For those of you that were keeping up with the ongoing Koolance saga that Dug, myself, and a few others were having, mine has ended well. 
After first blowing a liquid cooled power supply model, and then the replacement upgraded non-liquid cooled power supply model developing a leak, I was still convinced that the 3 pass radiator design mounted in a unit hidden in the case bottom worked well. I like the concept of watercooling, but did not want the hassle or looks of tupperware containers, radiators with 92/120mm fans, etc.. I wanted something hidden, quiet, and with good waterblock mounts, hence my interest in the Koolance.
It turned out that the liquid mix used in some of the Koolance units was contaminated with something that caused a reaction which resulted in a very high internal pressure build up. I sent an e-mail off to a person that modded his Koolance, and was given the mixture he used to cool his. ( 120 ounces of distilled water, 6 ounces of anti-freeze, and 2 ounces of Redline Water Wetter. )
The first thing that I did was to replace the 3 cheap rear exhaust fans with 3 50mm Deltas. They are very quiet, but push more CFM than the stock ones. The fans only operate at 40% speed at the first thermal kick on level, (which I have not exceeded yet after 4 complete days of testing) and 100% speed after that. I also replaced some of the hose clamps with easy to remove ones. After flushing the unit with distilled water a couple of times, I used the above cooling mixture to fill the Koolance.
For this test, I am using an ABIT KT7R which had the voltage mod done, and a 750 Duron. I cranked it up to 2.0V, and set the Duron at 1 GHz. My DigitalDoc 5 is installed in another system, and I have had some painful dental work done, so I am just going by the Koolance waterblock probe LED temp display for now. I will attach an external probe against the core for more accurate readings when I move on to re-testing a 1200 T-Bird in a few days. The important thing is that after 4 days of testing, there have been no lockups, overheating problems, or leaks.
Each day of testing consisted of complete 24 hour days. Day 1 was the Prime95 Torture test, Day 2 looping the 3DMark2001 benchmark, Day 3 looping the 3DMark2000 benchmark, and Day 4 looping the Q3A demo2 in Q3Bench, at 4 resolutions.
The Koolance waterblock probe never exceeded 37 degrees Cel. throughout testing, and averages 34 Cel. the fans are not running at 40% speed all of the time, and have been kicking on/off when needed. The inner case bottom and hoses are just barely warm to the touch. I am very satisfied with the performance of the Koolance, and had similar good results overclocking a T-Bird to 1400MHz before the leak occured during my other test awhile back.
I had an Acrylic case made with custom holes cut in the bottom to install the Koolance cooling unit. My next project will be to install my primary components in it, and see if I still am able to run at my below specs. Even if I have to underclock to the 1400MHz that I know the Koolance will do, it will be well worth the noise reduction. The Koolance is so quiet, all I hear is my hard drives, and the Quiet fan in the power supply.
Sorry for the long post, I just knew that there were quite a few people interested in the last Koolance threads.
After first blowing a liquid cooled power supply model, and then the replacement upgraded non-liquid cooled power supply model developing a leak, I was still convinced that the 3 pass radiator design mounted in a unit hidden in the case bottom worked well. I like the concept of watercooling, but did not want the hassle or looks of tupperware containers, radiators with 92/120mm fans, etc.. I wanted something hidden, quiet, and with good waterblock mounts, hence my interest in the Koolance.
It turned out that the liquid mix used in some of the Koolance units was contaminated with something that caused a reaction which resulted in a very high internal pressure build up. I sent an e-mail off to a person that modded his Koolance, and was given the mixture he used to cool his. ( 120 ounces of distilled water, 6 ounces of anti-freeze, and 2 ounces of Redline Water Wetter. )
The first thing that I did was to replace the 3 cheap rear exhaust fans with 3 50mm Deltas. They are very quiet, but push more CFM than the stock ones. The fans only operate at 40% speed at the first thermal kick on level, (which I have not exceeded yet after 4 complete days of testing) and 100% speed after that. I also replaced some of the hose clamps with easy to remove ones. After flushing the unit with distilled water a couple of times, I used the above cooling mixture to fill the Koolance.
For this test, I am using an ABIT KT7R which had the voltage mod done, and a 750 Duron. I cranked it up to 2.0V, and set the Duron at 1 GHz. My DigitalDoc 5 is installed in another system, and I have had some painful dental work done, so I am just going by the Koolance waterblock probe LED temp display for now. I will attach an external probe against the core for more accurate readings when I move on to re-testing a 1200 T-Bird in a few days. The important thing is that after 4 days of testing, there have been no lockups, overheating problems, or leaks.
Each day of testing consisted of complete 24 hour days. Day 1 was the Prime95 Torture test, Day 2 looping the 3DMark2001 benchmark, Day 3 looping the 3DMark2000 benchmark, and Day 4 looping the Q3A demo2 in Q3Bench, at 4 resolutions.
The Koolance waterblock probe never exceeded 37 degrees Cel. throughout testing, and averages 34 Cel. the fans are not running at 40% speed all of the time, and have been kicking on/off when needed. The inner case bottom and hoses are just barely warm to the touch. I am very satisfied with the performance of the Koolance, and had similar good results overclocking a T-Bird to 1400MHz before the leak occured during my other test awhile back.
I had an Acrylic case made with custom holes cut in the bottom to install the Koolance cooling unit. My next project will be to install my primary components in it, and see if I still am able to run at my below specs. Even if I have to underclock to the 1400MHz that I know the Koolance will do, it will be well worth the noise reduction. The Koolance is so quiet, all I hear is my hard drives, and the Quiet fan in the power supply.
Sorry for the long post, I just knew that there were quite a few people interested in the last Koolance threads.