dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
1226
share rss forum feed


Lauriel
Premium
join:2003-05-25
Sacramento, CA

motherboard caught fire after system battery replaced

I have an older system that stopped POSTing the other day. I would get all fans and LEDs, but no POST beep and no video.

RAM (4 sticks,) video card, CPU all checked out fine in another system.

I breadboarded (kept in CPU/heat sink/fan, 1 stick RAM, video card, PS) and checked the following:

If I took out all the RAM, I would get the correct beep code for no RAM. This was true with or without video card in.

If I put the RAM back in (any stick, even known working stick from another machine,) I would get no beep code - no POST - and no video, either. This is true with or without video card in.

I also used a multimeter to test power supply. It fell into tolerances, but I know that's not the be all and end all.

I did some Googling and decided to reset the CMOS as some people said they had success with this. I figured the battery was old, so while I was at it I put in a new battery.

As soon as I plugged in and turned on the power supply, the board started up on its own, without me needing to short the Power On pins. This was not the case prior to changing the battery. Before that, it DID require that I short the Power On pins to start up. As soon as it started up, the board caught fire between the battery and the CMOS jumper pins.

I am curious as to why this would happen. I'm looking at the board right now, and the battery is correct, and seated correctly. Why did this not happen with the old battery? Was there possibly a voltage regulation issue with the board the whole time causing all the issues, and it just culminated in this event after getting a new battery?

The board is an older board - Intel D945pvs with a Pentium D 830 chip and 4 sticks of DDR2 RAM. The board was NOT overclocked at all.

So, any thoughts? Thanks in advance for any help/insight people can offer.


Hank
Searching for a new Frontier
Premium
join:2002-05-21
Burlington, WV
kudos:3
Did you check the battery before installing? Did you install the battery correctly?


Lauriel
Premium
join:2003-05-25
Sacramento, CA
Hello, it was a brand new battery, and I did check to see that it was installed correctly. I'm looking at it now, and it's definitely in correctly. I didn't check the battery before putting it in, I guess just figuring that a brand new battery should be okay. Probably not always true, though, eh?


Lauriel
Premium
join:2003-05-25
Sacramento, CA
reply to Hank
Just took a multimeter to both the old and the new batteries, and they both read at 3V. New one was definitely installed/seated correctly.

r2d2

join:2012-05-29
reply to Lauriel
Possibly a trace or something shorted the board out causing a fire. But i think your system is toast and will need a new one, Puts a new meaning to who let the smoke out lol

asdfdfdfdfdf
Premium
join:2012-05-09
kudos:3
reply to Lauriel
I would say the board was ready to give up the ghost and that's probably why you had problems before. To be on the safe side i would also not try to reuse the power supply on that machine either.


Lauriel
Premium
join:2003-05-25
Sacramento, CA
reply to r2d2
I know, right? I'm not going to use the same power supply, either. I just came back from buying a new power supply. The old one was about 7 years old anyway. I didn't find a board I wanted for a replacement, but I'll look online.


Lauriel
Premium
join:2003-05-25
Sacramento, CA
reply to asdfdfdfdfdf
Yeah...it was a pretty old board. Since the RAM, CPU and video card are still working fine, I'm replacing the board and power supply. It's not my main machine, but I wanted to make a file server out of it. It wasn't getting used regularly anymore.


chrisretusn
Retired
Premium
join:2007-08-13
Philippines
kudos:1
Reviews:
·PLDT
·Comcast
reply to Lauriel
said by Lauriel:

I breadboarded (kept in CPU/heat sink/fan, 1 stick RAM, video card, PS) and checked the following:

Breadboarded? I guess what you meant to say was disconnected everything except.... I have never heard breadboard used this way before. Before I retired I worked in electronics (40+ years), perhaps things have changed.

said by Lauriel:

I also used a multimeter to test power supply. It fell into tolerances, but I know that's not the be all and end all.

You are correct, using a meter on the power supply really doesn't tell you much.

said by Lauriel:

I did some Googling and decided to reset the CMOS as some people said they had success with this. I figured the battery was old, so while I was at it I put in a new battery.

As soon as I plugged in and turned on the power supply, the board started up on its own, without me needing to short the Power On pins. This was not the case prior to changing the battery. Before that, it DID require that I short the Power On pins to start up. As soon as it started up, the board caught fire between the battery and the CMOS jumper pins.

Scary stuff in the above quote (bold and underline is mine). That sentence explains why you smoked your motherboard. Perhaps I am misunderstanding this, but why in the heck would you need to short out the Power On pins?? You state "it DID require", define "it". Certainly not the manual, if so what page? There is absolutely no reason to do this. To replace the battery, you simply replace it with the correct one (CR2032). Clearing (reseting) the CMOS is accomplished by removing the battery and waiting (about 60 minutes or so). There is no requirement to short any pins.
--
Chris
Living in Paradise!!


Lauriel
Premium
join:2003-05-25
Sacramento, CA
Hi Chris,

I had the board completely out of the case, with only 1 stick of ram, the CPU (and of course heat sink and fan) and a video card attached. (Incidentally, that's what I meant by breadboarded.) So, there was no power button to turn the machine on. The only way to start the machine when in this condition is to tap the power switch pins with a screwdriver. This isn't entirely uncommon. I've read about and heard about doing this quite a bit. It had nothing to do with the clearing of the CMOS, but was simply a means to start up the machine when not in a case.

At any rate, there were some issues with the board prior to the incident (as evidenced by my description prior to it occurring,) and I later found what happened to cause the short that started the fire. It was because I forgot to replace my CMOS jumper back to the regular (non-clear) position before turning the power supply back on. Got distracted or not paying close enough attention or something. Embarrassing error. Fortunately, my attached components (RAM, CPU, video card) are doing fine. I'm getting another board and power supply. The machine is a secondary machine that I decided to use for a project, but I had that board for a long time. Well, I guess the project will require a bit more than I originally planned now. But, I'll sure pay closer attention from now on.


chrisretusn
Retired
Premium
join:2007-08-13
Philippines
kudos:1
Reviews:
·PLDT
·Comcast
Not uncommon? OK, if you say so. Glad I retired. My choice would have been to leave it in the case. If that wasn't an option then at least use a switch taken out of an old case. That is my opinion anyway. For what it's worth, there is no CMOS jumper on the motherboard you stated you have (Intel D945pvs). There is a three pin BIOS Configuration Jumper located near the battery. Normal BIOS is jumper on pins 1-2, Configure BIOS is jumper on 2-3 and no jumper sets BIOS recovery mode. I fail to see how this would cause a short and fire.

Oh well, good luck with building your next machine. Replacing a older motherboard these days often means new everything. New memory most likely. If you have EIDE hard drives, CD drives make sure the new motherboard supports EIDE, some do not. Some motherboards no longer support PS2 for keyboard or mouse either.
--
Chris
Living in Paradise!!


Lauriel
Premium
join:2003-05-25
Sacramento, CA
When I said CMOS jumper, the three pins to which you were referring is exactly what I was talking about. I called it CMOS jumper. On the board it's referred to as CLR CMOS. I do realize that the manual refers to it as BIOS Configuration Jumper. (I really did read the manual.) I do know what the normal and the clear settings are. (I have both the board and the manual in front of me right now.) I know it was my error to start the machine up with the pins in the wrong configuration. A few folks do think that this could definitely have caused the short/fire, especially since the board already appeared to have issues. Whether it did or not, I guess I'll never know 100%.

Anyway, as I said, it's not my main machine. My main machine is an X58 setup, which is obviously not compatible with a D945 setup, for example. So, I do know that if I wanted to upgrade the board that went down that I'd be looking at basically building from scratch. I know how quickly compatibility changes as technology progresses LOL. But I have been able to locate some new compatible-with-my-RAM-and-CPU boards (yes, brand new in box) at our local large electronics store. So, all is not lost if I choose to rebuild what I've got with compatible (new) parts. They're still out there.

Thanks for your input.


aurgathor

join:2002-12-01
Lynnwood, WA
kudos:1
Reviews:
·Frontier Communi..
reply to Lauriel
I highly doubt that an improperly set BIOS configuration jumper would cause a motherboard to catch fire. A more likely cause is that you may have accidentally damaged the board somewhere, or grounded/shorted a power trace somehow.
--
Wacky Races 2012!


Lauriel
Premium
join:2003-05-25
Sacramento, CA
Yeah...I've had mixed feedback on that one. Some people suggested that the jumper in the clear position is what led to the "final blow" shall we say. Others say no. I've not heard of it happening before, but I'm no pro, just a hobbyist, so I wasn't sure one way or the other. I do know I left it in the clear position, and really shouldn't, so I am erring on the side of "I was wrong," when saying that perhaps that caused/contributed to the issue. The board already appeared to have a short or some other kind of damage, based on the way it was behaving, so....I don't know for sure. For all I know, it was gong to literally smoke, anyway, based on that damage, and it just happened to coincide with my jumper move. It surely wasn't working properly before the jumper move, either.

At this point, I guess I'm just glad the CPU and RAM are okay, in case I do decide to reuse them in another application, since I have found everything I'd need to do that.


psafux
Premium,VIP
join:2005-11-10
kudos:2
reply to Lauriel
Im guessing when the battery was inserted it delivered just enough power (or delivered it to the right place) to close a circuit that was, for some reason, open. The resulting arc caused the fire.

Just a semi-educated guess.


aurgathor

join:2002-12-01
Lynnwood, WA
kudos:1
Reviews:
·Frontier Communi..
reply to Lauriel
When I operate a mobo not properly installed in its case, especially if the PS is still in the case so the mobo has to be on top of the case, or very close to it, I normally place a 1/4" thick ply under the mobo to make absoluteley sure that I'm not shorting something to the case.

In any case, if I have to assign probabilities to your mishap, I'd say there's a better than 90% chance that you ended up shorting something.
--
Wacky Races 2012!


Lauriel
Premium
join:2003-05-25
Sacramento, CA
Yeah...a short, it surely seems to be. With the way the machine was acting BEFORE it ever came out of the case (described in the original post,) I have to wonder if there wasn't a short going on at that point. When I took everything out of the case, the power supply also came out of the case. The board and power supply were nowhere near the case at that point, and were both on top of a thick slab of wood (a wooden table.)

I don't know if the battery replacement was just merely coincidence or not. It was definitely in correctly. I check before and after the fact. I wonder about psafux's response with regard to that? If the old battery were weak, could the new battery have given enough power to send an existing issue over the edge, so to speak?

I guess at this point, my curiosity lies more in the original problem (what was causing THAT,) along with any relation between the original problem and the final spark. I don't know enough about that and would be interested in peoples' takes on all of that.

Thanks!


signmeuptoo
Bless you Howie
Premium
join:2001-11-22
NanoParticle
kudos:5
reply to Lauriel
It's also an old system with an old board. Every hear of tin fingers? They are tiny, microscopic (sometimes) solder fingers that shot things out, they grow like stalactites.


aurgathor

join:2002-12-01
Lynnwood, WA
kudos:1
Reviews:
·Frontier Communi..
Tin fingers will simply act as a low amperage fuse and blow harmlessly when there's any significant current. To cause a fire, there had to be a major short.
--
Wacky Races 2012!