SK8V Hi Temp(127C) With Zalman Cooler

[Darph_Bobo]

I recently purchased a Asus SK8V motherboard, Opteron 144 and the Zalman WB2 Waterblock. According to Zalmans website the waterblock supports the socket 754 cpu, and after reading info on amd’s website found out that both the socket 754 and socket 940 (SK8V) both use the same mounting bracket. So armed with this knowledge I proceeded to assemble the board and cpu as per the directions that came with the waterblock.

I then proceeded to power on the system, which appeared to post normally until I got the CPU OVERTEMP error, I proceeded to enter the BIOS and checked the temperature and it reported a temperature of 127C, oddly this did not automatically shutdown the system, after seeing the temp I quickly shut down the system. Thinking that this was a BIOS error I proceeded to clear the CMOS and rebooted, Same Problem. Next I turned on the system and let it run for a few minutes, as I am using watercooling if the temperature was indeed 127C the water would start to boil, which it didn’t. So I continued to install windows which went fine.

Once windows had been installed I quickly installed Asus Probe (Latest version) which reported the temperature as 127C. Frustrated I decided to take the system apart and try installing everything again just to make sure that the waterblock was not improperly installed. This is where I noticed the problem.

I had the AMD bracket sitting next to the Zalman bracket, and I noticed that the cutout in the AMD was larger (both LxW) then the Zalman, so I took a look at the back of the board and noticed that there was a small resister that had been touching the back plate.

I then took my dremel tool and removed a small area from the back plate so that the resistor would no longer be touching it. I then reassembled everything and hit the power button and went into the BIOS where it reported 26C and everything was normal.



I am unsure if this problem is happening on all AMD 940pin boards, or on just a select few of the SK8V as this is my first AMD 64 system.

*Edit* I Have talked with Zalamn and they have changed the backplate on the newer coolers

[cromozone]

Very interesting information, thanks! I didn't have this problem, I didn't change the bracket at all from the one that Asus supplied. Hopefully this was a one off issue, and won't reoccur with other users.

[Darph_Bobo]

Unfortunatly I have heard from 2 others with the same issue

[cromozone]

Hmmm, worrying indeed. While it doesn't seem to stop the system from functioning correctly, it will worry an awful lot of users.

[Ghostrider]

Originally posted by Darph_Bobo

I had the AMD bracket sitting next to the Zalman bracket, and I noticed that the cutout in the AMD was larger (both LxW) then the Zalman, so I took a look at the back of the board and noticed that there was a small resister that had been touching the back plate.

I then took my dremel tool and removed a small area from the back plate so that the resistor would no longer be touching it. I then reassembled everything and hit the power button and went into the BIOS where it reported 26C and everything was normal.



I am unsure if this problem is happening on all AMD 940pin boards, or on just a select few of the SK8V as this is my first AMD 64 system.

*Edit* I Have talked with Zalamn and they have changed the backplate on the newer coolers

If the backplate is properly installed (with the non-conducting sticky film on it) there should be no chance to bend the board or to shortcut it.
BUT - if they constructed the backplate wrong it's different.

BTW: If I remember right my K8W has NO resistors or anything else on the backside of the mainboard below the CPUs. It's completely blank there.

*searching for a picture online*
Like this: http://user.chollian.net/~xriona/2cp...5/1024/l38.JPG

Ooooh yeah... I like Server-Mainboards.

EDIT: Does AMD allow to place ANYTHING in this region below the CPU?

[Darph_Bobo]

This is the resistor that is causing the problem

Problem Resistor

[Ghostrider]

I never thought that a clear backside of a mainboard could be a feature
-------
I still don't know if the reference design of these mainboards allow parts to be placed there. Does someone know? Whitepapers from AMD should reveal it.

[xgman]

I ahd this happen once and when I reset the computer it went back to normal. maybe it did touch that for an instant.

[kaplansa]

Originally posted by cromozone
Very interesting information, thanks! I didn't have this problem, I didn't change the bracket at all from the one that Asus supplied. Hopefully this was a one off issue, and won't reoccur with other users.

This could be a problem long-term. I did the same thing :eek: The issue is that the stock backplate has risers that come through the mobo. Even thought the Zalman will screw into these risers no problem, you're not making a tight connection to the CPU.

If you look at the backplate that comes with Zalman, it's got no built-in risers, but you have to use their own screw-on risers. In fact, these risers are lower than the risers on the stock backplate. This is possibly what's causing your problem.

Completely remove the stock backplate and put a piece of waxpaper over the sticky back to preserve it's stickyness if you need it again. Then, take the Zalman backplate, and install it appropriately. Note that the Zalman may have come with little paper o-rings - my instructions didn't say what they were for, but I figured it out. Put a small paper o-ring on each hold, between the new backplate and the mobo. Then take a larger paper o-ring and place this on top of the mobo hole, betweem the mobo and the screw-on risers. These o-rings help insulate the metal from the backplate and the risers from shorting out the mobo. The stock backplate doesn't have this problem since the risers are physically attached to the backplate.

Anyhoo, just my $0.02.