Question Win7 PC Suddenly Fails POST

[Mantrid-Drone]

I think I know what the problem is I just want to check what will happen when I apply the 'fix' as I've never had to do this before.

This bit is just for informational purposes really:-

Win7 Pro GA Z77 MBoard i5-3550 16GB Kingstone RAM MX500 SSD primary drive.

Symptoms: with no previous issues PC now won't cold start by PS/2 keyboard.

Using main power button it boots direct into the BIOS screen with a message that the BIOS has been reset and offers a couple of options.

I try rebooting without altering anything and eventually get Windows diagnostic options when it fails POST at the Windows launch screen. Windows says it can't fix the problems which it 'suggests' are due error code 0x490 system file integrity check and repair failed and may be due to 'unspecified changes to the system configuration' or, elsewhere it suggests it may be due to hardware changes or failure.

I try rebooting again and go into the BIOS (Del) and see some of the few settings I chose had changed including the keyboard start return to the Default=OFF.

Left it overnight and when I tried keyboard start the next day it failed and I had the same BIOS reset warning screen again.

Decided to reset to 'optimized' defaults with the exception of the PS/2 keyboard, shut down and restarted. Keyboard start works but again Windows fails to launch at the same place.

Eventually try Windows Memory Test and then Memtest86 from disc as the only other problem I've had with PC was a bad, secondhand, RAM stick which went bad out of the blue and caused a range of problems. Both the tests showed RAM is perfect, so definitely not the issue.

Right from the start my first thought was CMOS battery failure but my one experience of that with a laptop is that it continued to work using the original default BIOS version. But until I researched a bit I had not realised that type of problem can cause serious Windows boot issues too. In fact there are multiple sources in a variety of forums that say the same thing.

So I'm now thinking that as the MB still has its original CMOS battery which, depending how long it was on the shelves before I bought it, could easily be over 12 years old and that was the age at which the laptop (not replaceable) one went too. Coincidence?

Seems the most likely explanation in this case Y/N? Will there be any problems caused by replacing it?

 

[In2Photos]

CMOS battery is where I'd start for sure.

 

[Iron Woode]

If a new battery doesn't fix it then you will have to do a repair install of 7. That may or may not be easy depending on various factors.

Repair Install

How to Do a Repair Install to Fix Windows 7

www.sevenforums.com

 

[Mantrid-Drone]

Repair install of Windows can only be done if you can log in to your account and as I can't do that as it is failing the POST it would have to be a fresh install.

That is my worst case scenario.

 

[tcsenter]

Failing CMOS battery has typical symptoms but then there are these edge cases where it causes weird things you wouldn't think attributable that seem it shouldn't be the CMOS battery. Worth tossing in a fresh one. CR 2032. Might have to reset/clean CMOS via procedure outlined in user manual for your board. And then you'll need to enter BIOS and configure all the settings back to what you want, including time and date.

 

[Mantrid-Drone]

If you Google CMOS battery failing symptoms, as said earlier, you can find plenty of reports dating back years that it can affect the PC start up and often prevent it from booting. Numerous cases too where simply replacing the battery fixed the problem 100%. In my case the fact is the CMOS battery must be in that end of life period 12+ years.

If I leave the PC in question off without any attached power for 12hrs the PC still starts as normal with the BIOS logo on screen (normal) and single, normal beep. However, the first screen that appears is not a Windows BSOD or Windows failing to load, retrying and then going to a Windows repair/boot option but the Gigabyte MB telling me the BIOS has been reset.

That has to be the most relevant clue, surely.

I've taken the precaution of downloading on another PC the last three BIOS available from Gigabyte for that particular Z77- DS3H MB so, if necessary, and it probably will be, by using their Q-Flash facility from the BIOS menu.

 

[bba_tcg]

If you Google CMOS battery failing symptoms, as said earlier, you can find plenty of reports dating back years that it can affect the PC start up and often prevent it from booting. Numerous cases too where simply replacing the battery fixed the problem 100%. In my case the fact is the CMOS battery must be in that end of life period 12+ years.

If I leave the PC in question off without any attached power for 12hrs the PC still starts as normal with the BIOS logo on screen (normal) and single, normal beep. However, the first screen that appears is not a Windows BSOD or Windows failing to load, retrying and then going to a Windows repair/boot option but the Gigabyte MB telling me the BIOS has been reset.

That has to be the most relevant clue, surely.

I've taken the precaution of downloading on another PC the last three BIOS available from Gigabyte for that particular Z77- DS3H MB so, if necessary, and it probably will be, by using their Q-Flash facility from the BIOS menu.

The most likely thing to change by the CMOS battery going bad - causing the BIOS settings to reset, in any PC, especially one that old, is the storage device addressing mode. A lot of old machines defaulted to IDE (even with a SATA drive), when it probably should be set to AHCI.

Many Dells defaulted to an array setup. I used to change this to AHCI, but invariably, if the battery went bad, it would change back. I finally just started leaving it that way.

 

[Mantrid-Drone]

That's useful info - I was trying to remember if I had to change that setting myself.

On my other Win7 PC with another Gigabyte MB I have a full WinXP installation set up for dual boot. But the hassle of swapping the settings from AHCI to IDE for that, then back again and other unavoidable annoyances means I haven't actually booted WinXP on that PC in over ten years.

I later installed it on a VM instead (XP Mode) and if I've needed WinXP, usually for particular video gaming reasons, which itself is rare, that is what I've used.

 

[Mantrid-Drone]

I'm glad I revisited this thread because I'd forgotten about that possibility of the BIOS resetting to IDE. If it is that simple I'm going to be annoyed with myself for not checking back here first.

Today after finally getting around to replacing the CMOS battery, which indeed was as dead as a dodo, with a brand, spanking new Duracell one I've just spent hours unsuccessfully trying to get the PC to boot into Windows again.

Same problem as previously and I've probably made more unnecessary work for myself in trying to fix that.

 

[mindless1]

What did you do in those house spent? Were all settings now put back to (best guess...) what they were previously, and saved again after the new battery was installed?

Where does booting stop now? What's on the screen?

Did you try booting to safe mode? Booting with debugging/logging/whatever shown on-screen to note where the process stopped, or outputting that to a log file?

Do you have an OS partition backup that you can restore? Gotta have one of those!

How old is the PSU? Just for the heck of it, I would pop it open, clean any dust out, and examine the 5VSB subcircuit's capacitor for signs of venting. Just thinking that if this box is mostly off but receiving power, then wear is continuing on the PSU's 5VSB subcircuit. "Cold start by PS/2 keyboard" (necessarily?) means the PSU has had extra load on 5VSB any time it's receiving wall power... not much but if the PSU is pretty old or lower than median quality grade, it's worth a look.

 

[Mantrid-Drone]

Thanks for the response.

I was going to post all that sort of information but I decided that if it turned out it was the simple resetting of AHCI to IDE problem described the information would be a bit redundant.

On reboot the BIOS had been set back to version f3 (GA Z77-DS3H) but I don't have a record of what BIOS I was actually using. Whilst I do not remember updating it I'm fairly sure that was not the BIOS version it had before this problem occurred.

After the CMOS battery had been replaced I tried booting again with that but as I expected it stalled again at the Windows loading logo, in fact just at the point it starts to appear. I tried re-seating the battery just to make sure that wasn't the problem - it wasn't.

I went through the whole Windows diagnostic/repair reboot process, twice just to make sure and the same single, unspecified, unrepairable error was flagged as the problem again each time.

The hours spent involved reflashing the BIOS with a fresh download, using another PC obviously, from the Gigabyte web site. Doing that and Installing all those available now onto a clean FAT32 formatted flash drive took some time.

When I eventually got to the BIOS Q flash facility I found that three out of the four BIOS were rejected as not being the correct size. I realised I made a mistake unpacking the 7Zip files so had to go back to the other PC to fix that.

In the end I actually reflashed the BIOS with version f10f which is marked as Beta. That went well but on reboot exactly the same problem. I then rebooted into the BIOS and, as recommended in the MB manual after a BIOS flash, used the reset to optimized defaults option, saved and rebooted to the same damned error again.

I've tried using SFC then CHKDSK with repair /f from the Windows Safe Mode menu options using CMD , no joy. Just rebooting in Safe mode likewise. Tried a restore point from the previous week before the problem occurred, again no joy.

Lastly tried the EASEUS recovery disk/flash drive I'd create over 3 years ago just after I replaced the original 128GB SSD with a 500GB one. Very simple to use and all went well. It replaced the MBR partition as well as the primary drive. MBR damage I thought might be the problem; apparently not.

That's when I gave up for the day.

I'll check the BIOS settings for the AHCI status problem tomorrow. I hope it is just that but now I'm worried my boot fix attempts may have caused additional problems,

The PSU is a 550W one that came with the CoolerMaster case so probably of the 'cheap' sort you suggest. However there are no obvious power supply symptoms and it is way over spec for the system so has never been stressed. Whilst 'old' it has been used comparatively little over the years compared to other PCs too.

In any case the PS/2 keyboard start has been off or not used ever since the problem first occurred. It was reset to off when I saved the 'optimized defaults so I think a related PSU issue is unlikely. But I'll consider it being that after I've checked the AHCI thing.

If not that I'll try reinstalling the original 128GB SSD (system cloned using EASEUS) in case it is an hardware issue with the 500GB one.