Power problem?

jynxed

n00b
Joined
Sep 22, 2004
Messages
5
Hey guys, first post. I tried to search around for the info I wanted, but I couldn't find a couple things I was looking for.

Basically, I got a new Tower + PSU + Motherboard + CPU + RAM.
The tower and PSU are called .... I-Cute. 400w power supply. Keep in mind that I paid $55 CDN for the combined.
Review of 400w I-Cute PSU
Then I picked up this motherboard from a mom and pops part distrobution store.
"MSI K8TM-ILSR S754"
And of course, my CPU was AMD retail 3000+.
and, RAM is PC3200, 512MB.
All the rest is pulled from a semi-new source.
Oh, and I also picked up some Artic Silver, 5 I think.

So anyways, here's my story.
I got all the aforementioned parts at once, came out to be about 600 after tax (not a bad deal), so needless to say I was pretty excited to put everything together and watch it blow away anything I've owned before. Heh. So when installing I wiped off the retail-heatsink's thermal pad. I thought I wiped it off pretty good, but it was just a household cloth. Then I went about applying the Artic Silver. I messed that part up. I guess I didn't think to read the instructions for applying the thermal paste to CPU's with a large heat spreader. Why? Because I emptied the canister of Artic silver onto the heat spreader, it was like, layered on there. I thought this was a good idea, at the time. Oh oh... I know better now.

Needless to say, I had a problem with my computer overheating. I caught the problem rather quickly, and finally "corrected" the problem. Upon taking the CPU off, I noticed that there was alot of thermal paste. Like, alot. There was even a gob on the side of the heat spreader. Needless to say, I cleaned off the heatsink and heat spreader as best I could. I got little plastic picks to get the hard to reach areas and everything. I worked long and hard with isoprophyll to get both pieces shiny clean. They looked fairly clean.
So anyway, I hooked it all back up and away I went, I had a stable comp. For a while.

Untill I was playing Quake3 and listening to MP3s in the background for about an hour or so.
My computer just rebooted. So I went and turned off the reboot-on-error in XP.
MACHINE_CHECK_ERROR.
I searched around, and couldn't find anything specific about this error, and my problem. Seems to happen when the CPU expects one thing, and gets another. Could be caused by overheating, or motherboard failure to provide requested functions.

So, I considered the possibility that it might be my power supply. How? Well, when I looked at my mobo+cpu temp, they were around 35-40, constant. Never saw them go higher, and never heard my temp alarm go off. So off I went to get a new power supply, and video card (needed to get it anyway =p). Well I put in an Entec 420w PSU, and it's a good product. I did the research, read the reviews, decent PSU.

Well anyway, I put that sucker in, and I got the same Voltage readings as I got with the I-Cute PSU. How annoying, see picture.

voltage.JPG


So, lastly... I got the chance to try Call of Duty. I ran that sucker in high detail (vid card i got = radeon 9800pro), and after a good 10 or 15 minutes, my comp just blue screens. I don't even see the error, it's just simply a blue screen, no text. Went into BIOS afterwards, and my CPU temp was around 58 degrees celcius.

So now my plan is to grab some fine sandpaper and sand down the heat spreader a little, and the heatsink, so that it's a good copper connection to transfer the heat. It's the only idea I got, assuming of course I didn't damage anything the first time around with my artic silver accident.

Anyone else got input on this?
 
This thread would probably do better off in General Hardware, since it's very possible it's not a motherboard problem.

Anyway, have you checked the RAM with memtest? Any idea how hot the CPU is getting while under load? I suggest playing one of the games you mentioned for a while and then checking the temperature before it reboots, or immediately after it reboots, in the BIOS. Definitely sounds like a heat issue, anyway. If you rule those 2 out (RAM and heat), then you can start checking other things.

It only happens while gaming or doing anything that stresses the system, right? And it is a clean OS installation, right?
 
Anarchonixx said:
And it is a clean OS installation, right?


What Anarchonixx said.

If you reused an old install of XP for another chipset (and some how got it to boot), you might want to do a repair install now that you have all the drivers on your hard disk. Windows will rebuild the low level hardware layer stuff and find most if not all of the new drivers you have installed. No need to reformat and reinstall the entire world. Although you may have to do the repair/install thing twice. I do a lot of board swapping and do repair installs all the time and I know it works. A repair/install can even fix clean installs that go wrong. Patience grasshopper. You just have to reapply Windows updates and activate again to get where you were. Or, if you have an XP SP2 install disc, you won't have to do anything except the activation thing again.

If you are overclocking everything, then all bets are off. Try running everything without the overclocks. Pretty simple troubleshooting technique.
 
Yeah, general hardware might of been a better, idea. Oh well.

Yeah, only during 3d gaming. My temperatues raise and raise.
I left it at the BIOS screen for a while, after a crash... and the CPU temp was right around 54-58 Celcius. And yeah, the RAM has been tested already.

I think it has to do with the temps, i would tend to agree. However, I'm concerned that my core voltage never changed, or even leveled out when I swapped power supplies. this would indicate a problem with power distrobution on the motherboard? perhaps?
 
jynxed said:
Yeah, only during 3d gaming. My temperatues raise and raise.
I left it at the BIOS screen for a while, after a crash... and the CPU temp was right around 54-58 Celcius. And yeah, the RAM has been tested already.
Wow, my bad for not reading the whole 1st post. Didn't notice it until now. I guess I missed the part after the picture about the temperatures after gaming. Sorry for making you repeat that.~ Anyway, I would say without a doubt it's a temperature problem, and all there's left to do is practice that thermal grease application.

jynxed said:
I think it has to do with the temps, i would tend to agree. However, I'm concerned that my core voltage never changed, or even leveled out when I swapped power supplies. this would indicate a problem with power distrobution on the motherboard? perhaps?
Well, it's still well within spec for a proper working system. Maybe all that's happening is the motherboard voltage sensing isn't perfect and has a certain margin of error or something... I wouldn't be concerned about the PSU at all.
 
If your temp is 58C after a crash, that sounds like overheating to me. Try reseating your heatsink. Sometimes even the best of us put them on backwards or something silly like that and get overheating. In my experience, Athlons tend to get unstable in the low to mid 60s for temp. I would try Prime95 and Memtest also...
 
jynxed said:
and Yes, i formatted the computer twice, during trial and error runs.

Well I'll bet that was fun. Almost never necessary for any reason unless your drive is full of spyware, Kaazaa sewage, or virus wormy garbage. I'm not saying I've never reformatted, but it never makes any difference if the problems are hardware related. The reformat and reinstall methodolgy is mythology passed down from the Windows 3.0 era. I know you young guns love to do it, but you don't have to, really. OK, Off the soap box and back on topic, sort of.

I was getting random reboots with a BFG 6800 OC that was about to die. I wasn't even overclocking it, much. ATI cards are generally hardier, but don't count an overclocked and unhealthy graphics card out.

I'm not so sure about the heat in and of itself being the entire problem. 58C is warm for sure, but not out of spec for the part at the factory clock speed. The combination of overclocking and high temps, now that is more likely the problem. Cool that baby down and if it still reboots look at the graphics card.
 
Yeah, well the first time I formatted because I didn't partition my HD correctly on initial install, that... and I was having some driver "issues". Oh, and no I didn't have Partition Magic on a bootable cd with me. The second time I formatted was when I installed the Radeon 9800Pro.

I was using a Radeon 9000 series in this machine before I bought that card. And even with the Radeon 9000 card, the comp would crash. Albeit I would actually SEE the text on the blue screen with that card... with the 9800Pro it's just blue.

I never said anything about overclocking, but I guess it's the [H]ardForums, what can you expect. :) Everything is running at stock speeds, and the video card is new.

Although I have a new thought now. I was playing Call of Duty, and I decided to watch my temps every couple of minutes. My CPU started out around 35, and MB was 33. My cpu started to slowly rise... but my Motherboard was going hotter than the CPU. My motherboard got temps of 40 real quick, then im sure it hit at least 45. My theory would be that I damaged the motherboard when on the first time around, caused by the overheating of the thermal paste (I put wayyyy too much). It makes sense to me anyway... and I was looking at my motherboard, and in the center of it, there's what looks like aluminum foil (it's not of course)... it was right next to the motherboards little heatsink there. I noticed a few bubbles on it, which would normally mean heat damage. But I don't know.

I also remounted the CPU and heatsink earlier today. That went all good, but unfortunetly didn't make any difference to my crashing problem. The only thing left I can think of doing is replacing the motherboard.
 
Back
Top