PSU, or what? (Page 1 of 2)

I've had this happen before, and the problem was that the power button had become disconnected from the case.  You should probably check that out before spending money.

Yes.

Most motherboards now have power up and reset buttons on itself, so you probably can look for those and bypass the case switch to check.

Well, as expected, everything is still connected (seemingly - unless there's a loose connection at the button itself, which is buried behind plastic). Replugged everything, made sure I had the pins correct, still nothing.

Re Errr: No such buttons to be found.

Next troubleshooting steps?

I'm looking into that at the moment. Any ideas why my computer spontaneously shut itself off, though? Don't think any critical updates were waiting to be installed.

We were also considering picking up a multimeter to make sure the button closes a circuit, but I doubt the switch is the problem, especially given:

The history of problems
Abruptly shutting itself off last night

I tried the reset button thing earlier this evening, it didn't work. Still unsure how a spontaneously broken switch would shut off my PC in the dead of night.

Wow, I really wouldn't have bothered to respond to your post if it wasn't because this is the tech support forum, and people could suffer real consequences because of your misinformation (not that many will listen to you, but a few might).

For starter, sensor-triggered shutdowns occur only during overheating, they do not prevent a power-up. So even if your argument is valid, it isn't relevant. Secondly, mobo sensor triggered shutdowns don't manifest in BSOD; they simply power the system down without warning (or with a mobo beep). Even the described behaviors before the Vista installation and replacement of parts didn't match this. And depending on the CPU, a dead fan on a heatsink that isn't designed to passively cool the CPU will certainly result in temperatures of over 100C rapidly (for reference's sake, the retail coolers that come with the recent generation CPUs typically only cool the CPU to 60C actively in reasonable environments - barely keeping them in safety range when under heavy load), and cause permanent damage. Your idea of being able to run a CPU for a while with no active cooling is so much a thing of the distant past that it's laughable.

As to isolating components, how do you propose him booting anything up if he can't send power to the mobo? He described his problem clearly, and the first thing everyone advised him to do is to get through it to do any sort of further testing. Perhaps you should pay more attention to everything, not just the fact that it's not a laptop. Or, perhaps you should pay less attention to this forum as a whole and just leave it alone, since all you seem to want to do is attract attention to yourself, instead of actually helping others. Go back to ESAD where you belong and troll there.

Viserys: I've had an incident in the past where the power button connection failure resulted in shutdown. I've also built a system for a friend some years ago where the power-up problem was solely due to the cheap case and its badly designed power button. It doesn't take much "effort" to trigger a shutdown - a bit of gravity on a loose wire or a gust of wind all can do that. I'd worry about the power button for now, and if you can verify that it isn't the power button, I'd then look into swapping the PSU.

Mikotren, even if your advice was valid, I deleted your post. This is tech support - there is zero fucking room for flaming or trolling. Think carefully before you respond, because if you attempt to troll or flame anyone else in this thread, I will simply ban you.

Funny that the first advice most of these threads received is always "check your power button and make sure it's not faulty"? Upon verifying that, the following advices have been "verify PSU isn't faulty"? What are you trying to prove here? That we're suggesting to him the correct course of diagnostic actions?

So he underclocked it to nearly half of the stock speed on a 65W C2D and still booted to 52C core? I'd say that's more than enough proof that a stock CPU (especially an AMD one or previous generation C2D) will fry itself. Thanks for making my point instead of yours, genius. Also, since that video is now posted (and irresponsibly positioned as a "proof"), I have to advise anyone who values his/her money NOT to try what was shown in the video without researching into his/her CPU's thermo requirement extensively. Even then, unless you have an external thermometer, I'd be very reluctant to actually trying it out. It's like one of those bomb making videos - consider yourself warned.

Viserys: If you can try to plug the PSU to a barebone system with definitely reliable hardware, then I'd try and do that. As I understand correctly, you replaced your old CPU/mobo/RAM because you suspected that they were the source of your problem, and I'm not sure it'd really help isolate your problem now if you used them. If you had a friend who's willing to give it a try, you can always hook up the PSU to his system.

Update:

Hooked up the PSU and power switch to my old system (mobo, CPU, 2 sticks of RAM) via the 24 pin connector, an aux 4 pin, and the 2 pin switch. As soon as I plugged it in and flipped the switch on the PSU, I knew something was up. The back fan's blue LEDs lit up, and the CPU fan came to life. Hit the front power button and everything functioned as expected.

So, next step?

Things I have yet to try:

Trying just one of my two RAM sticks in the new system
Praying fervently
Crying

Edit: Crast just suggested plugging my old Pentium D into the new mobo to try and isolate the problem further, but we're not sure if that's safe. Currently, everything is removed except for CPU, RAM, power. That includes sound, graphics, optical drives, hard disk.

I'd consider the current configuration pretty bare bones?