Quote:
Originally posted by airspirit
One other thing to consider as an improvement to my design is using multiple weaker (quieter) pumps, with one on each line after your manifold. As far as the radiator goes, size and shape doesn't matter, as long as it is easy for you to maintain when filling and draining. Plastic is outstanding, I think, because it will not corrode, and can look quite nice if you get the clear stuff, though that is up to you.
As for radiators, you can just get three or four heatercores and wire them up in parallel rather than use one big core. That shouldn't be a problem. I would, however, recommend using the largest fans that you can find to cool them all, as larger usually means more powerful per sound output.
For manifolding, you should be able to find much more efficient ways to do it than I did ... using X intersections of PVC would have done the trick for me, though for some reason I insisted on designing for four blocks ....
For cheap and good pumps if you go with multiples, try a stack of Danner 3 pumps ... they are cheap, reliable, and silent if mounted correctly.
Oh, and try to plan for adding temp sensors into the coolant and such ... I really wish that I did in the beginning! Also, make sure to provision ways to fill and drain it ... that will be your one greatest headache if you don't. I mean it ... a multiple system cooling circuit can be a bastard to maintain in the best of circumstances (though minie is easier than my old setup), and in the worst ... well, I wouldn't curse that on anyone!
|
Sorry I missed your post airspirit musta lost it in the jumble of replys.
As for the design I think I am going to go with two Velocity T1 pumps for the system. Both pumping in the W/C cabinet. One pulling water from the computer cabinet and on pumpig into it.
As for the temp probes I am planning to use sensors before and after the rad to get temps of the water there as well as posibly at each branch before and after the computer case to guage how much heat is being absorbed in the loop.
Now i just have to find some fairly accurate temp probes or build some to mount in the lines. I was thinking of using T fittings with the probe in the top of the T sealed in with a cap and silicon to prevent leaks and maintain the looks of things.
If I go with a water chiller like I am thinking about I might be able to use these in some sort of control circuit to cycle the chiller or something. That way the electric bll isnt quite so high.