be exceeding almost 2.volts on Vcore the PSU is a situation for you because it might burn everything up you keep using it

 

[Layback Bear]

I agree I would not use the computer with that power supply.

LeonDroid wait for the new power supply. Then get back to us.
Were open 24/7/365.

I would also install a new motherboard battery. (Cmos)
Vbat should be 3.0 volts or very very close.

 

[LeonDroid]

Temps look okay but the voltages are just horrible.
Do as AddRAM has suggested in post #12 and compare.
That will give you some idea if HWMonitor is working properly with you system.

The temp that concerned me was "TMPIN2 @ 117 C, or maybe that doesn't mean what I think it does. The computer idled from about 11:30 PM to 1:00 AM, so I wonder what caused the sudden temp spike.

I think the main issue here is the PSU is underpowering everything 12+ is way below 10+volts no way should you be exceeding almost 2.volts on Vcore the PSU is a situation for you because it might burn everything up you keep using it

I agree. I now believe the PSU is improperly grounded, which can cause all kinds of problems. These are not even the worst. I'm wondering if I should check this by running a wire from the chassis to the ground screw on an outlet. A replacement will be here next week.

 

[Layback Bear]

I will say again. I do not recommend turning on your computer until you replace the power supply.

With a bad power supply and a weak Vbat I would not trust any reading your are getting.
Replace the power supply and the motherboard battery and let us know when that is done and we can start again with any problems you might have.

 

[AddRAM]

Agreed, it`s not worth risking your hardware if you have a dodgy PSU

 

[LeonDroid]

Sorry it took an extra week to get this done. I had some unexpected things come up.
I got the new PSU, installed it, and put the computer to work doing what it will be used for most often. Here are 3 screenshots of HWMonitor at boot up, while idling, and working. It's the best I can do unless there is a program that can used to stress test the PC. The new batteries should be here in a day or two.