A PC question.

No - just Probe and the power management page in the bios. Both report the same thing.

Mine has always shut down after AVG free has done its checks. And the box that tells it to do this isn't ticked. ;-)

The fact is it was shutting down before XP had even started to load sort of eliminated that sort of software problem to me.

;-)

I can find my way around that ok but this new fangled Windose gets me head scratching. I only use the PC for AV work and browsing - everything else still RISC OS on an RPC. I was considering VRPC to get rid of some hardware - but perhaps not. ;-)

--
*Real men don\'t waste their hormones growing hair

    Dave Plowman        dave@davenoise.co.uk           London SW
                  To e-mail, change noise into sound.
Reply to
Dave Plowman (News)
Loading thread data ...

In message , "Dave Plowman (News)" writes

Ah, I only saw your post in uk.d-i-y, didn't realise you'd asked somewhere there are intelligent people as well.

--
Clint Sharp
Reply to
Clint Sharp

In the UK it is the responsibility under warranty for the retailer to deal with it. In the USA you deal directly with the manufacturer.

Reply to
GMAN

I can just imagine the hassle with what is an intermittent fault. Nor do I know if it's the MB or CPU which is at fault. I'm not sure I can wait for the weeks all this would take. And the cost of the carriage back to them probably several times...

--
*There are two sides to every divorce: Yours and shit head\'s*

    Dave Plowman        dave@davenoise.co.uk           London SW
                  To e-mail, change noise into sound.
Reply to
Dave Plowman (News)

Try the Memtest and drive fitness tests found here and at least rule out flaky memory or HDD problems. Have you looked at the crash dumps yet?

formatting link

Reply to
pipedown

Keep saying it can happen before Windows loads.

--
*How\'s my driving? Call 999*

    Dave Plowman        dave@davenoise.co.uk           London SW
                  To e-mail, change noise into sound.
Reply to
Dave Plowman (News)

In message , "Dave Plowman (News)" writes

Memtest boots and runs before Windows. Event logs are a different kettle of fish though, if you can't get into Windows then it's a little more difficult to get at them unless you have a second system.

FWIW, I suspect the CPU/socket or a thermal issue but that's all moot now anyway, the system is working.

You use the system for work don't you? IMHO you would have been far better off replacing the motherboard and CPU for the time you have spent on it and relegating the components to a less important use. You currently have a system that could fall over at any time with no warning and you can't be sure you have fixed it.

--
Clint Sharp
Reply to
Clint Sharp

Understood.

As I said I don't think it's actually a thermal issue as the heatsink isn't even hot when it shuts down. And it can happen so quickly after switching on from cold that the CPU wouldn't have time to overheat.

Yes - when I have a project. I'm freelance so it's not continuous work.

Had I needed it at this time then that would have been my approach. And I do have a new MB, CPU and memory standing buy - it arrived on Friday. But it will need a complete re-install of everything so I'm putting it off until forced. ;-)

Indeed. Although it's working fine at the moment. And my experience with PCs is that they can *all* fall over at any time. ;-)

--
*Even a blind pig stumbles across an acorn now and again *

    Dave Plowman        dave@davenoise.co.uk           London SW
                  To e-mail, change noise into sound.
Reply to
Dave Plowman (News)

In message , "Dave Plowman (News)" writes

You'd be amazed how quickly a modern CPU heats up, don't forget these things are dissipating massive amounts of power given their size (some of the Athlon XP64 chips, 90 watts for a die that's smaller than your little fingernail), if the heatsink isn't on correctly or not making good contact you can have an overheating chip by the time it's started to display POST.

Interesting read;

formatting link

Having said that, one of my systems occasionally shuts down because the BIOS reports the CPU temperature as over 90C on the first power on of the day, it's lying.

Hmm, unless your software is locked to the current system for licensing reasons it's rarely necessary to do a full reinstall, you just need to know how to wave the magic chicken correctly and reactivate Windows once the new stuff is installed.

--
Clint Sharp
Reply to
Clint Sharp

Indeed. Although mine is somewhat over an inch square. And I've not tried running it without the heatsink.

That's more or less what happened to mine - as soon as I could get to the bios and select the page from a cold start it was reading 40C or so and rapidly climbed to over 70c and shut down.

One snag is the new MB requires different HDs. I stuck with IDE with the old one and the new really needs RAID, since I'd want its one IDE for other things. I suppose I could fit RAID to the old and transfer everything? But many say a fresh install is no bad thing anyway.

--
*With her marriage she got a new name and a dress.*

    Dave Plowman        dave@davenoise.co.uk           London SW
                  To e-mail, change noise into sound.
Reply to
Dave Plowman (News)

Dave, Did you try replacing the PSU ?

--
Best Regards:
                     Baron.
Reply to
Baron

Yes - it was my first thought.

--
*The problem with the world is that everyone is a few drinks behind *

    Dave Plowman        dave@davenoise.co.uk           London SW
                  To e-mail, change noise into sound.
Reply to
Dave Plowman (News)

Thanks ! I couldn't remember if you had or not.

I possible answer: Some time back I had problems with a machine that kept shutting down shortly after boot. No O/S just a boot floppy. The only clue initially was that if you restarted the machine straight away the time to shut down got less.

It turned out that the voltage regulator for the CPU simply couldn't hold the CPU voltage at 1.65volts and over a few seconds the voltage rose upto 2.25volts at which point the PSU got switched off.

The cure was replacement of the six capacitors in the CPU PSU circuit. Non were bulged or swollen, no visible signs of damage. All the caps exhibited ESR's around 10 to 15 ohms.

However prior to its failure the machine showed symptoms similar to yours !

HTH.

--
Best Regards:
                     Baron.
Reply to
Baron

On Mon, 19 May 2008 04:54:55 -0700 (PDT), jango2 put finger to keyboard and composed:

Some motherboards put a temp sensor inside the CPU socket.

- Franc Zabkar

--
Please remove one \'i\' from my address when replying by email.
Reply to
Franc Zabkar

Dave Plowman: "The bios power management page again showed the CPU overheating - going from ambient to overheat in around a minute. But the heatsink was cool to the touch. ;-) "

Franc Zabkar: "There is a thermal diode on the CPU die. The diode appears on two pins, THERMDA and THERMDC (anode and cathode?). An external hardware monitor chip senses the diode voltage and adds calibration factors (eg temperature offset) as per the CPU's Thermtrip Status Register. Furthermore AMD's datasheet states that "if the temperature sensor has an ideality factor different from 1.008, a small correction to this offset is required".

Me: Find the diode and replace it.

Reply to
jango2

Inside the CPU? Good trick if you can do it.

--
*If at first you don\'t succeed, try management *

    Dave Plowman        dave@davenoise.co.uk           London SW
                  To e-mail, change noise into sound.
Reply to
Dave Plowman (News)

Dave Plowman: "The bios power management page again showed the CPU overheating - going from ambient to overheat in around a minute. But the heatsink was cool to the touch. ;-) "

Franc Zabkar: "There is a thermal diode on the CPU die. The diode appears on two pins, THERMDA and THERMDC (anode and cathode?). An external hardware monitor chip senses the diode voltage and adds calibration factors (eg temperature offset) as per the CPU's Thermtrip Status Register. Furthermore AMD's datasheet states that "if the temperature sensor has an ideality factor different from 1.008, a small correction to this offset is required".

Me: I vaguely recall replacing a defective temp sensing diode located ON the motherboard. Don't remember which MB. Pull up the datasheet of the hardware monitor and trace along to the temp sensor section. A faulty diode or peripheral component would cause the hardware monitor chip to report "Temp Hi" to the cpu.

Reply to
jango2

I editted my post, read again.

Reply to
jango2

Didn't stop you sending the original. ;-)

--

*The first rule of holes: If you are in one, stop digging!

Dave Plowman snipped-for-privacy@davenoise.co.uk London SW To e-mail, change noise into sound.

Reply to
Dave Plowman (News)

It's part of the same die, it'd be much easier to disable it.

Reply to
James Sweet

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.