I turned my PC off today before dinner, and it was working fine. I came back afterwards, turned it on and I heard a crack, with a smell of frying electrics. The PC rebooted (as it does when it shorts out. I know from when I decided to start wiring up my case lights with the PC on
It doesn't seem to do any damage though, it's happened a good few times before), and my fans stopped working.
The fans are coming through a 5.25" controller, which has a LCD display showing temps and fan speeds. The display (which uses 5v) is still working, but the fans and the LEDs inside the controller (which use 12v) aren't. When it rebooted the first time, I opened up the side with the ghetto wiring, and tapped it a few times. The fans started cutting in and out intermittently then. I figured it was a loose connection, so I re-did all of the joins and connections, and tried again. Still no luck.
All of the hard drives and disk drives are receiving full power, and the multimeter is showing 5v and 12v as expected on molex. The machine refuses to boot either, it stops right before it checks the disk drives for boot disks. It goes through its routine checks, and gives its usual list of hardware and controllers, which are all present. It stops on verifying DMI pool, which usually only takes a few milliseconds. Going into BIOS, it shows all voltages reading fine, and detects the HDDs and disk drives.
When I connect just the 12v molex wire straight onto one of the LEDs in the fan controller, it lights, even if I remove the molex plug feeding it. I'm fairly sure this means the negative is earthing against the case somewhere, as all lights work when just a positive is connected. I've tried the obvious solution of unplugging the molex plugs going to the case lights and fan controller (the only things I wired myself), but it still refuses to boot and earths somewhere. The weird thing is the fans still won't come on even if they're connected to the controller, which to me sounds like one of the negatives going to the controller is earthing along the way. I've traced it along and there's no signs of damage though.
I'm not sure if it's just the weather, but the PSU fan stays on about a minute after the machine is shut down, which it never does. Maybe it's the PSU itself earthing against the case, causing it to heat up, I'm not sure. The PSU is warm to touch on the outside, which is fairly usual.
I do have a spare 350W PSU floating about, but I'm not sure if it will be able to replace my 550W. I also don't want to have to find and remove all of the PSU's wiring, which is lost throughout the case at this stage after various lighting projects. I'm sure these aren't the problem though, I've removed them all to test it. I'm absolutley stumped at this, it was working flawlessly all day at TF2, and I shut it down and started it as normal. Hopefully I've missed something obvious, or someone smarter than me will see what I've cocked up on
Cheers
EDIT: I've solved the non-booting issue. I left my memory stick in which was formatted as NTFS when I updated Windows 7 from it, and it was trying and failing to find a boot disk on it :doh: (like I said, something REALLY obvious...) So at least I can short out the fans and play away for the mean time
And the PSU is acting normal, and not causing the boot failure. So it probably is just a negative earthing somewhere, or else the fan controller is toast...
EDIT2: After a bit of poking around, I found one of my fan controller wiring jobbies (held together with insulating tape
has come loose...Problem solved, and I feel like a tool after trying to find a problem for the last 2 hours
EDIT3: Actually, it's still earthing somewhere. The controller still works without a negative molex attached. But I don't care, it's working again