Are My BIOS Settings At Risk When Changing My Motherboard Battry?

[malletKATman]

I have a custom-built, 10 year old desktop computer using the ASUS AMD-chip M4A88TD-M motherboard. Here are all the pertinent specs (I hope) for my system obtained by the "Speccy" app.

Operating System
Windows 7 Professional 64-bit SP1
CPU
AMD Phenom II X6 1090T
Thuban 45nm Technology
RAM
16.0GB Dual-Channel DDR3 @ 803MHz (9-11-11-29)
Motherboard
ASUSTeK Computer INC. M4A88TD-M (AM3)
Graphics
FHD2401 (1920x1200@60Hz)
3067MB ATI Radeon HD 4250 (ASUStek Computer Inc)
Storage
931GB Samsung SSD 860 QVO 1TB ATA Device (SATA (SSD)) (C: Windows Partition)
931GB Seagate ST31000528AS ATA Device (SATA ) (E:)
3725GB Western Digital WD easystore 25FA USB Device (USB ) (F:)
The C: drive is all Windows Partition, the E: drive has all the User Profiles, the F: drive is for backup/restore and system restore points
Optical Drives
ATAPI iHAP222 9 ATA Device
Audio
AMD High Definition Audio Device
Operating System
Windows 7 Professional 64-bit SP1
Computer type: Desktop
Installation Date: 7/5/2010 9:45:39 AM
Serial Number: JY467-933T7-YWJWM-4CP8W-H6QJB
Windows Security Center
User Account Control (UAC) Disabled
Firewall Enabled
Windows Update
AutoUpdate Not configured
Windows Defender
Windows Defender Enabled
Antivirus
Antivirus Enabled
Company Name Webroot
Display Name Webroot SecureAnywhere
Product Version 9.0.24.49
Virus Signature Database Up to date
Antivirus Enabled
Company Name Webroot
Display Name Webroot SecureAnywhere
Product Version 9.0.24.49
Virus Signature Database Up to date
.NET Frameworks installed
v4.7 Full
v4.7 Client
v3.5 SP1
v3.0 SP2
v2.0 SP2
Firefox (default browser) V67.0 (64-bit)
Internet Explorer
Version 11.0.9600.19374
PowerShell
Version 2.0

About three weeks ago the Windows 7 Op/Sys clock stopped running overnight with the power off the computer. Then about a week ago, it resumed maintaining accuracy overnight with the power off. Go figure.

I attributed this probably to the CR-2032 Lithium 3V motherboard battery being on the verge of failire, because the battery has not been change since I purchased the computer. I downloaded the ASUS document describing how to change the battery, but I am hesitant to change it without knowing the effect on the BIOS settings without any power on the motherboard. The document does not mention any risk to the BIOS settings, but what is the purpose of the battery except to maintain the BIOS when the AC power is off? I need to know the risk of losing my extensive BIOS settings by changing the battery before proceeding.

 

[Digerati]

but what is the purpose of the battery except to maintain the BIOS when the AC power is off?

It is not the BIOS the battery is maintaining - that is set in firmware. It is the user changes to the BIOS defaults that are saved in the CMOS memory module the battery is maintaining. For most users, the only changes are the date and time settings. Other common changes might be the boot order for your disks. In other words, there typically is nothing to worry about. You change the battery then boot directly into the BIOS Setup Menu. You set the date and time, make sure your boot order is correct, then Save and Exit and be good to go.

But note most BIOS menus let you save your settings to a file you can restore later. So if you manually made a bunch of changes, you might you want to look at that. But if you didn't make a bunch of changes, I would not worry about it.

Note the battery also keeps the RTC (real time clock) ticking when power is removed. This is just a counter and when you set the Date and Time in the BIOS Setup menu, you simply sync the RTC to your local date and time.

10 years is a long time for those batteries to last so I would say you've been lucky there and yes, it is time to replace it. Just be sure to unplug the computer from the wall and touch bare metal of the case interior BEFORE reaching in to discharge any static in your body. Observe the battery polarity before removing (it only goes one way but can be forced the wrong way - not good). Do not touch the new battery with your bare fingers; skin oils attract dust and can promote corrosion. While in there, make sure the computer free of heat-trapping dust and make sure you didn't knock any cable connections loose. I would give each fan a little flick with your finger to make sure it still spins freely and the bearings are not beginning to freeze - 10 years is a long time for them too. Periodically keep touching bare metal as you can build up static in your body just by squirming in your clothes.

Then connect power and as noted, boot directly into the BIOS Setup Menu to set the date and time, and make any other manual changes you want, then Save and Exit.

One more point - note that mainstream support for Windows 7 ended over 4 year ago and all support ends this coming January. It is time to let W7 go. But since your computer hardware is already beating the odds too, it is time to starting looking at replacing your hardware too. At the very least, make sure you have current backups of your data.

 

[malletKATman]

Thank you so much for your prompt and complete answer to my question. I have been inside the covers to swap out my C: SSD drive, clean the interior and expand the RAM memory over the course of those 10 years so I am familiar with the precautions. The fans are running smoothly without any bearing noise so far, and the internal temperatures have remained well within the prescribed limits. So far the bearings on my 10+ year old Winchester E: drive are good as well.

One more point - note that mainstream support for Windows 7 ended over 4 year ago and all support ends this coming January. It is time to let W7 go. But since your computer hardware is already beating the odds too, it is time to starting looking at replacing your hardware too. At the very least, make sure you have current backups of your data.

I have reasons why I’m not moving off Windows 7, the main one is relying on the hardware assisted Windows XP Mode/Windows Virtual Machine to run database applications I wrote that use an RDBMS that will not run on anyWindows OS newer than XP. The RDBMS has a feature I made great use of: Procedural Static Views, which greatly simplifies coding either at the app level or within the DB schema as SQL Triggers. That RDBMS no longer exists, support for Windows XP VM died with Windows 8, and at age 82 I’m not interested in spending the rest of my life recoding my apps (C and C++). I do not have the reinstallation media for those components. Any new computer I buy can only have Windows 10 on in, on pain of MS death to the vendor that installs anything else.

Also I’m disgusted with the diminishing customizing control that MS continues with each “new and improved” version of Windows. I backup my system religiously using the perfectly functional Windows 7 Backup & Restore system which disappeared in Windows 8 so as not to compete with MS’s evolving Cloud business. There are many of us who think that Win 7 Pro SP1 was the last really fun version of Windows, and that user control has been dumbed down with each new version since. I’m hoping that this Windows 7 box lasts as long as I do . If not, I may switch to Linux. I have a long history as a Unix developer.

 

[niemiro]

I have reasons why I’m not moving off Windows 7, the main one is relying on the hardware assisted Windows XP Mode/Windows Virtual Machine to run database applications I wrote that use an RDBMS that will not run on anyWindows OS newer than XP. The RDBMS has a feature I made great use of: Procedural Static Views, which greatly simplifies coding either at the app level or within the DB schema as SQL Triggers. That RDBMS no longer exists, support for Windows XP VM died with Windows 8, and at age 82 I’m not interested in spending the rest of my life recoding my apps (C and C++). I do not have the reinstallation media for those components. Any new computer I buy can only have Windows 10 on in, on pain of MS death to the vendor that installs anything else.

Hi there,

I can understand and sympathise completely with what you are saying.

I think the most important thing to be aware of is this - old versions of Windows are staggeringly vulnerable, in part because they lack all of the very many (and effective) security mitigations modern operating systems have, and also because they may not be fully patched (depending on when support ends). What this means is that you are at a very high risk of catching malware - ransomware being the latest common tactic.

To mitigate this - make sure you have very effective backups. It is critical, when running old and insecure operating systems, that you are fully prepared at any time to have a ransomware attack which encrypts not just your internal hard disks but also any connected external hard disks and memory sticks as well.

If your disconnected, offsite backups are good enough, this should not concern you too much: get infected, wipe machine, reinstall, no big deal.

If this is a big deal - rethink, and rethink fast, before it is too late.


The best way to run Windows XP software on Windows 10 is to create a Windows XP Mode VHD, and then run it on a Virtual Machine. I personally like VirtualBox, which is free.

Instructions here: Get official Windows XP virtual machine for Hyper-V

All the best.

 

[writhziden]

Windows 10 also still includes Backup and Restore (Windows 7). In the Type here to search text box, type Backup settings and open the Windows 10 Backup settings interface. From there, click on the link to Go to Backup and Restore (Windows 7). I believe all the original features are still intact from the Windows 7 backup engine.

 

[malletKATman]

Windows 10 also still includes Backup and Restore (Windows 7). In the Type here to search text box, type Backup settings and open the Windows 10 Backup settings interface. From there, click on the link to Go to Backup and Restore (Windows 7). I believe all the original features are still intact from the Windows 7 backup engine.

That is news to me. Thanks. I will look for it on my wife's Win 10 notebook.

 

[malletKATman]

Hi there,

I can understand and sympathise completely with what you are saying....

I think the most important thing to be aware of is this - old versions of Windows are staggeringly vulnerable, in part because they lack all of the very many (and effective) security mitigations modern operating systems have, and also because they may not be fully patched (depending on when support ends). What this means is that you are at a very high risk of catching malware - ransomware being the latest common tactic.

To mitigate this - make sure you have very effective backups. It is critical, when running old and insecure operating systems, that you are fully prepared at any time to have a ransomware attack which encrypts not just your internal hard disks but also any connected external hard disks and memory sticks as well.

If your disconnected, offsite backups are good enough, this should not concern you too much: get infected, wipe machine, reinstall, no big deal.

If this is a big deal - rethink, and rethink fast, before it is too late.


The best way to run Windows XP software on Windows 10 is to create a Windows XP Mode VHD, and then run it on a Virtual Machine. I personally like VirtualBox, which is free.

Instructions here: Get official Windows XP virtual machine for Hyper-V

I will look into this, thanks. Can you recommend a source of a high-end desktop where I could order a 1TB SSD for my main C: drive and an excellent audio board/driver setup? (I am a musician.) I don't need game-level video. If writhziden is correct in this thread, I should be able to roll in my many music WAV files as a Restore from my Win 7 backup drive.

 

[Digerati]

What this means is that you are at a very high risk of catching malware -

Not just that you can catch malware but that you may then become a threat to the rest of us. That's the problem with XP and W7 is heading in that direction.

I hear you when you say W7 was the last "fun" version of Windows. But I also understand where Microsoft is coming from.

One of Windows 7's greatest assets was its flexibility.
One of Windows 7's greatest liabilities was its flexibility.

The problem there is, who got the blame when users dinked with W7 so much it broke? Microsoft. And the fact is, today, the vast majority of users are not into customizing Windows. Instead, they just want it to work as reliably as their toaster or TV. And frankly, if users just left W10 alone and left the defaults as is, it does work that reliably for the vast majority of users. The problem is, if even 1% of the 700 million W10 users have problems (and a 99% success rate is outstanding in almost every industry - though it typically is much less than that), that still leaves 7 million upset users. And 7 million upset users can make a lot of noise especially when their complaints are constantly being amplified, exaggerated and echoed by the IT media.

If you like to tinker (and nothing wrong with that in my book) then I would urge you to move to Linux soon rather than later.

 

[malletKATman]

There’s one more issue that gives me pause about changing my motherboard battery. I have made only one CMOS change other than Date-Time since taking delivery of the box and I can easily do it again, but because this was a custom-built box there could be a number of CMOS changes made by the guys that built the box for me of which I am totally unaware. I’ve searched my CMOS setup and I can’t find any Import/Export facility that Digerati described to save all the changes from the firmware to a file.

Unfortunately the people who built the box ten years ago are out of business and scattered in the wind. I’ve taken pictures of all the CMOS screens, and there are a lot of them, and maybe I could get all the possible changes back by hand, but I’m not confident of it.

Those “garage” guys really knew their hardware, so it’s probably unlikely they would have chosen a motherboard that needed substantial CMOS changes to build what I wanted. But not knowing the software that well, they were terrified of making a simple Windows 7 Registry change I knew that makes defining User Profiles off the Windows Partition “Walrus in Goal” easy. (Back then defining users off the Windows Partition was required for reasons I will explain if anyone cares.) So maybe they were terrified of making CMOS changes as well and I’m exaggerating this potential problem.

Any one care to comment?

 

[Digerati]

There’s one more issue that gives me pause about changing my motherboard battery.

You really don't have a lot of choice on this. Your current 10 year old battery will fail eventually. And when that happens, your BIOS will be reset whether you are ready for it or not. So you really need to do it on your terms. Understand a weak battery can result in corrupt file timestamps and that can result in corrupt operating systems. These are not common events, but not unheard of either.

I highly doubt the builders of your system made many changes to your BIOS unless you had them overclock your system manually. And because it is so easy to reset the changes to the BIOS that are saved in the CMOS memory module back to the factory defaults, I highly doubt they were terrified of making any changes either.

If you look on page 2-1 of your manual, it talks about saving and updating your BIOS. This is also discussed on page 2-29 via the Tools submenu of your BIOS Setup Menu.

 

[malletKATman]

If you look on page 2-1 of your manual, it talks about saving and updating your BIOS. This is also discussed on page 2-29 via the Tools submenu of your BIOS Setup Menu.

Thanks again. I don't know how I missed that; I have the motherboard manual on my PC. I will run both forms of CMOS backup (BIOS Flash, external USB memory stick) to get at least one to complete successfully before I shutdown and change the battery. However, there's an old system admin's aphorism about Backup/Restore that goes, "Backups are always perfect until you try to restore from them." , and I'm not going to run the Restore phases just to test them out because that does seem to be a non-destructive test. Perhaps I'm the one whose terrified - of making irreversible changes (for me, at least) to the hardware and CMOS.

 

[Digerati]

I would not do a restore. I would just unplug from the wall, touch bare metal of the case interior, replace battery, connect power and boot directly into the BIOS Setup Menu to set the date and time. If you boot from a disk other than c: change the boot order then Save and Exit.

 

[malletKATman]

I would not do a restore. I would just unplug from the wall, touch bare metal of the case interior, replace battery, connect power and boot directly into the BIOS Setup Menu to set the date and time. If you boot from a disk other than c: change the boot order then Save and Exit.

I was able to save what the motherboard manual calls “the current BIOS file to the BIOS Flash” (there are 8 Profile slots from which one can choose and name), although the manual then says it saves “your current CMOS settings in the Save to sub-item.” I was able to restore the settings as well and then save the changes and exit to the Windows Boot-up. Some random testing, including one of my complex XP VM database apps, shows the system to be working just as before this sequence.

The BIOS Backup/Restore to an external device won’t work on this box because it only supports FAT 32/16 format drives and it can’t recognize a memory stick on an USB port. It only recognizes hard drives (one of which is my USB-connected backup/restore drive) and they are all NTFS formatted. So that BIOS Tool feature is useless to me unless I read/write on my optical drive. Not worth the effort IMO.

So with my current CMOS settings saved to the BIOS Flash, I’ll power down and change the battery, after my better half finishes her Monday-scheduled work on the system.

 

[Digerati]

Sounds like a plan.

 

[malletKATman]

Sounds like a plan.

It was a plan, but it failed. After backing up the computer, I followed all the instructions I got on this thread and the very specific illustrated instructions I found on line at ASUS on how to change the motherboard battery as carefully as I could, carefully grounding myself and the small screwdriver I used to cam open the spring loaded retention clip on the battery. With the skin off I cabled the box minimally and started it up. It hung in an initial sending-reset-to-every-device/board-connection-jack loop (I could hear the optical drive reset endlessly at about a 3 second interval) and could not be broken out of it by any means including the ground-the-reset-motherboard-connector button on the front of the box that forces a reset-and-reboot. That button has never failed to interrupt anything in the boot up sequence before. The only way to break out of the loop was to power off the box. Hoping that perhaps I had installed a dead battery (highly unlikely) I swapped in yet another brand new CR2032-Lit battery I had ordered from Amazon. Same story with the second new battery. It acted as if the BIOS could not access the CMOS settings at all and hung hard in the initial reset loop.

So I was right to fear the process, but quite possibly for the wrong reasons. There are a host of possible explanations for what happened, including that I screwed up the mechanical installation, but I did that very carefully and slowly, following the instructions exactly. I have my own WAG – perhaps a SWAG: The original battery was not dead and removing it created probably the first power-off shock to the CMOS in the lifetime of the computer, which caused the CMOS to fail (and rendering my only CMOS backup in a BIOS Flash sector slot useless.

I found the best looking website for nearby computer repair/data recovery shops and talked to the owner. He recommended replacing the whole motherboard because if the CMOS failed on it, something else was likely to go soon. He looked it up the motherboard and found it was still available and reasonably priced. I accepted his recommendation and dropped my box off in his care, hoping that the one required CMOS change I knew about would be enough to get the VM emulation working for my XP Mode operations. Ten days later, today, I picked up the box and all the ordinary operations worked perfectly (somehow my UserID’s background wallpaper selection was lost which was not true for the other two IDs on the system). On the way up I had made the one CMOS change I mentioned and the XP VM also works perfectly. I got lucky.

We could close this thread now as SOLVED. The answer: leave everything to do with the motherboard to the pros.

I want to thank everyone who contributed to this thread, particularly Digerati and writhziden.

 

[Digerati]

I have never seen the problem you described before and I have changed CMOS batteries literally 100s of times. So I have no explanation.

You said you grounded yourself and that is great - but I don't see where you unplugged the computer from the wall first. Did you do that?

Regardless, removing the battery simply resets the BIOS back to their default settings. It still should have booted past POST (power on self test) and then hit attempt to load an OS from the first boot drive.

So I am really puzzled here and sorry for all your troubles.

BTW, you are right that it is highly unlikely you installed a brand new but dead battery - but twice in the last year or so it happened to me. I fear like so many things mass produced these days, quality assurance before products leave the factory is something cut to keep profits up.

 

[malletKATman]

…but I don't see where you unplugged the computer from the wall first. Did you do that?

Yes. I even changed the battery in a different room from where the box's power cord was.

I actually saw a technician change a CMOS battery on a live box once and get away with it, but that was in my music synthesizer, not a computer, even though the two devices have a lot in common. In the synthesizer (Kurzweill PC2-R), changing the battery loses a whole bunch of programming that accommodates the box to my MIDI controller (Alternate Mode’s malletKAT PRO). Then all that programming has to be downloaded again together with the voice modifications I’ve made, otherwise the controller can’t talk to the synthesizer and vice versa. (Who designs a system like that? Kurzweill does. ) She got away with it and we didn’t have to download the interface programming, which BTW can only be done from an Apple iMAC, not any PC. Needless to say, every five years or so when the battery is close to done, I box up my synthesizer and send it back to Alternate Mode to get the battery changed.

 

[Digerati]

I actually saw a technician change a CMOS battery on a live box once and get away with it

Yes. It can happen. I've seen electricians change light switches and wall outlets on live circuits too and "get away with it" too. Still not safe.

 

[malletKATman]

Yes. It can happen. I've seen electricians change light switches and wall outlets on live circuits too and "get away with it" too. Still not safe.

That's what the annual Darwin Awards are for: to "celebrate" people like that.

 

[Digerati]

Agreed.