Duallie's with H2O - series or parallel?

[krp]

Water-cooled Duallie's (2 cpu's), (and assuming only ONE pump to let the whole mess fit in a midtower case), which is the best way to route it?

1) routed in series? - 2nd cpu gets water warmed by 1st cpu... however flow rate is same for both...

or

2) routed in parallel? - each cpu gets cool water, BUT no guarantee flow rates are equal...

which is more important? Damfino... lets hear from the experts!

[#Rotor]

Parallel... is the only way for me...

[redleader]

Series. Less room wasted, higher flow per CPU.

Furthermore the idea that a second block heats the coolant is nonsense. The difference in temp before and after the second block will be a tiny fraction of a degree.

[ambient]

The difference in temp before and after the second block will be a tiny fraction of a degree.

Then where is the water getting hot? Does the water temp not change when run through the block, say if water was 21c coming out of the rad into the first block it would be how hot going into the second? only 22c? I really don't know the answer to above questions. I just read that people see a delta of about 8-10c load and idle in water temp. So if that is true where is the water getting hot?

[Brad]

in series one will be a little warmer, so I'd make sure this cpu is your weaker one.

in parrallel you will have half the flow going to them

if you have a high volume pump, it probably doesn't matter which way you go

[Brad]

btw rotor, your system plumbing is really cool. Does your NB actually get that warm?

[#Rotor]

NO ! but how will a heatsink in between all those blocks go over.....

[redleader]

Then where is the water getting hot? Does the water temp not change when run through the block, say if water was 21c coming out of the rad into the first block it would be how hot going into the second? only 22c? I really don't know the answer to above questions. I just read that people see a delta of about 8-10c load and idle in water temp. So if that is true where is the water getting hot?

Did you read my post? You're looking at tiny fractions of a degree. Water is getting hotter, just immeasurably so.

Off the top of my head:

2GPM = 125ml/s
100w CPU adds 100J to the coolant per second
SH of water = 4.2 J/ml

Therefore assumeing that it takes a molecule of H20 1 second to pass through your block,

100/4.2 = 24 / 125ml = .2C

So with 120GPH of flow a hundred watt CPU adds .2C to the water. Again you can't even measure a change that small, so its basically nothing.

[pHaestus]

I would run dual loops each with their own pump (common res perhaps?) if I were using a parallel setup and otherwise I would run in series. I think the performance decrease from cutting flow rate in half will be larger than the contribution of heat from the first block to the second. It isn't hard to get 1.5 GPM with a pump that will fit into a case, but it is much harder to get more than that. Cutting flow rates down to 0.5-0.75 gpm will markedly decrease performance (hitting the transition between laminar and turbulent for the water at such low flow rates).

[Mr. Thompson]

With a 700 GPH pump and Maze 1Cs, I had better results in series. I run my Xeon Quiet Power box in series too. If you look at the table from BillA below, you will note it's hard to get any more than a 0.5 C rise in water temp after a waterblock if you have decent flow.

load W . . flow . . . die T . . bp T . . inlet T . outlet T . C/W . . Btus*. . Watts . %eff
45.00 . .2.0/7.57 . . 36.5 . . 27.1 . . 25.00 . . 25.08 . . 0.256 . . 144 . . . 42.2 . . 94
45.01 . .1.5/5.68 . . 36.7 . . 27.3 . . 25.00 . . 25.11 . . 0.260 . . 148 . . . 43.4 . . 96
45.01 . .1.0/3.79 . . 37.2 . . 27.7 . . 25.00 . . 25.17 . . 0.271 . . 153 . . . 44.8 . . 99
45.01 . .0.5/1.89 . . 38.2 . . 28.7 . . 25.00 . . 25.33 . . 0.293 . . 148 . . . 43.4 . . 94

75.02 . .2.0/7.57 . . 45.0 . . 29.4 . . 25.00 . . 25.14 . . 0.267 . . 234 . . . 68.6 . . 91
75.02 . .1.5/5.68 . . 45.4 . . 29.8 . . 25.00 . . 25.18 . . 0.272 . . 243 . . . 71.2 . . 95
75.02 . .1.0/3.79 . . 46.1 . . 30.4 . . 25.00 . . 25.28 . . 0.281 . . 251 . . . 73.6 . . 98
75.02 . .0.5/1.89 . . 47.8 . . 32.0 . . 25.00 . . 25.56 . . 0.304 . . 251 . . . 73.6 . . 98

105.04 .2.0/7.57 . . 53.3 . . 31.6 . . 25.00 . . 25.19 . . 0.269 . . 341 . . . 99.9 . . 95
105.04 .1.5/5.68 . . 53.9 . . 32.1 . . 25.00 . . 25.26 . . 0.275 . . 350 . . . 102.6 . . 98
105.04 .1.0/3.79 . . xxxx . . 33.0 . . 25.00 . . 25.39 . . xxxxx . . 350 . . . 102.6 . . 98
105.04 .0.5/1.89 . . 56.9 . . 35.2 . . 25.00 . . 25.77 . . 0.304 . . 346 . . . 101.4 . . 97

[Hellion_Prime]

When i finally get the budget to build my dual amd box, I plan on going with seperate loops for each processor. Overkill maybe, but it'll sure as hell look pretty.

--HP

[Brad]

yeah, completely seperate systems are the best of course. better have a big tower case

[EMC2]

Comments on the flow rate - depends on the pump and the blocks. Yes, you split the flow in ~half when going parallel, but you also cut the flow resistance by almost half as well, meaning your pump's flow rate will go up with them in parallel. What the net result is regarding flow depends upon how restrictive the blocks are and where you are on the PV curve of the pump.

General comments as far as temps :

Pelts + CPU = significantly diffferent system from only a CPU.

Q = h * DeltaT (heat flux equals heat transfer coefficient times the temperature delta)

End result - there is no "one size fits all" If you are going with 1 pump and rad, try both ways and see which works best for your system configuration like MT did.

[pHaestus]

Also consider that testers of blocks and radiators usually use pumps that may not be typical for "in case" watercoolers. Joe uses the Hydrothruster and could get nearly 3 GPM, and I believe BillA uses a Little Giant and also can get about 3 gpm in a loop. People need to carefully choose the pump that they use if they want high GPM flow rates. The hydrothruster 500 GPH pump and the Little Giant pumps will push water through restrictions much better than any of the Danners or Eheims. The Hydrothruster just flexes its muscles and increases power when it encounters flow resistance, but eheims and danners simply can't. You need to look at the max PSI the pump can handle and not the GPH flow rate in unrestricted systems. I am in particular looking at the 500GPH Little Giant for its cost/performance ratio. Who cares if it fits in the case!

[Brad]

just buy a farm pump then