Rebooting when loading WinXP

**DtM** · 09-26-2005, 06:09 AM

Hello all,

I have a strange problem... When booting my carpc it will reboot as soon as windows starts to load. Let me describe a bit more in detail:

My carpc is a VIA Epia board, powered by a M1-ATX supply with a new firmware pic (the one with the new power down schemes). Fairly new laptop HD, 256mb RAM. Very basic. It doesn't matter if I hook up a slimline CD drive or not.

At the moment I'm testing the PC on my bench. I hook up the M1-ATX to a (quite heavy - amps should be no problem) lab power supply - so both PWR and IGN to +12 at the same time, GND to ground.

When I boot the PC, all goes well until the WinXP boot screen appears. Then the PC shuts down and reboots. And this cycle repeats.

When I leave the PC in BIOS setup, I can leave the thing running for hours. Voltage levels in the BIOS monitoring are ok.

Voltage on all rails are ok, and stay stable (only varying +/- 0,02V) when booting. Power is also ok, the supply is providing enough amps and my lab supply is able to provide the nessecary amps. No spikes or drops as far as I can see/monitor. Also measured CPU voltage, is also ok.

The thing just seems to go down when loading Windows...

When I use a normal ATX power supply, the PC boots perfectly.

Now my question: has anybody else had this problem? I've searched the forum here, but haven't found a problem exactly like mine. Any suggetions on what I could try or check to see what's happening? Is it the firmware of M1-ATX?

Thank you in advance for any views on this...

**turner brown** · 09-26-2005, 07:17 AM

reload windows should sort it
it sounds like a windows prob that i have seen a few times
that would be my first step and is a good test for your power supply
if it can load windows on the 12v power then shouldnt have any probs running with normal use

**Crazy-devil** · 09-26-2005, 08:42 AM

What you can try to do is right-clicking My Computer, choose Advanced, Choose the settings for Startup & Recovery and then tick off "automaticly reboot". Correct me if I am wrong, but then you will see a BSOD in stead of the reboot. Maybe that screen can give you some clues?

**numbers** · 09-26-2005, 11:59 AM

The fact that it runs fine on an atx psu pretty much answers it. The problem is that even though you can't measure it, the spike is there and it is what causes the computer to reboot. Either the m1 is faulty, or you're just drawing too much on one rail or another. If you have a scope, you can probably find the spike with it. The average voltmeter just won't cut it.

**DtM** · 09-28-2005, 09:33 AM

Originally Posted by Crazy-devil

What you can try to do is right-clicking My Computer, choose Advanced, Choose the settings for Startup & Recovery and then tick off "automaticly reboot". Correct me if I am wrong, but then you will see a BSOD in stead of the reboot. Maybe that screen can give you some clues?

Well, if I only got this far

The PC reboots before actually entering WinXP, it shuts down when you see the boot screen with the blue scrolling bar thingie...

Originally Posted by numbers

The fact that it runs fine on an atx psu pretty much answers it. The problem is that even though you can't measure it, the spike is there and it is what causes the computer to reboot. Either the m1 is faulty, or you're just drawing too much on one rail or another. If you have a scope, you can probably find the spike with it. The average voltmeter just won't cut it.

I thought as much... I don't have a scope at my disposition, so I only can guess there. Might the new firmware have to do something with this? Anyway, I should have the old one lying around somewhere, so I could try that out.

Personally I don't thing I'm drawing too much somewhere, as I only have the mobo and an laptop HD attached, and they work perfectly. So I doubt that they are drawing to much current.

Although, the strange thing that remains is that the PC only reboots when wanting to load Windows, since when I go to the BIOS setup and stay in there, the PC keeps running. So it's no timing issue. Does loading Windows draw more power or some strange thing like that than not loading windows?

Thank you for your replies already!

**numbers** · 09-28-2005, 12:58 PM

Yes - loading windows does draw more than just sitting in the bios setup. It draws quite a bit more. I wouldn't know about m1 firmware. I doubt it could cause this but I really shouldn't offer my opinion on that.

**IntellaWorks** · 09-29-2005, 08:20 AM

I ran into this problem a while back. I think it happened twice to me, with two different hardware issues.

It was a PITA to trouble shoot.

I'm thinking that its either the power supply or failing cd-drive. But to be safe I'd check other hardware first. Since a bad RAM chip can act like a lot of other hardware related issues I'd check that first. Swap out the RAM with another stick to see if you get the same problem. If you don't have another stick then download memtest and run the diagnostic on your system's ram.

I'd like to stop here for just a second and throw in some additional information. If your comfortable with troubleshooting the system but just need a platform to load then you can check outBartPE and just ignore the rest of this.

Once you've checked out the RAM and if it's good then check out the power supply. Do you have a normal ATX power supply you can boot and run the system with? One that runs off of wall power? If so, plop that in and see if your still getting the problem. If you are, finally, replace the CD drive with a non-USB drive. (I can't remember if you said you had an EPIA or Laptop.)

The reason why I say NO USB CD drive is because the second time my system had a problem similiar to this was with a failing USB CD/DVD drive. For whatever reason the drive would either re-boot windows before it loaded or take a LONG time to load. Except when I put the drive into another computer, windows seemed to find the drive just fine (except couldn't access it or use it) which is how I determined that it was bad.

Oh yea, a few more things. Is windows restarting at basically the same point or at random intervals? if its at the same point, you can't rule out power issues but I'd start to wonder if the CD or DVD your loading windows from is scratched or damaged. Also, I wonder if the system board is grounding out on something. You might want to check to see if there is a bare wire or soemthing touching the bottom if your systemboard. Make sure your mobo is elevated off whatever its mounted to and that there is no chance for it to ground out on something.

So, check RAM,Power, CD-ROm drive, CD or Media and or for Ground or something touching your mobo.

I wish you the best of luck in your troubleshooting woes. Just try and keep an open mind on this one, if you get frustrated walk away go have a beer (or juice if you don't like to drink)

Sometimes walking away from the problem allows you to come back and look at it with another perspective.

If you have any more questions feel free to PM me.

Beleive me I know what your going through and It's truely frustrating. But you'll get through it!

**numbers** · 09-29-2005, 01:44 PM

The fact that it runs fine on an atx psu seems to be getting overlooked too much. It doesn't mean 100% certainty, but 95% isn't unrealistic.

**ITL** · 09-29-2005, 02:32 PM

I've had this happen before...I want to say it was incorrect jumper settings on one of the hard drives.

**DtM** · 09-29-2005, 04:05 PM

@IntellaWorks: thank you for your extensive writedown.
I don't think the RAM is faulty, I've tried this one in another PC without problems.
As mentioned, the PC boots fine when running of a ATX power supply off wall power.
At the moment there is no slim-line cd-rom drive present. I only used it to install Windows, which I did running from the regular ATX power supply (my M1-ATX hadn't arrived yet at that moment).
So Windows is already installed on my system, and it always reboots at the windows boot screen (the first screen you see with the Windows logo and the running blue bar).
I doubt the board is shorted somewhere as it is (properly

) installed in a morex housing, but I'll check that.
Yesterday I've pulled the housing fan power connector from the motherboard and I thought the PC kept running, but I have to test this more thorough as I just tried that while actually having no time. This might confirm the fact I'm drawing to much power from my supply.

Anyway, let me follow your advise and take a beer first and leave the testing for tomorrow...