[SOLVED] boot up problems with my office computer

Pants

Active Member
Reaction score
21
Location
California, United States
My specs:

Case: Overseer RX-1 ...VN7000 Series
Motherboard: Asus M5A97 R2.0 (uEFI)
CPU: AMD FX 6100
RAM 8 gb
HDD: Seagate ST320005 42AS 2TB
Video: Video|ATI Radeon HD 5450
PSU: Coolmax zu series 900w


My house is in the country and we have brown outs a few times a week...ie Lights dim and other power sags.

When I power on my PC, particularly in the morning, I get various strange error messages from the motherboard. The only thing that seems to fix it is power cycling the power supply...ie Turning the switch off for a minute or so, then powering it back up. THEN powering on the computer and booting. The Computer is plugged into a surge protector.

Is this situation common for frequent brown out environments? I checked the voltage pin outs of the PSU while it was doing a chkdsk in Windows Recovery Environment, booted from the install disk...The output voltage on all the pins looks good, using my volt meter.

I put Vista on it recently as the host OS (since it's the only retail license key for Windows I have right now). When I first installed Vista it ran almost flawlessly. But Vista has been freezing, despite various optimizations. Afterwards I noticed that it freezes at the log on prompt before any programs even start up, which is potentially pointing to other issues besides the OS, like the power problem I just mentioned???
 
Last edited:
1. "various strange error messages from the motherboard" Well that really narrows it down, can you be a bit more vague here? Seriously as a tech you ought to be able to tell us more than that. End users do that crap to us you should know better. Tell us the exact error messages or don't bother.

2. You are in an area with bad power but you don't have your equipment on UPS unit? Take a revolver put one bullet in a chamber, spin it and fire. Same result. Looks like you spent a buck or two on this setup don't you think you need to protect it?

3. "Afterwards I noticed that it freezes at the logon prompt before any programs even start up, which is potentially pointing to other issues besides the OS" NO just the opposite. Hangs like that are usually caused by drivers that are stalling on boot up. Also if you have a server providing AD but you are not using it for DNS such a delay can be cause by that. Need to look through the boot logs and see if you can see where the delay is happening.

4. It is Vista. Hello? Microsoft's most sucky OS ever made and you wonder why you are having problems that are historically known issues with it?
 
The only thing that seems to fix it is power cycling the power supply...ie Turning the switch off for a minute or so, then powering it back up. THEN powering on the computer and booting.

Power issues like that are usually capacitor related.

When the SMPSU powers up, the controller IC/circuit monitors the output voltages that it generates and controls using pulse width modulation (PWM). If the initial voltages are out of tolerance or the PWM is unable to achieve the correct voltage, the controller shuts down.

Commonly the cause is insufficient capacitor smoothing, caused by faulty/leaky capacitors. Without sufficient smoothing the voltage fluctuations can fall outside of tolerance limits. Turning the supply on and off, or allowing the supply to warm/cool can sometimes catch the voltages within the limits, due to component temperature coefficients or pure chance, allowing the supply to power up normally.

When a supply output voltage is fluctuating due to poor capacitor smoothing, a multimeter will display the average voltage, which may appear normal. You would need to 'scope the voltage to properly check it.
 
My house is in the country and we have brown outs a few times a week...ie Lights dim and other power sags.

If, as you say, you're out in the country that's subject to brown outs, sags, etc, then I have to assume you're using an UPS. If not, it's just a matter of time before you get nailed and suffer a variety of power surge-related problems.
 
Seriously as a tech you ought to be able to tell us more than that.

I was insinuating that I believe the particular error messages are irrelevant. My theory was that the voltage sags and/or power supply was causing variety of motherboard processing problems. Sure I can write down the error messages but I don't think they're relevant.

2. You are in an area with bad power but you don't have your equipment on UPS unit? Take a revolver put one bullet in a chamber, spin it and fire. Same result. Looks like you spent a buck or two on this setup don't you think you need to protect it?

I had planned on getting a battery backup ASAP. Budget is really tight right now.

3. "Afterwards I noticed that it freezes at the logon prompt before any programs even start up, which is potentially pointing to other issues besides the OS" NO just the opposite. Hangs like that are usually caused by drivers that are stalling on boot up. Also if you have a server providing AD but you are not using it for DNS such a delay can be cause by that. Need to look through the boot logs and see if you can see where the delay is happening.

I believe I've narrowed it down to the video driver. Going to test an updated driver.

4. It is Vista. Hello? Microsoft's most sucky OS ever made and you wonder why you are having problems that are historically known issues with it?

Yes I agree with this. But I run Vista on another computer with far less computing power and it runs good.
 
Last edited:
Yes please, if you don't mind.

I'll write down the errors as they occur over the next week.


I believe I've narrowed down the freezing problem to an overheating video card. It was very hot to the touch a little while ago.. Got a fan blowing on it with open case. System isn't freezing anymore.

However, computer just started rebooting automatically, even while in safe mode. Don't think an updated video driver will fix that.

I reseated the ram and 24pin ATX PSU adapter, didn't seem to help. Gonna try removing some hardware and reseating some other stuff.
 
Last edited:
But your video card wouldn't be overheating when you first power on the system.

I've seen video cards overheat relatively quickly after startup.

CPUs will overheat pretty fast and cause problems, if they are not cooled properly. GPUs are not all that different.

I'm trying to get the heat sink off the video card so I can redo the thermal grease, but it's stuck on there with some really lame fasteners, where once they are pushed through the card, they latch on like a fish hook. Hard to get off :(
 
Logically if that was the case then the machine wouldn't boot at all. I'm sure you have found ONE problem. I'm thinking that you have more than one.
 
Logically if that was the case then the machine wouldn't boot at all. I'm sure you have found ONE problem. I'm thinking that you have more than one.

Yes, I think there is something going on with the power supply, for starters.
I'm thinking doing an RMA on it.

System just froze up again even while the vid card was cool. This was in Windows normal mode though, not Safe Mode. So I think the freezing in safe mode was due to the overheating card.

As for the freezing in Windows normal mode, I went ahead and installed the updated video driver. We'll see how it turns out.
 
If the GPU is overheating and it doesn't have a fan on it then likely your card is bad as it should be engineered to run properly with just a heat sync. Even with your replacement of old heat sync compound the damage may have already been done.

Time for a full slate of diagnostics. PSU, HDD, Video, Memory, etc.
 
Rule out the video card and swap it. If you don't have one invest $30 and buy one from Amazon. It will pay you back with huge future returns. You can't afford not to have one.
 
I have a couple of different video cards. I swapped in a PCIe card. Tried it in each PCIe slot (there are two on the motherboard) and got the same thing. Computer was freezing and rebooting.

However, I removed one of the ram modules to see if I could isolate it to one of them. The system ran stable for a long time and did not crash/reboot ever.

So then I tried the other ram module in the same slot as the other one, and it ran stable and did not fault.

So it seems that both ram modules are good. So I thought the other ram slot might be bad...So I installed and booted up with one ram module in the suspected slot, and still the system ran good...with one module installed.

THEN I thought I would put both back in and try again as normal and everything is running stable again.

This is a weird one.

It's my belief at this point that it's between the PSU and a slight voltage increase needed on the ram. When I built this system it was doing the same thing that it is now...ie Freezing and rebooting. And Windows wasn't recognizing all the ram...It was only showing 4 Gigs available. After long diagnostics and ruling out everything else, I increased the voltage supply to the ram by 1.5 TENTHS of a volt. It's now running at 1.65 volts.

The system ran stable from then on and Windows reported all the ram installed, although I did get a strange error at boot up from time to time, early on on the boot process. I had two OSes dual booted and when selecting one, I got an error...then after restarting it worked fine.

Furthermore, I discovered early on before I put the system together, that the ram I bought did not match anything on the ram comparability list for the motherboard. I went ahead and installed the ram anyway because I wanted to test the theory that somebody said, that if high quality ram is used, it generally doesn't matter if the ram is on the QVL for the motherboard. This system has been running good for over a year now so I don't see why it could be a ram compatibility issue. Perhaps another 1/4 of one tenth of a volt increase will stabilize the system.

Should I just RMA the PSU? Because like I said, it's not just the freezing and rebooting, there is also the POST errors like I described, where power cycling the PSU fixes it temporarily.
 
Last edited:
Did you run memtest and gsmartcontrol to at least verify any blatant issues with the ram or harddrive. Those are less likely problems, but they are easy tests to run. That is the first thing we do on every machine. If you already mentioned that I apologize.
 
Did you run memtest and gsmartcontrol to at least verify any blatant issues with the ram or harddrive. Those are less likely problems, but they are easy tests to run. That is the first thing we do on every machine. If you already mentioned that I apologize.

No. Nothing to loose though.. I'll give 'em a go.

edit: just noticed some bios updates available from ASUS on this model motherboard. Looks like quite a few have been released since I bought this board. Gonna look into them. Apparently I have not ruled out the motherboard, completely.
 
Last edited:
Should I just RMA the PSU? Because like I said, it's not just the freezing and rebooting, there is also the POST errors like I described, where power cycling the PSU fixes it temporarily.

It's almost certainly the PSU that's at fault, for the reasons I pointed out earlier.

I would just order a new PSU -- It's not like they're expensive or that, as a tech, you're not going to need a new one for a job sooner or later. If the new PSU proves the fault, then you can send your present PSU back with complete confidence.


Unless you rule-out the PSU, much of your other tests and diagnostics will be inconclusive and a waste of time. PSU voltage fluctuations can cause any number of misleading symptoms, indicating faults that don't exist with the motherboard, memory, etc. Having stable supply voltages is crucial.




Personally, I would check the voltages with my oscilloscope (a multi-meter is useless in this situation), but if you don't have a 'scope, getting a new PSU is your cheapest/quickest option.
 
Last edited:
Last night I updated the bios. LOT of updates have been released since I bought this board. Gonna test this out for a little while.

But I just powered on my computer a little while ago after I went to sleep for about 6 hours, and I got nothing on the monitor. Opened the case and reseated the video card. NOTHING. Powered down, reseated the video card again. NOTHING

Powered down, reseated the video card again...Ah ha! Monitor came on.

I think the PCIe slot is faulty...Maybe off by just a hair in it's physical dimensions in the way it accepts the card. This isn't the first time I've had problems with it, but now it's really occurred to me. So there's one problem solved.

Good idea about the PSU. If the bios updates don't help, that will be my next step.

Edit: Starting to think now the problem is the PSU. I rebooted after a windows update and got nothing on screen. I put the card in the other PCIe slot and it worked at first, but after a reboot on got nothing on screen.

I just swapped the vid card and tried it out. On reboot I get nothing on monitor. Gonig to go ahead and RMA the PSU.

Evidently, re-seating the video card is not what made the monitor come back up, it was the power cycling of the computer, which indicates a bad PSU.


Thanks


PS. Gonna buy another power supply too. For my tool arsenal. Was going to buy one anyway, but, you know...BUDGET!
 
Last edited:
Back
Top