Strange hardware troubles with new computer (possibly shorting?)

So, several months ago I built up a new gaming rig -- Core i7 920, GTX 285, ASUS P6T6 Workstation Motherboard, Corsair 750W Power supply, and 6 GB DDR3 ram. It is also watercooled with dual loops.

Within the past week or two, it will (seemingly) randomly turn off, no warning, just off. Several seconds later it will turn back on -- sometimes. Sometimes it will try to turn back on and it will just flicker the power several times until it turns on, but does not boot and has my CPU-DIAG led light on and blinking.

I have a couple of ideas as to the problem, but could use some feedback.

1) Power supply is dieing

2) Motherboard is bad

3) Ram is bad (just ran Memtest86+, no errors, but it did shut off during it...)

4) My two pumps (18W*2) are running off fan headers on the motherboard, maybe they are stressing it?

5) CPU is dieing (not overclocked, completely stock setup except watercooled)


I really appreciate any advice or ideas that you have. This is a pretty depressing problem with my new baby... :(

Quote:

---

1) Power supply is dieing

2) Motherboard is bad

3) Ram is bad (just ran Memtest86+, no errors, but it did shut off during it...)

4) My two pumps (18W*2) are running off fan headers on the motherboard, maybe they are stressing it?

5) CPU is dieing (not overclocked, completely stock setup except watercooled)

---

1) It is possible that your PSU may of died, however if it's brand new then its rare but always possible. Test the PSU on another system.
2) I think it may be a bad mother board, I have had a ASUS Striker formula board that was bad fresh out of the box, it needed replacing. Test your mobo on another system to be sure.
3)I would re-test the RAM again and if shutdown ouccurs during it then test it on another system.
4)The two pumps should be fine with that motherboard and power supply, but just incase run them stright from the PSU.
5) test the CPU on another system

Would be really handy if you had a test rig :P


Dweldom

Haha, yes, it would be vary handy. Maybe I can convince my buddy to let me use his, but since we are both very entwined in our watercooling setups, it makes everything much, much more difficult...

EDIT: I am going to go out on a limb and blame my issues on the cheap, chinese cold cathode UV lights. The problem certainly worsened when the computer was left on for a period of time, but since I didnt have temp sensors on the voltage step-up boxes (12V to 650V), I didn't even think to look at them. The problem definately looked like something was shorting (from my EE training), so this would kind of make sense -- the components would heat up, expand (or overheat) and short inside the box, causing my power supply to shut down automatically.

Now, having said that, I am going to go home and find my computer in a coma as usual...but I hope for the best.