Sign in with
Sign up | Sign in
Your question

New homebuilt - Random Crash **PLEASE HELP**

Last response: in Systems
Share
June 30, 2007 5:40:11 AM

I've recently built a new PC. It boots fine runs fast but will randomly shut down like someone pulled the plug.. I've always related these kind of crashes to an overheating problem. Most of the time it runs for 1HR + before crashing and wont crash idle but when gaming, etc.

The PC:
ASUS P5N32-E SLI LGA 775 NVIDIA nForce 680i SLI ATX Intel Motherboard
Intel Core 2 Duo E6600 Conroe 2.4GHz 4M shared L2 Cache LGA 775 Processor
ZALMAN CNPS 9700 NT 110mm 2 Ball Ultra Quiet CPU Cooler
Antec Performance One P180 Silver cold rolled steel ATX Mid Tower Computer Case
EVGA 320-P2-N811-AR GeForce 8800GTS 320MB 320-bit GDDR3 PCI Express x16 HDCP Video Card
CORSAIR XMS2 2GB (2 x 1GB) 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400) Dual Channel Kit Desktop Memory
HIPER HPU-4S730-MS ATX12V v2.2 730W Power Supply 100 - 240 V

All the above is new. I re-used my old Hitachi 80G SATA HD, DVD and CDR Roms.

What I've done:
I've used a few different programs to keep an eye on my temps (nVidia Monitor, RivaTuner and Speedfan. The CPU seems to idle around 32'C and the Vid Card around 60'C (after I turned the fan speed to 100%). Every time I crash I try to feel for hotspots but dont seem to find any so I dont thinks its a heat problem. My case runs 3 120mm fans (Top, Rear, and lower chamber) and the PSU has a front and rear fan.

I was running 4GB of the CORSAIR XMS2 and my BIOS found all 4GB but XP would ony show 3GB. I read some stuff on why that was (32bit OS) and even found a guy who was having the same problems as me and drop his comp to 3GB and it went away. Pull a stick out and ran 3GB, still shut down after an hr or so. Took out another and rag 2GB and still the same problem.

Pulled CPU fan off whipped thermal paste clean and applied Artic Silver and re-mounted the fan just to make sure the CPU was keeping cool.. I've never used a fan this size before so I didnt want that to be the problem.

And last I thought maybe a PSU issue. Rearranged some of the power, made sure everything was seeded tight and HD,DVD and CD had its own line with a fan on the CD, DVD and Floppy lines. Vid Card is using its own PCI-E power line. I used Speedfan to check the voltage but I have no experience with voltage, amperage, etc. Here is a quick view:

Vcore1: 1.25V
Vcore2: 3.20V
+3.3V: 0.00V
+5V: 4.95V
+12V: 12.03 - 12.10V
-12V: -16.97V
-5V: -8.78V
+5V: -4.87V
Vbat: 3.01V

Ran a stability test for 10 mins, 50 passes all completed. Ran idle all last night 11:45PM - 8:00 AM. Was going to run a Prime95 stress test but wanted to get the system stable first.

Please help.. PSU? MEM? Overheating?

Quogg

More about : homebuilt random crash

June 30, 2007 5:45:22 AM

If it ran for 8 hours fine, then I'd say it's not overheating. HOWEVER, run dual Prime95's for awhile and watch temps and see what happens. If temps are good AND no crashes, then it's not overheating.

Also run Memtest for at least a half hour and see if any errors. If not, then RAM is fine.

Does this only happen during gaming, or certain times?

If RAM is fine, temps are fine, Prime95s doesn't crash.....then I'd say mobo or PSU. If at all possible, try a different PSU. If it's fine, then mobo is probably the culprit.

All of this is assuming you haven't overclocked. If you have, then I'd say unstable OC. If no OC'ing, then go back to above points.
June 30, 2007 9:33:23 AM

Just to make sure that it's not a software problem, I would run defrag or a spyware tool in Safe Mode to see if the same thing happens.

But it looks like a hardware problem. I would suspect the power supply of being faulty.
Related resources
June 30, 2007 11:17:17 AM

Dude it's your power supply... look at the voltages you posted...
Vcore1: 1.25V ok
Vcore2: 3.20V ?????? not sure but i think it's ok it should be 3,3 i think...
+3.3V: 0.00V ----- no voltage in the 3,3
+5V: 4.95V -----ok
+12V: 12.03 - 12.10V ----ok
-12V: -16.97V ------ plus 4,97v??? wtf
-5V: -8.78V ------- 3,78v???
+5V: -4.87V .----ok
Vbat: 3.01V --- brand new...

It shouldn't even be working at all...
one odd thing is that the other voltages are ok
June 30, 2007 1:29:24 PM

No OCing, does crash mostly while gaming but has when not gaming, but thats usually right after it reboots or I'm running some test. It did crash last night running another stability test.

OK ran Memtest for an hr+ no errors. Ran Prime95 all night no problem (but the side of my case was off.)

I remember reading some about the +3.3V reading 0 being normal for some reason but did always question the others Xanxaz pointed out. Could the PSU get over worked after the PC is on and running a full load? Maybe it cant get my vid card enough power overtime or something? I'll try and find another PSU to swap out and test. Running another stability test right now, I was surprised it didnt crash during the prime95 test.
June 30, 2007 2:30:23 PM

YA that -12v rail at almost 17v is NOT GOOD :!: :!:

Try a different PSU.....a GOOD one.
June 30, 2007 3:14:40 PM

I was gonna say look for gremlins hiding behind your pc and pulling the plug out, but it looks like the PSU was the culprit after all.
June 30, 2007 10:32:21 PM

Crashed durning a 2hr stability test. Installed Asus Probe II error log:

ASUS PC Probe II

Alert Log Report

Number = 5
===================================================

[06/30/2007 at 05:24 pm] SB CORE Voltage Abnormal, 0.00 V
[06/30/2007 at 05:24 pm] SB CORE Voltage become normal, 1.54 V
[06/30/2007 at 05:44 pm] NB CORE Voltage Abnormal, 0.00 V
[06/30/2007 at 05:45 pm] NB CORE Voltage become normal, 1.23 V
[06/30/2007 at 05:45 pm] NB CORE Voltage become normal, 1.23 V

Is that the PSU for sure of could it be the MB?
June 30, 2007 11:38:02 PM

Don't put too much faith in speedfan readings. My voltage readings are the same and there is absolutely nothing wrong with my PSU (not saying yours is fine) However the ASUS probe readings are a little funny.. 0.00 voltage on your chipset and it still ran??
June 30, 2007 11:58:17 PM

Ok I'm going to swap out my PSU and hope that is the fix. Replacing with:

Thermaltake W0106RU Complies with ATX 12V 2.2 & EPS 12V version 700W Power Supply 100 - 240 V CE, CB, TUV, FCC, UL, CUL, and BSMI certified

Model
Brand Thermaltake
Model W0106RU
SPEC
Type Complies with ATX 12V 2.2 & EPS 12V version
Maximum Power 700W
Fans Silent 140mm Fan
PFC Active
Main Connector 20+4Pin
Dual +12V Yes
PCI-E Connectors 2 x 6Pin
SLI Support SLI Certified
Modular Cabling Support Yes
Power Good Signal 100-500ms
Hold-up Time 16ms min.
Efficiency up to 85%
Over Voltage Protection Yes
Input Voltage 100 - 240 V
Input Frequency Range 47 - 63 Hz
Input Current 10A @ 115V, 5A @ 230V
Output +3.3@30A,+5V@28A,+12V1@18A,+12V2@18A,+12V3@18A,+ 12V4@18A,-12V@0.8A,+5VSB@3.0A
MTBF >120,000 hours
Approvals CE, CB, TUV, FCC, UL, CUL, and BSMI certified
Features
Connectors 1 x Main connector (20+4 pin)
1 x 12V(4+4-pin)
7 x peripheral
6 x SATA
1 x Floppy
2 x PCI-E
Features Mirror Finish
Modularized Cable Management to eliminate clutter and improve airflow inside the case
Next generation four +12V rails(12V1, 12V2, 12V3, 12V4) supports high-end graphic card and PC system (combined loading of 56A)
Independent Voltage Circuit: offers unflappable current delivery under heavy load and makes voltage output more stable
Extremely good voltage regulation (±3%): provides steady voltage for system
Industrial grade components (capacitor, transformer, etc)


Anyone had any problems with this one?
a c 83 B Homebuilt system
July 1, 2007 12:12:45 AM

What are the GPU temps when it crashes? Could the 8800 be getting too hot? They do run hot, so 80c should not be alarming.
If memtest runs, the memory is ok.
If prime95 or TAT runs, then cpu, motherboard, and heatsink are ok.
That leavs the PSU or vga card. Both will be more stressed while gaming. Try replacing one or the other if you can.
One last possibility; You might have a marginal PSU cable, or even a short. Try using a different pci-e cable.
July 1, 2007 12:35:52 AM

woah, after reading this topic for the hell of it..i decided to start speedfan. im getting:

By r3trdedfkr at 2007-06-30

WTF?!? btw, im getting a new psu soon because this one's fan stopped working on its own :roll: do you think this is safe to be running? im having absolutely no crashes or heat problems with my cpu and gpu OCd... im kinda nervous now
July 1, 2007 12:35:52 AM

Speedfan does not label all the voltage readings correctly. You just got to look at the ones that look reasonable and make sense and ignore the rest. Make sure your vCore voltages and RAM voltages are set correctly. Go into the bios and manually set then. The vCore stock voltage should be 1.325V, I think. Speedfan's reporting it as 1.25V, which is low. However, try using CPU-Z and see what voltage it reports too. And set you RAM voltage to the manufacturer specs, which is likely 1.9V or 2.1V, depending which memory you have. Do this before you go and replace the PSU.
July 1, 2007 12:37:46 AM

Quote:
Speedfan does not label all the voltage readings correctly. You just got to look at the ones that look reasonable and make sense and ignore the rest. Make sure your vCore voltages and RAM voltages are set correctly. Go into the bios and manually set then. The vCore stock voltage should be 1.325V, I think. Speedfan's reporting it as 1.25V, which is low. And set you RAM voltage to the manufacturer specs, which is likely 1.9V or 2.1V, depending which memory you have. Do this before you go and replace the PSU.


in cpuz it reports my voltage right (1.2v undervolted =D)

July 1, 2007 2:12:20 PM

Ok I set my Vcore in the bios to 1.35V and the mem to 2.1. Once I booted and started Asus Probe II - Vcore still shows at 1.28. Rebooted bios still set at 1.35. And now I'm getting errors on my VHT (?) Voltage from Asus Probe II:

[07/01/2007 at 09:24 am] 1.2VHT Voltage Abnormal, 0.00 V
[07/01/2007 at 09:24 am] 1.2VHT Voltage become normal, 1.25 V
[07/01/2007 at 10:06 am] 1.2VHT Voltage Abnormal, 0.00 V
[07/01/2007 at 10:06 am] 1.2VHT Voltage become normal, 1.25 V
[07/01/2007 at 10:09 am] 1.2VHT Voltage Abnormal, 1.54 V
[07/01/2007 at 10:09 am] 1.2VHT Voltage become normal, 1.25 V
[07/01/2007 at 10:09 am] 1.2VHT Voltage become normal, 1.25 V

Will run a stability test to see if change the vcore and mem voltage helped. Thanks and I'll post the results.
July 1, 2007 2:15:03 PM

ohhh and 8800 runs around 59-65'C idle to load.
July 2, 2007 12:19:50 AM

Still crashed.. Going to replace PSU.
July 11, 2007 1:52:04 AM

Put in a new PSU (listed below) set everything back to auto in BIOS. No more crashing.. WOOOHOOO..! ONLY GET GOOD PSU's...!

SeaSonic S12 Energy Plus SS-650HT ATX12V / EPS12V 650W Power Supply 100 - 240 V UL, CE, CB, TUV, FCC
July 11, 2007 1:52:16 AM

THANKS ALL FOR THE HELP
a b B Homebuilt system
July 11, 2007 3:06:57 AM

Hey, even though I did not contribute on this thread, thank you very much for letting everyone know what resolved the problem.
That is a nice touch to round out an issue, get some advice, and get everything working again, let the posters who helped know what solved the problem.
!