Errr, weird motherboard explosion!

[james_2k]

Hi all,

wondering if anyone has had a similar experience. was stress testing my cpu with Linx, and the computer made a loud bang and turned off, naturally i thought it was the psu.

but the pc turned back on and was working fine, which was weird considering the acrid smell which pervaded the room..

i thought it was a bit dodgy so i restress tested it (the psu is under warranty and i wanted it to die now if it was going to at all!) and it banged again, large cloud of smoke, and two capacitors came flying off the mobo and landed in the bottom of the case.!
weirder still, the pc is still working!

anyway im RMAing the motherboard, but i cant imagine why it blew like that (2months old) and no voltages were exceeding maximum..

picture here (http://gigglehd.com/zbxe/files/attach/images/236/849/852/001/rampageii_gene.jpg) the caps which blew off are the three above the cpu (not the small ones just above it) and the heatsink style arrangement on the mosfets, just to the left of the cpu power connector

anyone got any insights? :)

[CaJaks2]

probably just defective capacitors

[james_2k]

true, but what were they doing, if the pc is still working?

[l4dr0cks123]

Re-do the stress test but record it. Upload on Youtube. Also open your case and put it on it's side to see how high the capacitors can go.

[james_2k]

hehe, sounds like a plan

[Astaco]

This is NOT a good idea.

Contact the support and exchange the defective part.
Running a defective system is dangerous and can damage other parts.

In addition:
If the part is not working correct you can exchange it under warranty.
You don't have to break it completely.

It may be fun for you to blow it up, but as I said before you may damage other components, which will cost you (more) money.

[crispkreme]

If you are planning to make it go completely, I suggest you use old parts you have no use for. That way if it damages them, it wouldn't matter anyway. That is if you have any old parts. If you don't, don't try it!

[james_2k]

hehe no im not. its already scheduled for collection on monday.

but its still bizzare how all three of them went..

[TheMG]

If I were you I'd take a multimeter/voltmeter and measure the voltages that are coming from your PSU. Too high voltages is one possible cause of catastrophic capacitor failure.

Furthermore I believe the caps that blew are on the input side of the voltage regulators, since the computer is still working. This makes the PSU even more suspicious!

So yeah I would NOT plug a new motherboard into that PSU until you can make sure that the PSU is putting out the correct voltages on the 3.3V, 5V, and 12V rails (especially the 12V).

For the record, these solid polymer capacitors don't normally explode during a soft failure. Something really bad must have happened for them to go like that.

[james_2k]

hmm, would it be sufficient to use the asus utility which monitors voltages?

3.3v was at 3.33
5 was 4.95
12 was 12.02

they dont vary with load. psu is a corsair tx650.

[marie pavie]

Guess they named it the Rampage for a good reason!

[james_2k]

it sure made a bang in the computing scene

[james_2k]

OK Checked 12v voltages with occt, hwmonitor and asus pc probe (and the bios) and all of them say between 11.94 and 12.03.

but speedfan says 11.3...

i assume that an undervoltage wouldnt cause a bang? and how likely is it for speedfan to be right and all those other things wrong? :S

[MikeBlaszczak]

Another victim of overclocking?

[RedOctober45]

If I were you I'd take a multimeter/voltmeter and measure the voltages that are coming from your PSU. Too high voltages is one possible cause of catastrophic capacitor failure.

Furthermore I believe the caps that blew are on the input side of the voltage regulators, since the computer is still working. This makes the PSU even more suspicious!

So yeah I would NOT plug a new motherboard into that PSU until you can make sure that the PSU is putting out the correct voltages on the 3.3V, 5V, and 12V rails (especially the 12V).

For the record, these solid polymer capacitors don't normally explode during a soft failure. Something really bad must have happened for them to go like that.

Even if the polarity is backwards on them, they won't explode?

[james_2k]

Another victim of overclocking?

its an overclockers motherboard and i would be most unimpressed if it exploded for that reason!

it has lights in various parts of the board to display a colour based on the voltages. green safe yellow caution and red, danger. they were all green. i dont like overvolting too much.

anyway i will watch my new board very closely i think.

[Enigma776]

Another victim of overclocking?

Would be my guess to, I do not see the need to overclock every bit of hardware just becuase you can, even if it was an overclockers motherboard I would still not put the extra stress on my hardware.

[SmudgePot]

My guess is faulty assembly. I would lean toward the caps grounding out and the massive shunt to ground popping the caps. You did use standoffs, right? You did check that there is clearance underneath the mobo, right? My prediction at the cause of failure is the mobo vibrating when the fan speeds ramped to full during OC and the bottom of the mobo touched ground on the trace/pads for some caps and they shunted voltage to ground and blew. This is common if the grounds arent good and the mobo flexs or is installed wrong. But its only a guess.:rolleyes:

[TheMG]

Even if the polarity is backwards on them, they won't explode?

They probably would, but that's not a "normal" failure. Catastrophic failures caused by overvoltage or reverse polarity can indeed be rather violent.

"normal" failure usually doesn't involve any physically recognizable signs for this type of capacitor.

However there is the less likely failure that does involve the capacitor going short, and with the current available from the computer power supply, it could definitely have an "explosive" behavior.

[Enigma776]

They probably would, but that's not a "normal" failure. Catastrophic failures caused by overvoltage or reverse polarity can indeed be rather violent.

"normal" failure usually doesn't involve any physically recognizable signs for this type of capacitor.

However there is the less likely failure that does involve the capacitor going short, and with the current available from the computer power supply, it could definitely have an "explosive" behavior.

All I know is the testers have a nice bit of kit to tell them what the motherboard was doing before it went the way of the dodo. They will soon know if the board was abused or not from there own tests.

[SmudgePot]

All I know is the testers have a nice bit of kit to tell them what the motherboard was doing before it went the way of the dodo. They will soon know if the board was abused or not from there own tests.

Engineers have a nice way to tell. Board testers are paid $8/hr and dont give a crap. They are more worried about getting high again at their next break.

[kdawgmaster]

its an overclockers motherboard and i would be most unimpressed if it exploded for that reason!

it has lights in various parts of the board to display a colour based on the voltages. green safe yellow caution and red, danger. they were all green. i dont like overvolting too much.

anyway i will watch my new board very closely i think.

just because its an overclocking board dosnt mean u wont run into some issues. what was ur CPU, PSU and the core volts u were going to?

[YahooElite]

a PC with ANYTHING exploding or popping i would automatically suspect the PSU above anything else, especially when it comes to overclocking, the whole overvoltage protection blah blah etc etc type stuff and so on.

[james_2k]

ok to run through some points raised:
-im not buying a high end board to run stock.
-It was certainly properly cleared from the case, but there are no fans around the motherboard to vibrate it
-cpu is the 920 @ 1.38v psu is a 650tx corsair. and never goes above 65c core
-id love for the engineers to find out what went wrong, 'enigma776' because i know for a fact i wasnt as you say; 'abusing' it.
-before i RMAd the motherboard i found (remember it was still working) we did have a multimeter and the tests showed 5.03v 12.03 (12 under load) and 3.4 on the rails.

[MikeBlaszczak]

3.4 volts on the 3.3 volt line is almost out of tolerance; the 3.3 volt line has a +/-5% tolerance in the ATX spec, which means the safe range is from 3.135 to 3.465 volts. Depending on the accuracy of the meter you used, you might actually be out of tolerance. Since your other lines are so close to nominal values, the 3.3 volt line is further suspect.

Have you done any analysis to see which rails the failing capacitors were tied to?

I think the main issue is still your overclock, though.

[Enigma776]

Where did I say you abused the board or not all I said was "They will soon know if the board was abused or not from there own tests"

Like you I would be very intrigued to find out what actually caused this to happen.

If you have the board out of the case i would take a close look at some of the other components as well to see if there bloated or weakened I would also check for singe/arc marks or faulty solder while I was at it..

[james_2k]

Where did I say you abused the board or not all I said was "They will soon know if the board was abused or not from there own tests"

Like you I would be very intrigued to find out what actually caused this to happen.

If you have the board out of the case i would take a close look at some of the other components as well to see if there bloated or weakened I would also check for singe/arc marks or faulty solder while I was at it..

ok my bad. getting a lil defensive i guess, Thanks for the help :)

when the mobo returns i will test the psu again and report back if they say anything.

i do actually have 2 multimeters now (ordered one then found out we had one) so will test with both. but i wouldnt know where to start to discern what voltage caps they were im afraid mike.

[james_2k]

ok i got the motherboard back and ran some tests on my psu. its fine.

however i noticed the caps which blew get really hot, i mean, burn your fingers if you touch them hot.

so installed a nice slow fan over them and bingo, they are cold now.

this is in an antec 900 case, water cooled so no cpu fan there ofc.

take this as a lesson to pay attention to your motherboad and any hotspots it may have :)

class dissmissed

[Sassycat3172]

Another victim of overclocking?

Exactly. which is alwys 1 of 4 things.

#1 Uneducated overclocker.
#2 Pushed the OC to far.
#3 Bad components.
#4 Inadeqaute cooling, see #1 & #2

[james_2k]

Exactly. which is alwys 1 of 4 things.

#1 Uneducated overclocker.
#2 Pushed the OC to far.
#3 Bad components.
#4 Inadeqaute cooling, see #1 & #2

shooo shoo. ive been overclocking for years, and we never stop learning.

[Sassycat3172]

... ive been overclocking for years, and we never stop learning.

Aint that the truth.:)

[james_2k]

:) .