Shuttle SN41G2 keeps rebooting - PSU problem?

RostokMcSpoons

RostokMcSpoons

RostokMcSpoons

Associate
Joined
6 Oct 2005
Posts
135
Hi,

I'm having to use my old Shuttle system cos my 'proper' pc's case is currently RMA'd to OCUK :( Unfortunately my bad luck hasn't stopped there, cos the Shuttle is playing up a bit too!

Current config:
Shuttle SN41G2 (v1 or 2, probably v1)
Athlon 2500XP
2x512mb Corsair Value Select DDR400
Radeon 9700 128mb
120gb WD hard drive
Samsung DVD

Software / drivers:
Win XP Pro SP2
Latest NForce drivers
Cat 7.2 (but no CCC cos of the installer bug)
Ati Tools

Overclocking:
FSB at 166mhz, so cpu not overclocked.
I did have the memory running at 400mhz, but I've tried running it in-sync witht the processor at 333
Graphics card at stock.
I've recently reseated the heatsink with a fresh application of Arctic Silver, and I've seen no CPU Overheat warning light on the motherboard.

Symptoms:
System sometimes operates completely normally, even during intensive Day Of Defeat : Source sessions (highly stresses cpu and gpu)
Othertimes.. it will restart even during the bios startup.

I've had it run normally with the case on, and I've had it do reboots with the case off so I don't *think* it's an overheating issue. It reboots when starting Windows, idling at desktop, during internet browsing, or mid-game.


When the system restarts often the BIOS screen appears slightly corrupted - small lines appearing randomly across the screen, parts of text missing. It still seems to work ok. Obviously that looks like the graphics card might be playing up.


Tonight has been particularly odd - I almost gave up trying to get it to run for more than 2 minutes, then gave it 'one for luck' and it's run fine ever since (for 4 hours of full-time DOD:S, with TeamSpeak running too)

The graphics corruption made me think about the Radeon card, but the air coming off that is cool, and it happens even during the BIOS start up, so surely that rules out the card, it's drivers, and the motherboard drivers for that matter.

But there is one thing I noticed when I checked the BIOS setup: I went to the PC Health Status page, and noticed the PSU voltages aren't on-spec for one setting: there is a "-12v" monitor that is reporting "(-)13.46v" which seems an awful long way from -12v! The "12v" reading was normal (can't remember exactly .. 12.2v?), as was the 5v (5.1v?)

Is that likely to be my problem? Do I 'just' need a new PSU?

Any thoughts would be gratefully received!



edit: I've changed WinXP so it won't auto-restart - so I can catch any BSOD messages. I'll post 'em when I get 'em!
 
Last edited:
Go and reseat the RAM and graphics card, set the BIOS back to defaults and if you have a spare, change the BIOS battery on the motherboard.

Tell us what happens next :).
 
Arc,

I've read the -12v rail isn't important (indeed I've found someone posting where his read at 3v! And his pc was still working fine)... so that rules out the PSU... perhaps.

I've just switched on my pc, with the case off, and it got halfway through WinXP starting up before it switched itself off. I think that guarantees it's not a simple overheating issue, as everything was still cool to the touch!

I've rather pre-empted your advice - I swapped the ram sticks around. And the pc has booted successfully, so *fingers crossed*.

I'll buy a new battery tomorrow too.
 
you need a good 12v rail to keep your system stable. i'd say anything down to about 11.3v should be ok. anythin below and it could be whats causing it.

have you got a spare powersupply you could try? even an ATX one just running outisde the case so you can rule out the PSU?

aslo do s arc suggested with reseting everything. one other thing to try is dont bother with a g-card and use the integrated ones and see if it still does it?
 
RostokMcSpoons said:
I've read the -12v rail isn't important (indeed I've found someone posting where his read at 3v! And his pc was still working fine)... so that rules out the PSU... perhaps.
Click to expand...

If everything else fails to resolve the issues then do as cokecan72 suggests, try another PSU. That will definetly rule it out.

RostokMcSpoons said:
I've just switched on my pc, with the case off, and it got halfway through WinXP starting up before it switched itself off. I think that guarantees it's not a simple overheating issue, as everything was still cool to the touch!

I've rather pre-empted your advice - I swapped the ram sticks around. And the pc has booted successfully, so *fingers crossed*.
Click to expand...

You'd be surprised how quick a CPU heats up if the heatsink isnt on correctly :eek:. In this case however, due to the length of time you've been able to play games, it doesnt sound like a temperature issue.

If you still get lock ups then run the machine with only one stick of RAM. Any more lock ups after that then run it with the other stick of RAM on its own.

RostokMcSpoons said:
I'll buy a new battery tomorrow too.
Click to expand...

Due to the age of the SN41's the battery is likely to go. Mine failed about a year ago but you do notice that with the BIOS settings being lost and the like.

Tbh I think its a RAM or driver problem, just a case of tracking it down.
 
can you get to the event viewer in cpanel? any of the error reports in there showing the restarts? my desktop started doing the same and it turns out my hard disk is faulty. if you could post the codes here it might help see what's wrong. (normally displayed as hex)
 
Hi guys, thanks for the suggestions.

Well, it's all gone down hill :(

I wondered if the graphics driver being incompletely installed was a problem, so I did what any sane(?) person would do - ran the ATI uninstall utility and tried to re-install a set of old (known-to-work) drivers.

Except it buggered up :(

I got a message 'INF driver install failed, try loading a standard VGA driver' (or something similar).

Well, I couldn't get Windows to download anything automatically, so I fell back to the integrated graphics. That was ok, until I tried to change the screen res up from 800x600... firstly it didn't show 1280x1024, which was odd, but not the end of the world, so I chose 1076x768, and that's when the world went dark. Never mind, a reset should fix it I thought. Nope. Tried reinstalling the ATI card. Nope. Tried the integrated graphics again. Nope.

And of course all the time I *still* had the random reboots happening (more and more frequently. No BSODS, just an instant power down and restart)

I've also tried changing the ram sticks around, and using the old ddr266 ram.

Unfortunately I don't have a spare PSU (what with my other psu currently being RMA tested at OCUK! (And yes, I do have a surge protector on my power!)

Now when I press the power on button it either doesn't respond, or there's a long delay. Although I noticed that when I pulled the Clear CMOS jumper it powered up instantly!

I don't *think* the battery was a problem - the time and date and all the settings were all fine.

So there you go, it's a dead duck.

<blub>
 
so can you not get into windows at all?? have you tried safe mode (F8 on boot up) as that should jsut load up generic default drivers for basic display purposes. If you can get to safe mode, uninstall all graphics drivers in control panel and then run driver cleaner (you'll need to download and install) then restart back to normal windows and see if it helps

i just thought of another suggestion that may be too late by meh :)

right click My Computer - Properties - Advanced tab - start up and recovery settings - take the check out of automatically restart

with this unchecked it should give you a blue screen instead of auto restarting. may give you an error code to help track down the problem
 
cokecan72 said:
so can you not get into windows at all?? have you tried safe mode (F8 on boot up) as that should jsut load up generic default drivers for basic display purposes. If you can get to safe mode, uninstall all graphics drivers in control panel and then run driver cleaner (you'll need to download and install) then restart back to normal windows and see if it helps

i just thought of another suggestion that may be too late by meh :)

right click My Computer - Properties - Advanced tab - start up and recovery settings - take the check out of automatically restart

with this unchecked it should give you a blue screen instead of auto restarting. may give you an error code to help track down the problem
Click to expand...


I did the Recovery settings thing a couple of days ago, I still didn't get any BSOD's :(

Nope I can't even see the BIOS screen anymore - something else has gone awry, it doesn't matter if I use onboard gfx or my ati card, I get nothing... it seems to power up ok. It does seem like the old 'no POST' problem though, so I'll probably try pulling and re-installing the old memory, and resetting the CMOS jumper cos I'm sure it's not *completely* dead (though there's always the chance I static-zapped something cos of all this fiddling, though I always tried to discharge any static through my rooms' radiator, or the Psu case before I handled components)
 
Static is almost a non issue nowadays so dont worry about it :).

Theres a special CMOS clear function listed in the manual (simply putting the jumper over the clear pins wont do it properly). Do that, take out/disconnect any other non essential kit* and try booting it with the monitor connected to the onboard graphics card.

Sounds like Windows is completey goosed to with all the constant restarts and driver issues. You'll need both sets of driver cleaners. Reinstall the Nforce motherboard drivers from the Nvidia website (making sure to uninstall the current ones), then the Nvidia graphics drivers from the Nvidia website. Leave the ATi card out until you narrow down the problem.

*By non essential I mean your ATi graphics card, 2nd stick of RAM, Hard Disk and DVD drive.
 
Right, I did get it boot again last night - I think I had the CMOS jumper across the wrong pins! I installed an old set of Omega drivers, but the rebooting started again and I gave up.

I've just heard that Antec will be providing me with a new psu for my main systems' case - seems like I'll just wait for that to turn up.

In the meantime I'm missing my fix of internet / DOD:Source / Hammer and am having to watch god-awful pap on telly every night with the wife. It's killing me, I tell ya!


I don't think it's a driver problem because it can happen even during the BIOS POST, it really does seem to me the psu of the mobo is the problem. In which case - it'll go and gather dust in my attic :( I can't be bothered to spend money on replacing either or both those parts
 
Last edited:
When your new PSU comes back, give it a go and see if it still reboots.

If it does then store it until you have the time/cash/patience to fix it. I'd be half tempted to reinstall windows with brand new drivers when you have the new PSU hooked up to it.
 
can you get into event viewer (in control panel) to see if you can grab the error codes for the BSOD's? it will help show what is causing it to crash. one of my PC's doesn't BSOD it just resets, but in the logs it shows as it BSODing.
 
Steeps said:
can you get into event viewer (in control panel) to see if you can grab the error codes for the BSOD's? it will help show what is causing it to crash. one of my PC's doesn't BSOD it just resets, but in the logs it shows as it BSODing.
Click to expand...

thanks for that tip - I'll hopefully be able to take a look tomorrow - I've got a cheapo case (with psu) on order to get me active again (I've found out Antec are apparently likely to take weeks rather than days to give me my new psu)

So I'll plumb my Shuttle into the 350w psu, and hopefully it'll work long enough for me to look at the logs :)

I'd like to fix the problem - if I can sell it it'll help fund my next upgrade!
 
Ok....

good news - with the cheapo 350w Psu, it's stable and doesn't reboot. So now I know what's wrong, I know a £40 'investment' of a new SilentX psu will make it saleable.

Bad news - the logs showed some errors. There were quite a few regarding 'corrupt file systems' which no doubt go hand-in-hand with CheckDisk getting run on Windows startup. I've re-checked the drives - they're all fine.

The only other error relates to a service not starting properly... something like ULIAGP.dll (?) It looks like a de-installation of a set of ULI chipset drivers I had when the drive was temporarily used to boot my 'big' system didn't work properly. Running CCleaner didn't zap it either. Are there any other clean ways of removing whatever is calling it?
 
You must log in or register to reply here.
Back
Top Bottom