For the love of God..............

[ghost gamer 0]

hey guys, I really need your help on this one. I migrated my pc over to pci-express. The following parts came over to the pci-express and are known working parts:

fx 53

600 watt enermax(35 amp on 12 v rails)

1 gig mushkin ram(2 sticks of 512 pc3200)

sound blaster audixy 2 zs

western digital 120 gig/7200 rpm 8 mb cache

new parts for the pci express upgrade:

asus a8n5x

powercolor x1900 xtx

Everything went fine until I tried to install windows xp home edition. The install will reach the point where it reboots your pc for the very first time. Then wham, my pc just shuts off. No beeps, no error messages, nothing. Sometimes I will see the windows splash screen and once I actually seen the screen after that where it shows the over all progress of the installation.

Here is what I have tried so far:

1)-ran memtest 86-2 passes-standard tests-0 errors

2)-ran western didital diagnostics,both quick scan and full media

scan-0 errors

3)-microsoft article id #305681 describes my symptoms to the letter

and said the fix was to upgrade to newest bios or hit f7 during windows setup

and change the HAL to standard pc instead of ACPI HAL. The reason for the

shutdown according to the article was that the BIOS was notifying the system

the pc was too hot to run. I also cleared my cmos. Upon restart, I got

a "cmos-checksum error-defaults loaded" error message. According to Phoenix

Technologies that is possibly the result of a weak battery.

Silly question, but I need

to rule stuff out: Is it possible the cmos battery is weak and when the install gets to that point I mentioned above, the cmos battery craps out causing my shutdown?

Well, I flashed my BIOS to version 902 and still shutdown occurs. I also

tried changing the HAL, no effect.

4)-Temps on the cpu ranged from high 40's to high 30's, per BIOS

depending on whether I had my window fan in or not. Last night after

running all the tests and so forth which was maybe 4 hrs of continuous

running, the BIOS cpu temp was 39. The shutdown still occurred.

So I checked the heatsink of my zalman cooler. I felt around the base and it was by no means

hot. Oh one more thing as a side note. I checked the temp in my room and the temp in the BIOS upon start up and found this: room-22 / BIOS cpu-37.

When installed the cpu, I used maybe a bit more than a bb of artic silver

5 and placed it squarely in the center of the heatspreader. I then sat that zalman

heatsink/fan directly on top of it and rotated it left and right. I of course completely

cleaned the heatspreader/heatsink off before applying new compound. Is there anyway I used to much artic silver 5 and am somehow still triguring a thermal shutdown? The zalman cooler is of such a design that there is no way the heatsink could get hung up and not be sitting squarely on the heatspreader.

5)-I tried another copy of windows xp home edition(both copies are legit)

and still shutdown at the same place.

6)-Well, here we are. Power supply. I know the x1900 xtx taxes the system. ATI states 30 amp on 12v rail. My psu provides 35 amp. I figured that since it

was a brand name with more than the required amperage, I was ok. I am starting

to wonder about that. Since my psu is not pci-express compliant, I had to use s 20 to 24 pin adapter for the motherbaord and a 4 to 6 pin adapter for the x1900 xtx. I know this is not ideal as these adapters are creating weak points, resistance in the power flow. Not much,theoritcally, but still. I also realise that the 600 watt rating is peak and that power supplies lose efficency as they get hot. I read somewhere that the x1900 xtx draws like twice as much juice as the x800xt p.e., the card its replacing. Hmmmmm .......

What do you guys think? PSU?...............anything else I should check?.....common setup mistakes I could have made?....any help is greatly appreciared.......

Edited April 9, 2006 by ghost gamer

[=warcloud= 0]

I would of thought your psu is plenty of power for sure. Well if all you have upgraded is the mobo & gfx card i would be thinking it's one of those two parts that maybe causing the issue? Double check all connections, what do the manuals say. Check your BIOS. Could try Clear RTC RAM, clear CMOS. Check fans working on cpu gfx card etc. Best bit of advice is to keep cool & calm & positive, hope you sort it out mate.

[CrowmanUK 0]

Cant see any problems with the way you installed the cpu, temps dont seem to be high, check in the bios to see what temp its set at to cut out, it might be set way too low. I'd also try your ram sticks one at a time, I had a real swine of a time installing windows on the kids pc, turns out it was a dodgy stick of ram, I'm not saying that this is your problem but it doesnt hurt to rule it out. If the battery is on the way out you should be able to see that cos the date in the bios would keep resetting itself, it'll never keep a current date. Give it another try mate.

[ghost gamer 0]

Thanks guys...all fans are spinning. The shutdown temp setting on the mb is a possibility. I didn't see a setting for telling the pc when to shutdown due to thermal conditions. The only setting related to thermal that I saw was one where I could specify at what temp the cpu fan would kick up to high speed. I was running my zalman cooler with qfan enabled, but it would run the fan way too slow so I disabled that and the fan appears to run at full speed.

I presume since my pc goes through post fine everytime that the cpu is ok? I am not sure how to test the video card. I don't know anyone who has pci-express and since the a8n5x has no onboard video, if you yank out the video card you get a quick reminder on boot that its missing.

I have not tried testing the ram sticks one at a time. I figured if they passed memtest86, they were good. But none the less, a good idea and an easy thing to do.

Is my shutdown a symptom a bad motherboard would give?

Ok, bear with me I have a theory and want to know if it holds any water or if I am reaching for straws. I suspect the power supply. Heres why. It has 2 12v rails and according to the manual, these rails are independent and provide seperate power to the cpu/mb, drives ect.

Since the pci-express/x1900 xtx really draw on the 12v power (ati says 30 am on 12v rail)and my psu has two 12 v rails one with 18 amp and one with 17 amp, is it plausible that the x1900 xtx is trying to draw on one of those 12 vrails and not getting enough juice? And if so, when that happens, the psu notes an overdraw/overload on that rail and thinks there is a short and trips a circuit to shut the pc down?

Perhaps it does not work this way. I do have enough amperage on the 12 rail in total, 35, but if these rails are independent, each feeding different components, that might be my problem.

I was trying to figure out exactly what is taking place or trying to take place at shutdown. Windows setup has partitioned, formated my drive. It has then copied over its files. It then saves my configuration. Then it reboots to continue the install.

When it brings the pc back up, I presume windows is now reading data from the hard drive to continue the install? Is there a major draw of power at this point as compared to the demand that was on the system prior to that? Silly question, I presume I don't need to install any drivers since I am running an ide drive, is that correct? I know that you go to a gui,graphical user interface, upon the reboot I mentioned above.

Edited April 10, 2006 by ghost gamer

[ghost gamer 0]

problem solved. reseated heatsink/fan and installed new 480 watt antec psu. not 100% sure what the culprit was, but all is good for now.