Jump to content

Watercooling results


red1776

Recommended Posts

I recently replaced my 'Big air' Silver Arrow with a dual rad setup and have been trying to put a metric on how well the cooling is working.

 

I placed sensors on the tubing an equal distance from the corresponding exit fitting using foil tape ....

 

post-73619-0-01553100-1330068380_thumb.jpg

 

I took reading this way, and then covered the sensors with hot melt (which i have learned has the same thermal properties as tubing +/- 5%) to protect from outside heat sources and ran the temps again

 

Both tests Delta temp were identical. The sensors were used on a known temperature to start and were spot on.

 

The results were taken after 30 minutes idle and 30 minutes into 100% load test

 

@ 22 Ambient room

 

4.9Ghz @1.52v

 

Delta T Radiator out / CPU block out 3c

 

Delta T Idle to 100% load stress test 18c

 

***if anyone else plays with this stuff I would like to hear your results if you care to post them :)

 

 

 

 

 

Share this post


Link to post
Share on other sites

Red go buy a bitspower g1/4 screw in thermister and report back with some other temps, not sure that I can trust the accuracy of readings taken through the tubing.

 

Either way though temps are looking pretty damn sweet, how is the noise? Cougers doing well?

Share this post


Link to post
Share on other sites

Red go buy a bitspower g1/4 screw in thermister and report back with some other temps, not sure that I can trust the accuracy of readings taken through the tubing.

 

Either way though temps are looking pretty damn sweet, how is the noise? Cougers doing well?

 

 

 

Yeah I am going to get a top notch thermo after this, I am really enjoying it. I was guided in this by a WC'ing Guru on this test, so I trust the rad out/CPU block out test to be reasonably accurate. It's also the reason that i am going after Delta temps. The old fashioned method backs it up as well. You are really hard pressed to feel a difference between the two.

 

 

Would you throw me a link to the one you recommend? ***** I found it*** :)http://www.sidewindercomputers.com/big1tesestfi.html

 

The Cougars were a great choice, very quiet.

 

Choosing the rad thickness was a guesstimate, and I apparently got lucky. I think 45mm was the way to go :thumbsup:

Share this post


Link to post
Share on other sites

Nice temptatures & setup.

But don't you think that 1.52V is a bit high? :dunno:

Look at ambient and delta T as long as his vrm and mosfet is still being well cooled (look at the spot fan in his pic) he should have no issues at all. Also it is bulldozer not sandybridge.

 

That is the one Red also comes in black. Also check this out for a cool addition to the loop.

Share this post


Link to post
Share on other sites

@ d6, Like stoner said. This is BD not SB. The max Vcore for normal cooling methods is 1.55v. I am not coming close to the max thermal for the CPU.

 

@ Stoner, The NB and VRM are running in the high 30's , but I ordered a chipset/VRM block anyway for it.

 

If I can find the sensor/readout set you linked to, I have a spot for it :) 

 

 

Share this post


Link to post
Share on other sites

I have that liquid sensor and the specs are 10k @ 25c. When I purchased mine I searched all over the web for those specs but ended up ordering it with my fingers crossed hoping it would work. Other sensors are 50k @ 25c so that got me thinking that the resistance difference is probably the reason that some software temp monitors report temps off by the 10 or so degrees sometimes.

Share this post


Link to post
Share on other sites

You get 18 C delta T under full load? You need more cooling!!! With my 5970 @ 1.3 volts, 1000/1200, and my Phenom II at 1.55 volts, 3.4 GHz (tri->quad unlocked), I only get a 4 C rise over ambient. :teehee:

 

 

The biggest delta T within my loop was less than a single degree when I had my two 4870X2s (at 800/900) and the same Phenom II. It looks like your flow might be a bit low.

Share this post


Link to post
Share on other sites

I added a 2x120 rad in my loop and seen my flow indicator spin slightly slower, but my occt test for 3.08 ghz dropped ~6c despite using the two 80mm fans on the top of my case (60 cfm). I wonder what two 120mm fans would give me?

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...