D5 speed question

[Snorlaxxx]

So, my GPU block for my 290 came today anf I finally finished building my loop.
Until now I had only a 30mm 480 radiator and cpu + universal gpu block, now I have a full cover gpu block and I added a 30mm 240, a 45mm 240 and a 45mm 420 radiator. I have about 2 Liters of water in my loop.

I had my D5 always on setting 1. Quiet and sufficent, but now I don't know if it's enough. I will do some test and observe it when my pc is up and running. It is still in bleeding and leak testing phase.

My question is: Do you think settings 1 may be still enough for this loop?
I'm not concerned about flow rate, since it has been proven not to be that big of a factor and with this much rad surface I really doubt I get any temp issues.
I'm more thinking about the pump itself and if it suffers on a lower setting.
Usually its the higher the speed the bigger the strain on the pump, but might a too slow turning pump have to work against some sort of a resistance/gravity, thus beeing exposed to an even bigger strain?

I know the D5 is a centrifugal pump and maybe my thinking relates more to a impeller pump, but I still wanted to ask you guys

 

[RnRollie]

that is a lot of raddage , it'll definitely keep cool

The biggest restriction is not so much come from the rads... assuming these are Alphacools and not Feser Admiral variants... it's coming from all the fittings which are inevitable.

Maybe you can work out the total drop by deriving some numbers from martins site and err on the side of caution (add up all the losses & multiply with 1.5).

http://martinsliquidlab.org/pump-planning-guide/


Without flow meter, we can't know of course.. but my guess is that you're very near 0.5 GPM. Which might still be enough for cooling purpose, but bleeding becomes an issue.
The be safe, i'ld switch to setting 2

Gravity doesn't play a role unless you've got a Res in a very "tall/high" loop.
But i assume you didn't install the res on the floor above you and the pump in the basement below

 

[Snorlaxxx]

I don't know.. the theory behind the calculation is simple, but confusing at the same time, because I don't have any exact numbers.
According to Martin's list @1gpm, I'm around 3,1psi for my components.
That would mean I have to run my pump on setting 3 to get approx. 1,5gpm.

The confusing part is the curves are not linear and the higher the pump pressure goes, the more pressure drop every component has.
Where I get 3,1psi drop @1gpm, I might only have 1psi drop @0,5gpm or even less.
This would mean settings 2 is enough.

From my observation settings 1,5 still gives good movement. Going down to 1 shows some bubbles in part of the loop flowing backwards. Maybe this is an indication of too slow pump speed.
Keep in mind I am still bleeding and this may change when the air is out.

For me, as long as it is quiet and cool I don't really care, but like I said, my concern lies whether the pump takes damage or works under load at too low speeds.

 

[Snorlaxxx]

Oh, and... yes those are Alphacool rads. I'm a low rpm / silent fanatic. I know I have way too many radiators for what I am cooling. I just took the opportunity to add them, since I already opened my loop for the gpu. And who knows, maybe I'll add a second R9 290 down the line... Haswell + dual 290 = lots of heat.

Fittings wise I have almost exclusively straight ones. Just two rotary 90° ones I had to use. 3/8 ID tubing.

And no, my pump is not in my basement, but if I have to turn it way up I might put it there
For now I have a aquacomputer D5 pump/res. (450ml).
This shouldn't be too restrictive.

 

[RnRollie]

all in all .. setting 3 while bleeding, then setting 2 for daily usage
setting1 might be too low.
there shouldn't be an audible difference between setting 1 & 2 anyways, unless you are Clark Kent

 

[GTXJackBauer]

I wouldn't go less than setting 4 with such that much heat surface and other components involved but that's just me. Plus D5 pumps are quieter from my understanding than DDC pumps.

 

[Spawn-Inc]

i run dual d5's (one vario set to 4 to match the one fixed) and they are very quiet, the only problem is a bit of vibration noise i fixed by unscrewing them from the brackets.

BUT, i've got 22 fans in my system so i'm not worried about noise.

i also did a very un scientific test with my 1 D5 vario with just my new GTX780, some tubing and a 2 gallon bucket of water. with evga's oc scanner and the pump on setting 1 the temp was 23c. on setting 5 it dropped to 20C. that's the gpu core temp not the water temp.

 

[Snorlaxxx]

Thanks. For now I have it running at setting 2. This gives good visible flow and is quiet at the same time. I might turn it down when i get to test a little more.

Right now I'm busy figuring out how to set my Aquaero's fan profiles. I don't really need all radiators running at idle, so I disabled the 480 until my water gets to 30°C.
Too bad we have very hot days right now. Need to focus more on the delta, rather than the water temp itself. Everything over 30° looks wrong to me somehow
I keep thinking how can I get higher temps with three additional rads?! But looking at the delta I can clearly see there's a strong difference in temps and fan speed.

But I'll have to play a little more with it.