Pro/Forums - View Single Post - Outrage: cooling 10 comps with an Uber pump and a bong

jaydee · 12-31-2002, 04:39 PM

With that said it starts to not make since how it will be cheaper? You will still need somekind of active cooling for that external heat exchanger as just air going through it is not going to work in warmer climate areas like Arizona. Or any other place that has warm summers. So it will still need to be airconditioned somehow. this maybe take a little less power than A/C'ing the computer room but being A/C is most likely already setup in the building for the employees of the place so ducting shouldn't be all that hard. Lot easier than running pipes and setting up pumps. I am not sure how this would be cheaper. The initial installation of water cooling something like this is going to cost a fortune, not just for the parts but the manual labor. Putting on a HSF takes seconds, putting water blocks on computers and switches will not. So if it is cheaper to run watercooling that difference is going to take a very long time to offset the initial installation. And then this is all on speculation that the roof of this server room is also the top of the building. What if it is a multi level building and the server room is on the bottom.... I think it would be cheaper to push cooled air down a duct than cool water to the computer then back up to the exchanger. Then you have all the extra maintenance in a system like this. You would have to either have trained employee's checking the system regularly or have the installation company do it. That will not be cheap and it is a requirment. I don't think anyone in their right mind would let a system like this go unchecked regularly. That is some seriuosly expensive equipment as risk. One quick leak and you just lost all the money you may have gained by going with liquid cooling. You would also have to have overheat and leak detection systems on everything, which will not be cheap. And you would want a flow sensor of somekind. And all this has to work flawlessly 24/7.

I am sceptical, but not saying it cannot be done. There are just to many variables. I think each system would have to be designed for each application. That alone could be $100,000's in engineering costs alone as it would have to be drawn up and approved by engineers. It all has to fall under building codes to boot. Non of this is needed with air cooling.

12-31-2002, 04:39 PM	#48
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	With that said it starts to not make since how it will be cheaper? You will still need somekind of active cooling for that external heat exchanger as just air going through it is not going to work in warmer climate areas like Arizona. Or any other place that has warm summers. So it will still need to be airconditioned somehow. this maybe take a little less power than A/C'ing the computer room but being A/C is most likely already setup in the building for the employees of the place so ducting shouldn't be all that hard. Lot easier than running pipes and setting up pumps. I am not sure how this would be cheaper. The initial installation of water cooling something like this is going to cost a fortune, not just for the parts but the manual labor. Putting on a HSF takes seconds, putting water blocks on computers and switches will not. So if it is cheaper to run watercooling that difference is going to take a very long time to offset the initial installation. And then this is all on speculation that the roof of this server room is also the top of the building. What if it is a multi level building and the server room is on the bottom.... I think it would be cheaper to push cooled air down a duct than cool water to the computer then back up to the exchanger. Then you have all the extra maintenance in a system like this. You would have to either have trained employee's checking the system regularly or have the installation company do it. That will not be cheap and it is a requirment. I don't think anyone in their right mind would let a system like this go unchecked regularly. That is some seriuosly expensive equipment as risk. One quick leak and you just lost all the money you may have gained by going with liquid cooling. You would also have to have overheat and leak detection systems on everything, which will not be cheap. And you would want a flow sensor of somekind. And all this has to work flawlessly 24/7. I am sceptical, but not saying it cannot be done. There are just to many variables. I think each system would have to be designed for each application. That alone could be $100,000's in engineering costs alone as it would have to be drawn up and approved by engineers. It all has to fall under building codes to boot. Non of this is needed with air cooling.