Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Computer freezes or restarts 1

Status
Not open for further replies.

gregju

Technical User
Nov 4, 2002
6
BE
Hallo,

In September I put together my new computer: ASUS A7V333/PA, AMD Athlon XP 1800+, WD800JB 80 GB 7200 rpm HD, 512 MB PC 2700 DDR333 RAM, Gainward GeForce 4 Ti4200 64 MB, and from my old computer a Sis900 Ethernet card, a Guillemot Maxi TV/Video 3 PCI tv tuner card, a generic firewire card, a Creative CD ROM drive and CD RW in a brand new AOPEN HQ45PRO 250 W case. I installed Win 98 SE and Mandrake Linux 9.0. It worked all right, and I was very pleased :) with it until yesterday, when suddenly, in the middel of my work (in linux) it rebooted. After rebooting it started up again, but after about 5 minutes the computer freezed. After a reset I booted into windows, where it freezed again in a few minutes. Then I rebooted into safemode in windows and after a few minutes it wanted to reboot but stopped at the post beeping continuously. After turning the computer off and on again, it produced the same things: either it freezes in a few minutes or it restarts automatically or it stops at the post after a restart, telling me that the cpu is overclocked (which is not the case, and never has been) and the memory test is failed. And it does it just as well in windows as in linux. Once it even freezed while I was in the bios setup!!! :-(
I checked the temperature, but the CPU is arpound 56°C and the mobo is around 47°C, which I think are normal temperatures.
Any ideas? Is it the power supply? Or the mobo or the cpu or the graphic adapter???
Any help is highly appreciated!

Greg Juhasz
 
Continuous beeping during the POST is usually a memory error. Try reseating the memory first.
 
Continuous beeping usually signals the overtemp alarm...I'd boot to the BIOS page that monitors it...(after opening the case to make sure all fans are running and there's nothing crazy going on in there...) if it'll let you.
The fan running slow or the system running hot would sound the alarm. Settings for the temp alarm are usually in the BIOS...and quite a few I've seen are defaulted to 60c.
The monitoring section of the BIOS might be a place to start...just letting it run and watching it. Don't think I'd let it run like that too long, tho, as some real damage could occur. It's easy to overlook a stray setting in the BIOS so might be a good idea to restore default settings in there and watch it.
Tell you the truth 56 seems high for very little running.
 
Hi,

Thanks for the suggestions!!! After a day rest, I started anew.
The bios hardware monitor started at 43°C for the CPU, and within a few (ca. 5) minutes it climbed up to somewhere 50°C, which I thought is rather quick for no cpu activity at all (being in the cmos setup), but still quite acceptable. In the mean time, it froze once (right at the beginning, when the temperature was still 43°C)! The cpu's fan was spinning at 3000-3200 rpm. The mobo' temperature rose from 29°C to 34°C. The temperature alarm was set to
Code:
auto
, which I checked and is 90°C.
I restored the original default settings, took out all dispensable hardware (tv card, ethernet, firewire), reseated the memory (which appears to be fine), and still, it would produce same symptoms: it lived approx. for one minute in Win98, when it automatically restarted, and did not even get as far as the dm in linux, when it froze. :-(
So there I am, still with the same question, still waiting for a solution.

Greg
 
Just today noticed the biggest thing I'd overlooked in your 1st post:
250W power supply?
Suspect power supply!
 
Sometimes in the BIOS there are settings for minimum and maximum temps. This can set the alarm off. Also 250 Watts is not enough for a AMD in most cases. AMD maintains a list of approved power supplies. Reccommmend Antec or Enermax. If you do not like my post feel free to point out your opinion or my errors.
 
Continuous beeping during post may have many meaning depending on the BIOS-
1. Memory error
2. BIOS checksum error

check manual of your motherboard.

Prime suspect in memory, try changing it and setting in BIOS. Try to use failsafe settings.

TV tuner card is also suspect for freezing.

Suggestion for temperature are good ones, try them.

Just remove the fast/quick boot from setup and see.
Also try booting from bootable disk having HIMEM.sys on it and check for memory error, it shows error in many cases for failed memory.

But after all I found with experience that freezing and rebooting is mostly due to hardware drivers and installing too many devices in a single PC is not recommended for normal user.

Chetan
 
Hi,

Thanks for the new suggestions, though some of them, as I it is obvious from my previous posting, are things I already have tried. I did load the failsafe settings and took out everything which was not necessary (including tv card), without any improvement. Again, the temperature alarm is on
Code:
auto
being 90°C. I even tried underclocking it, with running it at 1153 MHz instead of the normal 1533 MHz.

The continous beeping during post occurs only when the machine restarts automatically (and even then not always), and never when I start it up from being turned off.

I did suspect the power supply. Actually I had wanted to buy a 300 W psu, but the vendor persuaded me that 250 W is more than enough for me. I don't feel like giving out another 90 euros (aprox. $90) for a new psu if I am not sure, that this is what's wrong. That is why I turned here for your help, for which a grand merci as the French say!

Since the problem occurs, with exclusively new bits installed, which are still under guarantee, I am planning to take the computer to the vendor, from whom I bought the pieces, and ask him to sort out what it is.

If you still have any suggestions, they are, of course, still welcome. Otherwise I'll let you know what it was, just in case anyone has the same problem.

Thanks again,

Greg
 
Greg,

I suspect the power supply also. I'd swap in a larger p.s. first and see if that doesn't get it done.

Phantom reboots have also been known to result from low voltage within a building or community, but that's less likely to be the cause.

If the problem persists, and you have two 256MB memory modules, you might want to remove one DIMM at a time. If the problem stops after rebooting with one DIMM removed, you've found the bad part.

Good luck!
 
Try to borrow a psu from someone and put it in your machine.

Your CPU fan is spinning too slow at 3000-3200. Either you haven't got the right fan or the fan is damaged or there isn't enough power to drive it at between 4000 (at the lowest end) up to 6000 rpm.
 
I would add as an after thought that if this has been the case for a few months the cpu is highly likely to be damaged. I would guess that the temp you are quoting of 56 degrees is what you are reading after a few mins of run time before it hangs.

If it has been under power for a few months I bet it has been running at a lot higher temp - I would suggest taking it to a shop ASAP to get it sorted. Also, if you are going to run it in this condition, whilst trouble shooting, I would take out the tv card - they are notorious for blowing when there is a faulty chip, mobo or memory on board.
 
Hi,

Thanks again. I checked the volatges, too:
According to the power monitor of the bios:
Code:
vcore 1,77 V +/-0,03 V
3,3 V 3,30 V +/-0,02 V
5,0 V 4,81 V +/-0,02 V
12 V 12,36 V +/-0,06 V
Isn't the 4,81 V too weak?

Greg
 
That's about -3.8% off the target so I don't think that's gonna be a problem. +/-5% is generally acceptable.
 
I'd upgrade the power supply! 250 watts is not enough, you can buy power supplies in the 350 Watt range for $30.00 online ie (e-bay)maybe less, just do your homework and make sure you get one that will fit your case.
 
I also would say POWER SUPPLY, 250watt on an XP1800+ system is boarderline, if you have a replacement choice go 350watt plus and you can get them alot less than 90euros I would say more like 45 for something reasonable (400watt Qtech)
The low fan speed I am guessing is because it's a 70 or 80mm and not the normal 60mm.
Most ordinary 60mm fans spin around the 5,000rpm mark with exception of the fast delta's (7,200) but the bigger the fan the slower it gets 70mm fans around the 3,800rpm and 80mm fans around 2,500-3,000 mark of course these are just rough average figures, there are high performance exceptions (faster rpm)as well as low noise exceptions (low rpm) to this rule.
Martin Please let members know if there advice has helped any.
 
Dear All,

Thanks for the help! After all, it was the memory (an APACER 512 MB PC 2700 DDR333 RAM). The vendor has replaced it for free (being still under guarantee). He is still convinced that a 250 W AOPEN psu is better than many 300 or 350 W power supplies of other brands, and I think he is right. On the long run though I'll have to change the psu, I guess. For the time being, it's working OK. :)
The cpu fan is indeed an 8 cm fan (Coolermaster Low Noise).
Thanks again for all the help!

Greg
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top