Hi guys, had a regular in some time ago to upgrade their computer - done. She contacted her sister who eventually contacted me to repair her system. Boy, what a sight - Have you ever seen the inside of a hoover bag when full, well this was worse!!!!!!

Anyway, completely serviced it & found out why it was slower than normal (asides from the dust accumulation), it had burnt out a ram stick(found out after it wouldn't boot up).

After rebuilding, booted up & within seconds of logon screen appearing, it automatically shut itself down. Never come across this before, so took side panel off & rebooted. Cpu/HSF & case fans spun up so assumed that they are ok. Again it shut down. Checked gpu in another system, card ok.

Made an assumption that the psu is dying as it was clogged with dust & told customer this. Am I correct in thinking psu problem?

If not, any ideas as to what causes this?

This is NET-20 & running WCG, but offline at this time, but have built 2 new X2 rigs, both running WCG. Also got ram, tb drives (boy do they take an age to format) cpu to upgrade server, but couldn't get mobo(not in stock) - undecided as to whether or not to upgrade or wait for an AM3 mobo!!!

Flaky PSU will definitely cause this issue as it will overheat and cut power.

Thanks User312. Wasn't 100% sure, so thought I better seek advice. Unfortunately for them, the only spare psu I've got left is a 20 pin & require a 24 pin. It's up to them how they proceed. It's a P4HT & I hope they get it sorted. The lad wanted a better computer & I had to bite my tongue as I know that his parents already bought him one (from guess who - it's NET-18 & will start running on Xmas day).

If they decide not to, I'll put in a bid for it, repair & upgrade ram & sell it on.

They've ordered a new psu & ram, & hopefully, it will solve the issue. I've already purchased them but will have to wait until Friday before I can do anything (they're away). Yeeeeeeeeeeesssssssssssssss, getting closer to that magic target I set myself - 50 cores

They returned early & dropped the rig off this evening. I've installed a new & better PSU (the one they had was only 300w). Here's the weird problem - It boots up & shutdowns still. However, this time, the logon goes further, it completes the desktop icons & notification area, then shuts down.

Have checked bios & hardware & everything's ok - could this be something that Antivirius 2009 left on the system so that if it couldn't find it's momma, it shuts down?

HEY SIRIUS B
THIS ONE IS A BIT ABOVE MY TALENTS BUT I WOULD START WITH A PUPPY LIVE CD.
IF IT SHUTS DOWN WITH IT THEN ITS HARDWARE. IF IT RUNS PUPPY OK THEN ITS ON THE HD.
THEN I'D START SWAPPING OUT MEM CHIPS THEN HD'S ETC
JUST A THOUGHT. (AND THE PUPPY PUSHER STRIKES AGAIN !)

I JUST HAD A SIMILAR PROBLEM IN REVERSE.
ON AN OLD DELL 350 PII I WOULD SHUT IT DOWN, THEN 5 - 10 - 15 EVEN 20 MINUTES LATER IT WOULD BOOT BACK UP.
ALL ON ITS OWN. I TRIED ALL SORTS OF IDEAS, ALL WITH NO JOY. LIKE SWAPPING POWER SWITCHES, POWER SUPPLY ETC.
THE WEIRDEST PROBLEM I HAVE SEEN.
I FINALLY SOLVED IT BY FLASHING THE BIOS.

Zappdog, you're brilliant - WHy didn't I think of that!!!!!! OK, will try it shortly as I'm busy building another rig (X2).

Thanks Zappdog. Booted up with Puppy 3.01 live & it lasted 4 min before shutting down, so it's hardware related. Stripped it down again & rebuilt. Same problem so rebooted & entered bios. Checked hardware monitor & found it was hitting 94 degrees. Couldn't understand that so stripped down again, cleaned HSF/CPU & reapplied thermal paste.

Rebooted & temp down to 75 degrees. Can't understand this as I used AS5. Fans going full blast - Is it possible that the thermal cutout has burnt out? I'm thinking this because I've installed PC Wizard on the rig & it's showing CPU temp @ 36 degrees & mobo temp @ 86 degrees.

I'D SAY THATS A GOOD PLACE TO START TO LOOK AT.
THE MB SHOULDN'T BE HOTTER THAN THE CPU.
CAN YOU FEEL WITH YOUR HAND WHAT SECTION OF THE MB GETS HOT ?
AFTER 4 MINUTES THERE SHOULD BE A NOTICEABLE HOT SPOT.
IF POSSIBLE FEEL THE CPU HEATSINK TO CONFIRM ITS NOT GETTING HOTTER THAN THE 36 DEGREES.
IT SURE SOUNDS LIKE A MB PROBLEM.
I'M NOT SURE IF IT STILL WORKS ON NEWER BOXES BUT ON OLDER STUFF, YOU COULD PULL THE BACKUP BATTERY AND HOLD POWER SWITCH DOWN FOR 10 SECONDS. THIS USED TO CLEAR BIOS MEMORY, I THINK. IT USED TO SOLVE NUMEROUS GLITCHES AND THAT MIGHT BE WHAT IT IS.

Thanks, already cleared & reset bios. With the amount of dust buildup that was in this rig, I'm pretty sure that something got burnt out besides the psu & ram. when I first checked this out, the noise was too high for my liking but Luke stated in was like that for a long time & he had got used to it. Rig totally clean & running but fans running at full tilt, so thought of installing Speedfan & controlling them that way. I'm only worried that it could increase mobo temp if I reduce fan speed - I'll try it & see - I have the rig until tomorrow evening.

edit: - Because of the high mobo temp, I've reduced boinc's usage to 50%

HEY SIRIUS B
DAMN ! THATS A FRUSTRATING PROBLEM AIN'T IT?
IS THERE A UTILITY PROGRAM FOR MEASURING CURRENT DRAW ?
AMPERAGE OR WATTAGE. I DON'T KNOW OF ANY MYSELF.
I WOULD HOOK UP AN AMP METER, EITHER DC OR AC
I BET THE WATTAGE USAGE IS PRETTY HIGH FOR THOSE 4 MINUTES.
BUT THAT WOULDN'T TELL YOU WHERE THE PROBLEM LIES, ONLY THAT ITS DRAWING MORE THAN IT SHOULD.
MY BROTHER SAYS THE MB IS TRASHED AND I'M STARTING LEAN HIS WAY THE MORE I HEAR.
YOU MENTIONED ONE OF THE MEM CHIPS WAS FRIED. ARE THE OTHER CHIP(S) OK ? I CAN REMEMBER AN SDRAM CHIP THAT GOT FRIED AND IT TOOK OUT ONE BESIDE IT. I BURNED MY FINGERS ON IT WHILE PULLING IT OUT.
ANOTHER THOUGHT IS SOMETHING I USED IN ELECTRONICS TROUBLESHOOTING. A COLD SPRAY REFRIGERANT (OR C02) ON THE PARTS WILL SHOW WHICH ARE HOT. FROST UP A SECTION OF THE BOARD AT A TIME AND SEE WHAT MELTS OFF FIRST.
WELL THATS ABOUT THE EXTENT OF MY IDEAS ON THIS ONE.
GOOD LUCK WITH THAT ONE.
LET US KNOW WHAT FINALLY WORKED.

A THOUGHT ON FANS. MOST OF THE FANS IN MY BOXES ARE YEARS OLD. WELL PAST NORMAL LIFESPAN. WHEN THEY START SQUEALING AND MAKING NOISE, I PULL THEM OUT, THEN HOLDING THE FAN WITH THE BLADE HUB FACING DOWN I FILL THE
CUP OF THE HUB WITH WD40, GIVE IT A SPIN THEN DRAIN OUT THE EXCESS. A WIPE DOWN AND THEY ARE GOOD FOR ANOTHER YEAR OR SO. FILLING THE CUP SUBMERGES THE BEARINGS. IT WORKS WELL IN A PINCH.

Thanks. I've installed Speedfan & set to Automatic - gone into fan settings, & set them at 50% min - 75% max - that killed the noise. CPU temp steady at 35 degrees but mobo up 1 degree to 87. Will leave running overnight.

Can't suggest he gets a new board as his dad has bought him new rig (NET-18 ) which he's getting Xmas Day. So tomorrow evening's going to be a bit tricky for me.

I agree with your brother - I'm pretty sure the mobo is on the way out.

Well it's been running continuous since yesterday afternoon & cpu temp still @ 35 degrees & mobo temp @ 87 degrees. No problems with any program so far. I think it's just the temp sensor gone awry.

Another situation that's got me bugged..................Workbench. Posted this photo on Crunching Room Thread.

The rig on the right keeps shutting itself down after approx 5/10 minutes of running.

The specs are identical to the one in the middle....
Advance Slimline LCD Case - 480w PSU - Maxtor 250gb HD - LG 22x DVD-RW - 2gb PC6400 - AMD X2 5000+ - Gigabyte GIG M61PME_S2 Mobo - XP Pro SP3.

It's done this constantly since yesterday afternoon, but since rebooting an hour ago, it's remained running so far..........Weird! Any ideas?

It's been running continuously since 10pm last night - probably gremlins at work

This is a new one on me. Andrea noticed it several days ago. It still shuts itself down - system itself is pretty clear of any short, everything seated correctly in it's slots, all cables correctly fitted. Checked event viewer but could not find any incident to cause power down. As it continually bugged me, I posted over at S@H & Fred W came up with a solution.

Remove factory paste & reapply AS5 which I did & no more problems. Did the same for the other 2 & noticed temps on all 3 down by 4c. Up to now, I've never had problems with stock coolers.

HEY SIRIUS B
DARN GOOD ADVICE. I'M GLAD THAT SOLVED THE PROBLEM.
SO NOW WE KNOW WHERE THE TEMPERATURE DIFFERENCES CAME FROM.

I NOTICED THAT A COUPLE OF OEM CPU'S I'VE PULLED OUT HAD VERY LITTLE HEAT SINK COMPOUND AND WHAT WAS THERE WAS PRETTY MUCH OVERCOOKED. MOST LIKELY MACHINE DISPENSED. I'VE ALWAYS APPLIED THE STUFF FAIRLY LIBERALLY WITH NO ILL EFFECTS, IN FACT CLEARING UP A GLITCH. I ALSO FEEL THAT IT SHOULD BE CLEANED OFF AND RE-APPLIED EVERY YEAR OR SO. IT DOES BREAKDOWN WITH TIME AND HEAT. AND A CRUNCHERS BOX WOULD BE MORE PRONE TO DRYING IT OUT. MINE RUN FULL TILT 24/7 AT TEMPS A FAIR BIT HIGHER THAN NORMAL. I TRY TO DO IT WHEN I DO A DUST CLEAN OUT. (NOTE TO SELF: PICK UP SOME PASTE, I'M OUT)

I quite agree Zappdog, I always reapply paste every year on annual servicing. However, these rigs are not even 3 weeks old! Up till now, I've never had problems with retail CPU/HSF's, & always found them to be sufficient. Only thing I can think of is that due to their financial situation, AMD is cutting back on their quality control.

Unless.............The middle & right rig have identical boards/cpu's/psu's/ram sticks/drives & the only difference to the left one is that it has a workstation mobo & X2 6000 cpu, everything else is identical, leaving me to think it could be the gigabyte boards.

I always used to use Gigabyte until having numerous problems back in the late 90's when I switched to Asus. I hope they are not starting to have similar problems again. I only bought these boards as my brother recommended them & I got a very good deal.

HEY SIRIUS B
I DON'T KNOW ABOUT THOSE ASUS MB'S. I'VE BOUGHT 4 GIGABYTE MB'S FOR INTEL AND AMD.
ALWAYS LIKED THEM. NEVER HAD ANY PROBLEMS WITH THEM IF YOU DON'T COUNT DROPPING A SCREW ONTO A WEEK OLD MB AND FRYING IT.