System fails to POST. Motherboard failure?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
- Friend of SPCR
- Posts: 279
- Joined: Sun May 06, 2007 10:37 am
System fails to POST. Motherboard failure?
I have a small file server I keep running all the time, and just overnight it mysteriously shut itself off. This morning when I tried turning it on, I got absolutely no video output. No bios messages, nothing. After tinking around a bit, I'm inclined to believe the motherboard is just dead. Here's what I tried so far:
1. Swapped out the ram. No luck there. The alternative set I used is a pair of Ballistix Tracers with the fancy LEDs that give you a light show with any activity. These lights were completely off, so this gave me the impression that the failure is at a very basic hardware level.
2. Swapped out the PSU. No change.
At this point the only major point of failure left to test is the cpu, but that would involve taking it out and disassembling another good system to test it in. I don't have an extra motherboard handy, so it will be a real pain. What do you guys think?
EDIT: I found a pair of suspect capacitors.
The tops are slightly bulged out.
There is this brown spot on top. Hard to tell if it's discharge or just sharpie marker. I did buy the motherboard new, so I'm leaning toward it being a leak.
1. Swapped out the ram. No luck there. The alternative set I used is a pair of Ballistix Tracers with the fancy LEDs that give you a light show with any activity. These lights were completely off, so this gave me the impression that the failure is at a very basic hardware level.
2. Swapped out the PSU. No change.
At this point the only major point of failure left to test is the cpu, but that would involve taking it out and disassembling another good system to test it in. I don't have an extra motherboard handy, so it will be a real pain. What do you guys think?
EDIT: I found a pair of suspect capacitors.
The tops are slightly bulged out.
There is this brown spot on top. Hard to tell if it's discharge or just sharpie marker. I did buy the motherboard new, so I'm leaning toward it being a leak.
-
- Moderator
- Posts: 7681
- Joined: Mon Dec 09, 2002 7:11 pm
- Location: Maynard, MA, Eaarth
- Contact:
Re: System fails to POST. Motherboard failure?
The brown spot is ooze from inside the capacitor, and yes this usually means the motherboard has failed. You can replace the capacitor(s) and possibly revive it? There is a thread here on SPCR about doing this.
Another possibility is replacing the video card?
Another possibility is replacing the video card?
Re: System fails to POST. Motherboard failure?
those caps do look suspicious, but caps in general tend to leak out the bottom.
what you are looking for here is the beep code that's present when you first turn on the computer... you should be able to pull the ram and still get a beep code.
what you are looking for here is the beep code that's present when you first turn on the computer... you should be able to pull the ram and still get a beep code.
Re: System fails to POST. Motherboard failure?
How old is the mother board? What is the CPU?
What kind of case, with what kind of heat sources and what kind of cooling has it been run in?
What kind of case, with what kind of heat sources and what kind of cooling has it been run in?
-
- Friend of SPCR
- Posts: 279
- Joined: Sun May 06, 2007 10:37 am
Re: System fails to POST. Motherboard failure?
I used onboard video, so no separate video card that can be replaced.NeilBlanchard wrote:Another possibility is replacing the video card?
I don't have case speakers to generate the beep code. I have this cheapo chinese pci card that's supposed to display an error code, but even that doesn't show anything. The "reset" light on it is constantly on though, which is making me think it's the motherboard.danimal wrote:what you are looking for here is the beep code that's present when you first turn on the computer... you should be able to pull the ram and still get a beep code.
Motherboard was bought in August of 2009. Cpu is an Intel e2140. Case is the Antec NSK 1380. The system has just been a basic air-cooled file server. I have a heatsink on the cpu, and it's running off onboard video so there's really not a lot of heat in this system. The hottest components were the hard drives at like 44C in the summer.ces wrote:How old is the mother board? What is the CPU?
What kind of case, with what kind of heat sources and what kind of cooling has it been run in?
Re: System fails to POST. Motherboard failure?
looks like a ZOTAC NF610I-E-E? sure is ring chokes and chinese caps in thar.
have you cleared the cmos?
try it out with just one stick of ram. try the other slot, etc.
if you cant get it to post, then you might have to try reseating the cpu. after that i dont know of much else to try really.
have you cleared the cmos?
try it out with just one stick of ram. try the other slot, etc.
if you cant get it to post, then you might have to try reseating the cpu. after that i dont know of much else to try really.
Re: System fails to POST. Motherboard failure?
That is probably not true.PartEleven wrote:The hottest components were the hard drives at like 44C in the summer.
See those white donuts with the large gauge copper wire wrapped around them?
I have one of them on a motherboard. Mine has about 15 turns of wire compared to your 6 turns. At idle I estimate mine runs about 65C. (it feels about as hot as an incandescent lamp that has been turned on and running on for a while).
You have two of them. If yours behaves like mine, yours may have cooked the capacitors located between them.
Does anyone know what these components do... and why not all motherboard have them? My guess is that it performs some kind of transformer like function. Whatever it is, it throws off a lot of heat while doing it.
-
- Moderator
- Posts: 7681
- Joined: Mon Dec 09, 2002 7:11 pm
- Location: Maynard, MA, Eaarth
- Contact:
Re: System fails to POST. Motherboard failure?
The toroidal "donuts" are chokes aka coils aka inductors. They function as low pass filters, if I am not mistaken. Capacitors are high pass filters. The number of windings has a direct affect on their values -- which is measured in microhenrys. Capacitors are measured in microfarads.
They should not be bulging, or oozing. Those capacitors are failed, or are nearly failed. They are probably part of the voltage regulation, and if they are failed or nearly bad, you will get no post. If you are handy with a soldering iron and can desolder them, then you can replace them with new capacitors, and it may well work again. They don't cost much -- maybe 60 cents, maybe $2? There is a thread here on SPCR with a how to -- the motherboard that was used was one of mine, and I mailed it to him and he got a discontinued motherboard for essentially the cost of shipping.
They should not be bulging, or oozing. Those capacitors are failed, or are nearly failed. They are probably part of the voltage regulation, and if they are failed or nearly bad, you will get no post. If you are handy with a soldering iron and can desolder them, then you can replace them with new capacitors, and it may well work again. They don't cost much -- maybe 60 cents, maybe $2? There is a thread here on SPCR with a how to -- the motherboard that was used was one of mine, and I mailed it to him and he got a discontinued motherboard for essentially the cost of shipping.
Re: System fails to POST. Motherboard failure?
toroidal ring chokes. yep, they're low pass filters.
not all motherboards use the ring style chokes because there are better performing options. solid state chokes can handle higher loads, and typically have reduced electrical noise. you still see ring chokes being used on new motherboards sometimes though, probably a cost saving measure.
for instance, on the op's board (if its the one i think it is), the voltage regulation for the cpu is handled by solid ferrite chokes (maybe an "R68M", if op would confirm?) but it looks like the memory voltage regulation was handled by the ring chokes and adjacent caps. your psu might have had too much ripple or was simply dipping out of atx spec, and the v-reg system couldnt handle it anymore...
btw: which psu were you using, how old is it, and how much has it been stressed over its lifetime?
not all motherboards use the ring style chokes because there are better performing options. solid state chokes can handle higher loads, and typically have reduced electrical noise. you still see ring chokes being used on new motherboards sometimes though, probably a cost saving measure.
for instance, on the op's board (if its the one i think it is), the voltage regulation for the cpu is handled by solid ferrite chokes (maybe an "R68M", if op would confirm?) but it looks like the memory voltage regulation was handled by the ring chokes and adjacent caps. your psu might have had too much ripple or was simply dipping out of atx spec, and the v-reg system couldnt handle it anymore...
btw: which psu were you using, how old is it, and how much has it been stressed over its lifetime?
-
- Friend of SPCR
- Posts: 279
- Joined: Sun May 06, 2007 10:37 am
Re: System fails to POST. Motherboard failure?
Close enough. It's actually the NF630I-E-Efumino wrote:looks like a ZOTAC NF610I-E-E? sure is ring chokes and chinese caps in thar.
have you cleared the cmos?
I haven't tried clearing cmos yet, but I might just RMA this board since I think it's still under warranty.
I used a pico-PSU-120WI-25V with an Antec NP-100.fumino wrote:btw: which psu were you using, how old is it, and how much has it been stressed over its lifetime?
-
- Moderator
- Posts: 7681
- Joined: Mon Dec 09, 2002 7:11 pm
- Location: Maynard, MA, Eaarth
- Contact:
Re: System fails to POST. Motherboard failure?
Sure -- RMA it if you can! Hopefully they have improved the quality of capacitors?
Re: System fails to POST. Motherboard failure?
Yeah... just rma it.NeilBlanchard wrote:Sure -- RMA it if you can! Hopefully they have improved the quality of capacitors?
To me it looks like a design defect. The capacitors got cooked between those two coils. They really need their own little 80mm fan.
Re: System fails to POST. Motherboard failure?
the beep code comes through the tiny speaker that's attached to most motherboards; if the mb doesn't have the speaker, there is usually a connector on the mb where you can plug any speaker into it.PartEleven wrote:I don't have case speakers to generate the beep code. I have this cheapo chinese pci card that's supposed to display an error code, but even that doesn't show anything. The "reset" light on it is constantly on though, which is making me think it's the motherboard.danimal wrote:what you are looking for here is the beep code that's present when you first turn on the computer... you should be able to pull the ram and still get a beep code.
no beep code at all is not a good sign.
Re: System fails to POST. Motherboard failure?
Ouch, another Zotac board bites the dust overnight (my GF-8200-C-E did so about a week ago: solid HDD activity light, fans on, no other signs of life). At least the RMA request experience was pretty painless for me (sent a support request "just in case" and was pretty much told with what I'd tried it was the motherboard, submitted an RMA request, and within 24 hours it was approved). Currently waiting for the motherboard to make its way to CA to hopefully be repaired/replaced.
Hopefully we both get satisfaction from the RMA process (too many ways you can be denied as something was "your fault").
Hopefully we both get satisfaction from the RMA process (too many ways you can be denied as something was "your fault").