Crash Override

Here is an odd issue for you folks here


I am putting together a system for a customer of ours and our owner prefers to use Antec Cases with the Power Supply already in them.

Well I personally have had nothing but trouble with Antec and wanted to avoid using them again.

Well we ended up buying th 1040 SOHO case with the 400W Smart Power inside.

Build the system, check all of the connections, power it up, and poof smoke.

I physically disconnect the power supply from the wall by pulling the cord out of the power supply less than 3 seconds after this happened.

I managed to save every single component inside of the computer other than the floppy drive upon which the power lead had melted onto the drive. This floppy drive was from one of my systems. It was only intended to go in temporarily to copy over the raid driver files for XP setup, so yes it was a known good floppy drive. It seems something was wrong with that particular line as the insulation on it had literally burned off.

So we replace the power supply with a 400W Allied (AMD recommended) PSU and power the system back up and the temps are much higher than they should be.

Using the F11 bios the temps are currently at 52*C idle. This is using the AMD stock heatsink for the 2500+ plus 2 Antec fans in the back spinning at 1300RPM's each and one vantec stealth in the front spinning at 2200RPM's. I manually set the CPU voltage down to 1.65 as it kept attempting to go up to 1.7 even though its labeled at 1.65. The room its in right now is air conditioned so I know it cant be correct. Not to mention this isnt the first 2500+ system that I have built and none of them have ever ran this hot so my theory is that maybe the thermal diode on the motherboard that picks up the CPU temperature is damaged since the System Temperature reads at almost 30*C


In case anyone is wondering here is the full specs on the system so you can see what if anything could cause it.


AMD Athlon XP 2500+ (Week Code 0327)
Gigabyte GA-7N400 PRO F11 Bios
1024 GB (512x2) Corsair XMS 3200 Dual Channel Memory Kit (Memory Timings: 6,3,3,2)
240GB (120x2) Western Digital SE HD's running in RAID 0
ATi Radeon 9500 PRO
Lite-On 16x DVD-ROM
Lite-On 52x32x52 CDRW
Antec 1040 SOHO Case
Allied 400W Dual Fan PSU
2x Antec 1300RPM 80mm Case Fans
1x Vantec Stealth 2200RPM 80mm Case Fan
Soundstorm Onboard Audio


I cant think of anything that I missed in my description above

If anyone needs more details let me know.

Crash Override

Customer is picking up his system after the weekend so giving this bumpage so I can get as much help as possible between now and then

RIV@NVX

1) Are you sure that the cooler is really powerful enough?
2) Are you sure that nothing is melted on the motherboard?

__________________
-- Vedran

PoopyTheJ

Check the coverage of the Thermal Grease? I've had high temps caused by poor application of grease before, also improper seating of the heatsink. I know you do this for a living just my random thoughts. Also, are you using the most recent BIOS for the Gigabyte mobo? I know in the past motherboards have had temp issues based on the BIOS not correctly reading the thermal diode, can't say I've heard of it with that board but never know.
My thoughts so far.

__________________

Crash Override

I'll use this post to answer your post and the post from gaborn415


1)Im sure that the cooler should be adequate, I have built many systems with the 2500+ CPU and have always been happy with the cooling it provides as were building systems mainly for people who arent all that tech saavy so they wont really be overclocking at all

2)On that I am not completely certain. Everything on the motherboard appeared to be in good working order, and before the power supply went kablooey the temps were a tiny bit higher than I would have liked. After putting in the new PSU the temps have risen almost 30*C according to the bios (only 20*C at first but the new bios revision kicked it up to 62*C so it most likely was adjusting the diode).

I pulled the heatsink off of the CPU and took off the stock paste and replaced it with the TIM that came with our SLK-800 and the temps went down 10*C to 52*C at idle. The system temperature throughout all of this has stayed steady at 33*C and lower so you can see why I am confused considering that the System Temp diode is usually close to the CPU socket and would be picking up that much heat.

tigerarmyAFI

Heres a thought: Open up the side of the case, and place a desk fan blowing on full power directly into the open case. This should drop the system and cpu temps dramatically (at least 5-10c), i know it does because ive done it many times when overclocking. watch your temps.. if they dont drop, you know the diodes are lying to you, then it doesnt really matter what it says if you know for a fact the system is cooled effectively for everyday use, what the hell does a diode matter anyways?

__________________
AMD AthlonXP 1800+ ThoroughBred B JIUHB @ 2137MHz
Geforce4 Ti 4200 128Mb @ 310/600 (ASUS V8420)
ABIT KX-7 333 v1.1
512Mb Micron PC2700 DDR
Creative Live! 5.1
Aircooled (45c Full load)

Crash Override

I know that would make the temperatures drop as when the room cooled at night the temps went down according to the thermal sensor but as the power supply could have damaged components Im wondering if anyone knows if the diode could be damaged without any other component receiving damage at all.

Also I dont like for a customer to see a big difference in temperatures like that as this system is a gaming rig for his kids and while he might not be tech saavy his teenages may be.

Anyone willing to drop almost 2000 on a system from us deserves all of the best we can give them

I just hope their monitor gets here on tuesday. They decided to get a 19" Sony LCD monitor and have us keep the 19" CRT that we had already ordered for them so they pay us 455 dollars for it and we call it even