Unfortunately you were only speculating and then shotgunning. A 3.5
digit multimeter would have accomplished more than that power supply
tester - and without making all those disconnections and swapping. For
example, if +5VSB had too much ripple voltage, then your strange
problems are identified. Excessive ripple could cause power supply
controller to triggers motherboard's reset line. Meanwhile that power
supply tester may not detect same. Same applies to other critical
voltages that may only measure defectively when under full load - such
as when connected to computer. Just more reasons why a multimeter is a
better investment; why a Go-Nogo tester reports so little useful
information.
Best test of power supply is under load. Test supply when connected
to computer while many computer peripherals are being accessed
simultaneously. Once meter verifies the power supply, then move on to
other suspects. (Of course, you are quickly running out of suspects
having replaced so many components). If meter finds a problem with
the Antec, then use the Thermaltake. Confirm that same erroneous
numbers no longer occur. Numbers from those readings (not just 'good
or bad') also provide useful information.
Do not worry about a system damaging that power supply. Nothing
connected to a power supply can damage a minimally acceptable supply.
If system damages the power supply, then you did not want that supply
anyway. Power supply specs may even say how large a shorting wire
should be to short all power supply outputs together - and still power
supply must not be damaged.
Nothing from wall receptacle should cause any problems as long as the
safety ground is properly connected - again something that the
multimeter can confirm. Just another thing that power supply tester
cannot confirm. The Belkin does nothing useful here. However, to
eliminate other reasons for problems, computer and all peripherals
should share a common wall receptacle.
Other useful information comes from executing comprehensive hardware
diagnostics for each peripheral, and information from Windows system
(event) logs and in device manager.
wrote:
> Merry Christmas, and now happy new year as well.
>
> Sadly, the problems continue, even after a new motherboard... and now I
> really have no idea what to do. I just drafted this post, and put it on
> several forums detailing the problems from the begining, I'll just
> copy/paste it here, and maybe somebody here has some thoughts.
>
> --
>
> Alright, this is fairly long, but I'll go step by step with what's
> occured, and hopefully somebody can give some suggestions or help in
> some way.
>
> Original system setup (core components, no PCI cards/drives listed):
>
> ABIT IC7-G MAXII Socket 478 Intel 875P ATX Intel Motherboard - Retail
> Intel Pentium 4 3.0C Northwood 800MHz FSB Socket 478 Processor Model
> BX80532PG3000D - Retail
> Antec NEOPOWER 480 ATX 480W Power Supply - Retail
> CORSAIR XMS 1GB (2 x 512MB) 184-Pin DDR SDRAM DDR 400 (PC 3200)
> Unbuffered Dual Channel Kit System Memory Model TWINX1024-3200XLPRO -
> Retail
>
> So one night, I come home, try to boot the machine, and nothing at all
> posts, nor does the motherboard beep. First I thought maybe it was the
> power supply, and my friend gave me a Thermaltake 430w supply. I tried
> swapping them, still no post, still no beeps.
>
> The only thing that seemed odd was that the RAM, which has LED lights,
> one stick would turn on at boot, the other wouldn't, but after no post,
> a few seconds later, the LED lights on the one stick that did light
> would also go off.
>
> Tried the RAM, one chip at a time, in different DIMMs to see if maybe
> the RAM was bad, nothing. Tried no RAM at all to see if the board would
> at least beep, nothing. [with only the CPU and videocard in the board]
>
> So then my friends and I decided it had to be the motherboard.
>
> I bought this as a replacement board, Asus P4P800 SE 865PE 800MHZ Ddr
> Ram 4 Gb.
>
> Luckily, I forgot the thermal paste my co-worker brought to work for me
> at... well, work, and as such, I had to go to Circuit City to buy some.
> While there, I also spotted an Antec Power Supply Tester, and it was
> $9.99, so I bought it as well for future usage.
>
> Now, given that we thought the board was dead, and the original 480W
> supply was fine, I was going to actually use that with the new Asus
> board. For whatever reason, I guess curisoity, I tested it first with
> the power supply tester. The green light didn't go on, and I started to
> smell pungent burning smells. So, apparently that supply is dead. I
> then tested the Thermaltake 430w supply, and the green light on the
> power supply tester went on, and no burning smells.
>
> My theory at this point was that the Antec 480w supply died and fried
> the motherboard with it.
>
> We installed the Asus board, with the same CPU, RAM, and the
> Thermaltake 430w supply. Everything worked fine for about 24 hours.
> Just now, I was ripping some music to WMA lossless format, and I went
> away from the monitor to take a phonecall. I came back, no image on the
> screen. I thought maybe it went to sleep, nope. The RAM LEDs were
> stuck, and not moving.
>
> Tried resetting the computer, nothing. Power off and on... nothing. No
> post, no beeping, same RAM thing as before where only one stick would
> light up, and briefly, and with no post the lights would go off again.
>
> What the hell is going on here? I've yet to try the new Thermaltake
> 430w supply in the power supply tester again after this event for fear
> of it blowing up or something.
>
> Was this RAM somehow damaged, and is now killing boards and power
> supplies?
>
> As for maybe an outlet problem, one, it's a Manhattan apartment, and
> the whole building's power was apparently re-wired a few years ago, so
> the wiring should be fine, nor have I ever had a problem with it
> before. Secondly, it's in a Belkin heavy duty surge protector, and
> these past 24 hours I was using the machine with only a keyboard and
> mouse attached, and hadn't gotten around to replugging in all my
> external drives (five), which would usually be on and taking power as
> well without a problem.
>
> The only thing I can think of that's weird is that while the RAM is
> 3200 dual channel, at one point I noticed the board detected it as 233
> or 267MHz single channel. I was going to trouble shoot this later. If
> this is a sign that the RAM is indeed bad, how would bad RAM work for
> 24 hours, kill components, and be indentified wrong as opposed to the
> board just not booting and giving an error beep?
>
> Again, any advice or comments would be very helpful. I don't know what
> to do at this point, and obviously am afraid of killing more parts.
> Thanks.