HX1000 vs Non-corsair 1200W and MB IOH temp differences

[shoes]

Hi,

 

I came across a peculiar problem today that I was hoping maybe a PSU expert could provide some insight on.

 

I have 2 "identically" configured computers running side by side but there are some small differences between the two. Here are the identical specs:

 

Case: Cooler Master HAF X

Cooling: Asetek 240mm liquid cooling (OEM)

Motherboard: MSI X58A-GD65

CPU: Intel Core i7-980X

GFX: 3x NVIDIA GeForce GTX 480

Audio: Creative SB Fatality champion whatever (in the PCI-e x1 slot at the top of the expansion stack on the GD65 board)

 

DIfferent specs:

 

Machine 1

PSU: Non Corsair but reputable brand TR2 1200W

Memory: Non corsair but reputable brand 6GB @ 1600

HD: 2x 1TB 7200RPM in RAID 0

 

Machine 2

PSU: Corsair HX1000

Memory: Non corsair but (different) reputable brand 6GB @ 1600

HD: 2x non corsair 128GB SSD in RAID 0, 2x 1TB 7200RPM in RAID 0

 

When running CPUID's HWMonitor software, the value reported under TMPIN1 for each machine is in the 70C's for Machine 1, and in the 90C's for Machine 2.

 

Some background: The TMPIN1 value is reporting the IOH chip temp. I confirmed this using a k-type thermocouple connected to a digital thermometer. The MSI X58A-GD65 board being used runs notoriously hot, but not this hot.

 

I can accept temperatures in the 70's, but certainly not in the 90's and 100's. (temps in celsius)

 

I have run some preliminary tests using different power supplies in the system exhibiting the higher temperatures.

Test1: Swapped in a 1100W old non-corsair PSU i had laying around. Temperatures were markedly lower, nearly 20C.

Test2: Swapped in a Corsair TX950. At first temperatures were 30C lower, but they rise back to the 90s.

 

I have not yet ruled out that this particular motherboard just has a hot chip or bad thermal connection on the heatsink, but I am not ready to make that diagnosis yet given the discrepancies when using different power supplies.

 

So I guess my question is:

Why/how would using different power supplies in a system effect the temperature on the chipset?

 

Is there any precedent for this? I have been building many computers for many years, most of them very high-end like this, and I have never encountered this particular problem before.

 

Thanks for any help.

 

shoes

[va1erian]

I am no electronics specialist. But I reason if there was any problem with the Corsair PSU, then the computer would not be able to perform well. Did you try to run any programs which load the CPU and/or the GPU 100% ?

[RAM GUY]

The most likely thing that I can think of would be that the other PSU has a faster spinning fan that is pulling more air out of the case. The fans on our PSUs will not do a good job of keeping the case cool, they will spin very slowly until you get to about 50% loading. This is to keep noise levels down. You may just need to add an extra cooling fan to keep the chipset temps a little lower. We can definitely try replacing the unit if you would like to try that, but from what you describe it may or may not improve temperatures.