PC boots for two seconds, then stops - MB or CPU?

Hey. I recently bought myself a Cooler Master HAF X and a Corsair HX 750w PSU, together with a replacement MSI GeForce GTX 460 HAWK edition graphics card. It's been standing still for about a month without any action, until I got my graphics-card today. After putting in said GFX-card, I turned on the power switch behind the PSU. Immediately, the PC started. I thought this was odd, since I hadn't even been near the start button, but figured it might be some weird glitch. However, it constantly restarted itself (I mean a full restart, everything stopped for a second before restarting). I've managed twice to get to the BIOS, but only for a brief 10-20 seconds before another restart. I can't remember to hear a POST-beep either, but that could just be my brain misremembering. There was light in the DVD-drive, and I could hear at least one HDD spinning (I got 2).

So I plugged every accesory such as mouse and keyboard out and took it in the hallway, where I don't have a wall-to-wall rug. I built the PC in the hallway to avoid static electricity from the floor, even though I put the GFX in in my room. Anyways, I placed the case on it's "back" with the motherboard facing upwards to look for any loose screws, and I also looked for any of those things you put your motherboard on top of (the long, bronze things ) which could be causing a short circuit, but found neither. When I then turned on my PC, it didn't restart, it just stopped completely after a second, it didn't matter whether it stood on its back or on its wheels, it didn't restart again (even though it still starts the second I turn on the PSU). I've ruled out the new PSU as a cause with my old PSU, same problem.

The obvious suspect now is that the motherboard is fried since I also had the occasional slipped screwdriver (I'm pretty clumsy, not a good mix with PC building). But I also suspect the CPU to be the problem. While changing the PC case, I accidentally pushed over a lever which apparently was the one holding the CPU cooler on. I heard a sound like something freshly glued is being ripped apart, which I suspect was the thermal paste. Could it be the CPU? I've read somewhere that it will automatically short circuit itself if it gets too hot or the CPU cooler isn't connected properly, but would that also cause the restarting I had in the start? Also, is there any way to test these teories other than buying new ones (I don't have any viable old ones), and could turning the power on to check if the PC is okay potentially damage my other components such as my GFX-card?

One last question. If it's the thermal paste between the CPU cooler and the CPU that's causing the problems, can I simply replace the thermal paste with new?

 

One thing I forgot to mention, I was kind of obsessed with getting the cables away as much as possible, so I routed them all behind the MB tray (the case is built specifically for this to be possible), perhaps not as pretty as possible. No connectors are behind there, just the cables themselves, but could it be the cause?

 

Yes - make sure you clean off all of the old paste. The same applies with any shipping tape.

Couple of quickies: try resetting your BIOS, either by jumper or pulling the battery - in fact, if the machine's been dormant for a while, you might consider replacing the battery.

If the mobo has onboard graphics, try those instead of the GTX for troubleshooting purposes.

If you disconnect all peripherals from the mobo (hard drives, opticals, etc.) and power just the board, do you get any beep codes?

As far as the cables: that's a common practice, which should be fine as long as nothing is crimped or pinched.

 

Thanks, I'll buy some new thermal paste and try that one when I get a chance

As for the battery, a couple of questions: my motherboard is an Asus M2N32-SLI Deluxe, do I need a special type of battery, or do all MBs use the same kind? And how much do you reckon I'll have to pay for it?

I'm unsure, but I haven't seen any onboard GFX mentioned anywhere about my MB. I tried running it without anything other than the motherboard itself and the CPU, same problem.

I do have some additional information though:
- The GFX has a red LED-light. I dunno if this is significant or normal, since it's a brand new card
- The fan speed could sometimes be higher than usual during the failed boots I had in the start

Also, when I reset the BIOS I had to remove both the battery and the CMOS jumper. The jumper was placed in an awkward position so I had to use a small pair of needle-nosed pliers to remove it, and I managed to drop it. Does it matter which way it stands? There was no way to identify which way was up.

 

I had a simular thing happen to me i was originally was going to do a minor hardware upgrade (cpu (intel duo core 2 cpu to intel core 2 quad)) and a 400 watt psu to 630 watt psu) and my old mobo (an asus p5ne sli supported quad core so when i installed the psu and psu and connected everything i plugged in the power and turned it on nothing happened i got power (the led light on the mobo was lit it just wouldn't post) so i put the old cpu, psu back in to check and see if it worked on it (still nothing) so i had it taken in to computer repair shop and had them test it (about a few days later i heard from them and they told me my mobo was dead but my components where fine (so i bought an intel mobo (since i had no issues with them in the past (i had a gateway pc with a intel mobo that lasted for years, if your curious my new mobo is a intel dp43tf
now i'm running windows vista home premium 32 bit with my quad core cpu and vista runs real nice.:drool

 

Ah yes, I was afraid it might be my motherboard, I just wanted to try out some other possible causes first.

 

the only thing i can suggest if you do need to replace your mobo, get a intel motherboard as long your case is a atx formfactor design

 

I always hate to say this but did you triple check all your connections? The case connections to the mobo could be wrong as they are very small and easily misplugged. Oh, have reseated your memory/GFX card? I guess there are no beeps either?

 

For the most part, the batteries are the same, and only cost a few dollars. Usually, it doesn't matter about jumper orientation.

Below is a pic of the M2N32-SLI Deluxe - you should see the VGA connector.

 

What do you mean by my case connections? I'm sure the USB, FireWire and the "misc"-connections (pwr led etc etc) are all connected properly. My memory can't be put anywhere else, because there's some Asus cooling technology-pipes that blocks the other two RAM slots, and I've tried booting without the GFX-card on, same problem. Also, no beeps.

Weird, I have no on-board VGA connector.

Anyways, I'm thinking I'll just give it to the computershop where I bought it, they also perform repairs. It'll cost me a few bucks, but it will save me the hassle of worrying about it.

 

Do an MG Forum search for 'capacitor', and you'll find some good pics of what to look for on the mobo, if you haven't already.

Quick side question: does that board have WiFi-AP Solo? If so, have you used it, or are you going to? That Skype feature looks cool.

:major

 

I believe I do have the Wifi-edition, although I'm not sure. It was bought about three years ago, at a time my computer knowlegde was quite non-existant, so I haven't looked into the features of the MB.

By capacitor, what do you mean? Should I look for a "bad" capacitor to indicate if the motherboard is fried?

 

Yes - quite often, bulging or leaking caps are signs of motherboard failure.

.

 

Thanks, I'll look for any bad capacitors when I get home.

Just wondering, what could theoretically cause the PC to turn itself on the second it's connected to the power, when a PSU fault is out of the question? Seems like an odd occurence which can't have a lot of causes.

 

Okay, I've looked all over the board. No damaged capacitors. My current thought is that I messed up when I accidentally hit the CPU cooler-lever, and I'll try to replace the thermal paste. I've heard acetone and a q-tip is good for removing old thermal paste?

 

I fixed it! Or, well, semi-fixed it. It turns out the entire problem was that the pwr-switch cable was plugged in the wrong way (which is odd, on the MB manual it showed a picture where it stood with the text downwards)

Anyways, I still have one question and one issue. Should I still replace the termal paste on my CPU cooler? It seems okay, and it was still "sticky" when I put the CPU cooler back in, but it was out in the air for at least 10-20 minutes. Should it be okay?

Also, I still have that red light on the GPU. I don't find any reference to it anywhere, anyone know if it's just for show or if it's an actual issue? It comes the second I turn on the PC. The GPU is an MSi GeForce 460 GTX Hawk edition, brand new.

 

Woot!!! I got another one right!:major:-D Though there really shouldn't be a difference as to polarity for the pwr/switch, maybe check for updated mobo/case manuals online?:confused I think you just missed your shot as it's so easy to do. As to the paste, as long as you didn't touch it with your fingers you should be fine though I personally would go the clean/re-apply way.

 

Ah yeah, I think I'll redo the thermal paste, as I did touch it with my fingers to check if it was sticky

Thanks for the help

 

Good job - old Occam's Razor rears its ugly head again.

 

As far as the Hawk's red LED, I couldn't find a decent manual, but I did read that the N275GTX has green, blue, and red LED lights on the back to indicate low, medium, and high power loading status - since the Hawk is PWM overclockable(sp?), I'd imagine it's got similar indicators.

 

i looked on google for what the red leds where for on your card, i found a review on one site that talks about the leds and what they are for http://www.hardocp.com/article/2010/06/22/msi_r5870_lightning_video_card_review/ and it seems like the red leds are for when you are playing a demanding game, and to answer your question about leaving the thermal paste exposed for 10 to 20 minutes, i wouldn't trust it, i would remove the paste and reapply it just to be safe.

 

I suppose it's about power load. Now that I got my PC up and running (replacing the thermal paste today went smooth ) there's actually 5-6 diodes with a red light, but I don't see anything wrong with it in any monitoring software.

I just gotta say, going from NZXT Hush to Cooler Master HAF X is an extreme upgrade temperature-wise. I went down from 70C to 30C idle, and 80C to 50C under heavy load, overclocked.

 

Wrote the previous post before I saw yours, thanks for finding that for me. I have googled it up and down, and didn't find any reference to it

 

Wow man, that is extreme!:eek I almost bought a CM haf but decided to go with a CM Sniper case for the cost. Instead of 240mm fans I have 3X200mm fans and I forget what else. I'm sure your new application of TIM also helped a lot. 50C @ full and OC'd load is just great!:cool How much of an OC may I ask?

 

About 950 MHz core clock, still pretty new at the whole overclocking-game so I haven't touched much else, I'll look further into it later.