[happyrock]

For general information, the 5v, 12v, and 3.3v rails need to be running within 5% of spec, or your power supply is defective as defined by industry standards. Voltages may need to be off by less or more than that to cause problems, depending on hardware used

I had a client install speedfan and it reported 11.02 on the 12 V rail ...I ,of course , suspected the PSU....when the dust settled it was in fact the video card that was causing the problems...after installing video card problem quit and speedfan still reports 11.02 to 11.05 volts....this is a intel mobo with a 845 chipset ..still running I guess...he has not called back yet

[Advertisements]

Yeah, the only place to get 100% accurate readings is the BIOS. Sandra, Speedfan, Everest, and PC Wizard, can misreport. Of course, if the sensor on the board itself is misreporting, then there's no help for it.

[Kurenai]

Yeah, the only place to get 100% accurate readings is the BIOS. Sandra, Speedfan, Everest, and PC Wizard, can misreport. Of course, if the sensor on the board itself is misreporting, then there's no help for it.

[happyrock]

Yeah, the only place to get 100% accurate readings is the BIOS. Sandra, Speedfan, Everest, and PC Wizard, can misreport. Of course, if the sensor on the board itself is misreporting, then there's no help for it.

all the tools get their data from the same places..the sensors...that includes the bios...I have run motherboard monitor...everest...and speedfan...all at the same time to see if there were differences in their reporting.....they were the same numbers with the only difference I could detect being the refresh rates of the apps...once the system settled down and i quit doing things they all reported the same info....

Edited by happyrck, 10 November 2006 - 07:26 AM.

[Kurenai]

No, different apps can sometimes list our different values. PC Wizard, for example, can sometimes be wildly inaccurate on the 12v rail, even to the point where it halves the actual value. Each of these programs does not read from the sensor in exactly the same way, and it causes some variation among the readings. Trust me, you look at ~15 machines per day, and have people run these programs, you start to notice patterns.

[happyrock]

I never used PC Wizard....but launch the 3 I mentioned at the same time and test them yourself

Edited by happyrck, 10 November 2006 - 10:08 AM.

[burnoutnotfadeaway]

I took out the graphic card(NVidia FX5200) and swapped it with an ancient agp 32mb graphic card. After leaving the machine off for a few hours it booted, I ran speedfan, and this was the result:
The first picture is after the machine was running for about 10 minutes, the second was taken about 5 hours later.


The machine does cut out with the old CRT monitor, but it tends to run much longer with the CRT than the LCD. The LCD monitor runs off a seperate power supply. I've had the same problem running the computer on two seperate power circuits in 2 seperate buildings.
Unfortunately the motherboard doesn't have onboard graphics so I always have to attach a graphic card to run the monitor and see if it's running properly. The voltages in the speedfan tests seem to be not what they should be, the readings from the BIOS can be seen on the previous log page, although that shouldn't make a difference as SpeedFan takes the recordings from the same sensors. I think the '-5V' reading looks alarming, although I don't have much of an idea about voltages, so please help!
Thanks for all your help

[happyrock]

[quote name='burnoutnotfadeaway. I think the '-5V' reading looks alarming, although I don't have much of an idea about voltages, so please help!
Thanks for all your help
[/quote]
you never did post the -5V readings....
I reread the posts from the start and I keep leaning to the PSU ...BUT you say you tried three other psu's...sooo that leaves video card or the mobo itself...maybe someone else has has something else to try...I'm at a loss right now

[burnoutnotfadeaway]

Well the same system was tested using a high-end PSU that was taken out of a gaming-PSU to test the power supply in my system and it made no difference. If all else fails, do you have any idea where I could buy a relatively cheap Socket 462 AMD motherboard to test my processor and run it for a little while before I can afford to upgrade to something more preferable? Thanks

[Kurenai]

Actually...I'm curious about processor temp. It's the only thing that hasn't been changed...I'm not so good with speedfan, so unless happy can pull a cpu temp out of that screenshot, pick up Everest free edition from here: http://www.majorgeek...wnload4181.html
Go to the Sensor item, and let us know what it's showing for cpu temp?

[burnoutnotfadeaway]

At boot-up it shows at 25C when I boot it and go into BIOS, within about 3 minutes it rises up to 40 then up to 45 when Windows is running IDLE. I'm not entirely sure what the safe temp is for this processor, but even I think that seems a little high for running totally idle after 5 minutes.

[Kurenai]

Those processors top out around 70C, so it's probably not a processor temp issue. It could just be a damaged processor. You've changed absolutely everything else, right?

[burnoutnotfadeaway]

Yep, literally everything. Ran it out of box, changed PSU's, all cables, tested with 2 monitors, changed harddrives, disconnected cd-rom drives, as well as tested other drives, tried other floppy drives. All other testing hardware all came from other working systems. Since I took the FX5200 graphic card out and put the old NVidia GeForce 3 card in it hasn't cut out yet, and it's been running for about 12 hours now. That may/may not be progress as it can run one day for a 12 hours, then I try to turn it on the nexy day and it cuts out. Very unpredictable.

[burnoutnotfadeaway]

Bad news, the computer cut out again tonight. Strange though, with the FX5200 graphic card when the system cut out it continously cut out when it got to the Windows loading screen unless I left if for a few hours. With this low-end graphic card when it cuts out it boots back up again as normal. Usually when it cuts out I have to turn it off for a few hours then start it up again or it cuts out when Windows begins to load. At this point I'm thinking I need a new Socket 462 motherboard...any idea which to get?

[Dan1887]

sounds like maybe an agp issue, your cpu temps are normal anything over 60 cause for concern, one thing i did notice on the speed fan report is your mobo is running at over 40 degrees, increasing in temp the longer it is on. Safe mobo operating temp is considered to be from 25-35 and sometime a little over 35 depending on case cooling, there is a good chance that after running for a sustained period ur agp is experiencing a failure and shutting down. The newer video card runs faster and hotter and takes up more power which may account for the older card's ability to stay on longer.

[burnoutnotfadeaway]

Thus meaning I need to replace the motherboard or the graphic card, or both? I've got the power supply blowing hot air out the back, a case fan I added to the back of the case sucking air out, the processor fan blowing air onto the processor and two side fans on the side of the case which are built-in. So perhaps it is a motherboard fault if with all that cooling it's still getting that hot? Thank you