Power Supply Blew Up

[coloradopc]

I am building a system with an Enermax EG701AX-VE(W) 600W power supply and a new Asus A8N32-SLI MB. After initial assembly, I turned on the PC for the first time and the room lit up with blue sparks out the back of the power supply, which then promptly sent out a cloud of smoke and died. Rather spectacular, really. Besides wishing I had a photo to share, can anyone think of something that might cause such an amazing death in a new hookup? I had a few drives hooked up, as well as supplemental power to the video card via the 6 pin connector. Since none of the connectors CAN be hooked up backwards, I am at a loss on this one. The power supply is fancy and supports 110/220 automatically. Any ideas on what to check or look into further?

I figure I'm off to a PC lab to figure out what survived of the initial install, but this is bugging me that I cannot figure out what might have gone wrong.

All ideas appreciated!
Thanks, Colin

 

[wpshooter]

I don't think you will figure out what was wrong other than you just happened to get a bad power supply. I had it happen to me once. Like you say the real problem is trying to figure out if it damaged anything. You might want to contact Enermax and look into filing a claim with them for any damaged conponents they you may wind up discovering.

Good luck.

 

[FlyingPenguin]

Dead short is the only reason for something like that - been there, done that, bought the T-Shirt. Incredibly unlikely that it was a defective PSU. PSUs are tested when they leave the factory.

The 5 volt or 12 volt rail got shorted to ground. Crimped wire, a power connector poked into a point piece of case metal, maybe you left a stand-off in a place under the mobo where there was no hole, maybe a screw got jammed under the mobo. Seen it happen plenty of times.

PSU is shot of course. Depending on what you shorted on the mobo the mobo may be a goner too. I'd be careful testing the mobo. Removit from the case, disconnect EVERYTHING and connect it to PSU you don't mind killing for testing. CPU and RAM may be okay - test them on another good rig.

 

[wpshooter]

IMO, it is possible that it was nothing he did wrong because when this happened to me all I did was have them send me a replacement power supply of the exact same brand and model and everything worked correctly without me altering anything.

If they are like most large industrial manufacturers, they may do a sample testing of their products for quality control but I have my serious doubts that they test every unit that goes thru their production line.

 

[FlyingPenguin]

Not impossible, I agree, but doubtful.

 

[RebateMonger]

I've had dead shorts in systems (it's easy to do with a floppy drive connector). A decent power supply will sense the overload and shut itself down. Wost case, it'll blow an internal fuse (if it has one). But there should be no sparks.

When sparks fly out of a power supply, it's generally a defective power supply.

 

[mechBgon]

coloradopc, welcome to the Forums, and consult this page and you got those standoffs in there, right? Muy importante.

 

[coloradopc]

Ok, here's an update. I have disassembled everything. Except for the power supply, everything else looks just like it did when it came out of the box. I have no idea what is alive or dead, of course. I have measured the +12 to GND and +5 to GND at the ATX 24pin and it is >1MOhm for the +12, but only about 4KOhm for the +5V. Does anyone have a measure of +5V to GND from a known good, unconnected-to-anything MotherBoard that they can share for comparisons? I would bet that this is telling me everything I need to know - I have a low-impedance connection (short) on the 5V rail.

The next question, why, is tougher. I have the standoffs in there, and now with the MB completely out and unconnected I get the readings above. Of course, if it is fried it's fried without external help now. I can still see no indication of anything shorting anything, which was what I was afraid of. This is an expensive saga. I wish it were an expensive lesson, but I haven't learned anything from it!

 

[stevty2889]

I hope you didn't mix up the 4 pins of the 24pin connector with the 4pix 12v atx connector(2 yellow and 2 black wires), thats one possiblity, but only if the 4pins are detachable from your 24pin connector. Quality PSU's usualy have protection, so it's unlikely that anything else got damaged. Really does sound like there was something shorted in the power supply, and that it came that way.

 

[stevty2889]

Originally posted by: coloradopc
I have measured the +12 to GND and +5 to GND at the ATX 24pin and it is >1MOhm for the +12, but only about 4KOhm for the +5V.

Were you measuring on the motherboard or on the power supply? Either way, it's not going to tell you much, as there are many differant components connected to those rails, so unless you got something like a 100 ohm or less resistance it's not likely shorted.

 

[RebateMonger]

One thing I'll do with a new motherboard....

BEFORE putting it into the case, install the CPU, memory, a video card, and a keyboard and plug in the power connectors. Set the motherboard on some cardboard.

Power it up and see what happens.

This avoids the "tragedy" of spending an hour screwing the motherboard into the case, attaching drives, LEDs, USB, etc., and then finding you have a defective motherboard. If you get to the POST screen, you know that everything is, basically, OK and can finish up building the PC.