Flow rate issues (Pump not the problem)

Reality

[H]ard|Gawd
Joined
Feb 16, 2003
Messages
1,937
I just finished tubing up my build, it's using acrylic tubing, this is the loop order:

Cylinder Res > Pump (Tried both DDC 18w & PMP-500) > 240mm/60mm Rad > 2x GPU's in Parallel > Mobo Block > CPU > 360mm/60mm Rad > 360mm/60mm Rad > Flow Meter > Res

Now, the pump & 240mm rad are in the pedestal of my Caselabs M8, so the flow is being pushed upwards against gravity (the two 360 rads are in the roof), and the return goes back down and into the bottom of the reservoir

http://www.youtube.com/watch?v=q9Uwtzn1MZ0

This is a video from a few days ago before I installed the pump, but you can see the loop order

My flow meter is a Koolance unit which sends it's data to my fan controller's LCD screen

It's reporting 2200rpm (which is 2.2L/m) initially, but then the flow rate gradually decreases down to under 900rpm (0.9L/m)

The Pump RPM stays constant at 4960, so the pump seems to be working perfectly (again, DDC 18w, with a Petra acetal top)

Should I add a second pump? I'm about to do it, just to see what happens
 
Do the temps rise as the flow drops? If you're still seeing good temps over a long run at full load on the system i'd say either the flow meter is borked or the fan controller can't handle the data.

Can you visually see the flow in the res? Does it appear to stay consistent? That is a fairly big loop but radiators aren't usually horribly restrictive, i would think that pump could handle it.
 
How many 90's did you put in there? That plus the motherboard block is your flow problem.
 
that wouldn't explain the gradual decrease, it would just be restricted from the start, there's 7 90°'s, I added a pmp500 in series and the flow starts out at 4 L/m, then gradually drops to 2.6 L/m
 
Last edited:
Yep. A drop of 60% in fluid movement should make a few degrees difference in temps.

What seems to happen to your temps?
 
Temps are irrelevant, the system isn't powered on yet, still trying to get the pump situation resolved first
 
Could air in one of the rads be the culprit? Getting air out of roof mounted rads in a case this size is very difficult
 
How long timewise is the actual decrease taking from first turning on the pump? I'm guessing an air pocket in a top radiator, vacuum lock somewhere, or perhaps pulling some air into the pump when your tubing turns 90 down out from straight coming from reservoir to go down to pedestal pump(s)
 
How long timewise is the actual decrease taking from first turning on the pump?

the decrease is slow but gradual. the flowrate goes from 4L/m to 2.6L/m over the course of about 2 minutes

Well. No, it could be a malfunction of your controller or flow sensor.

I've doubt the fan controller is faulty, as it's simply reporting RPM's. The flow sensor... I guess I can't say for sure it's faulty, but I'd only start investigating that as a last resort
 
update, it was an air/vacuum lock in one or more radiators, just as an aside, the amount of head pressure I'm getting with a DDC-1T + PMP-500 in series is insane.. lol
 
update, it was an air/vacuum lock in one or more radiators, just as an aside, the amount of head pressure I'm getting with a DDC-1T + PMP-500 in series is insane.. lol

Glad you got it sorted out, top mount radiators can be a bitch to get all the air out of. When I had one I would actually set the case on its back so the barbs were at the top of the rad then turn on the pump to help it push that air out. It also helps to not have the loop sealed off completely, least I think it helped me.
 
Last edited:
Glad to hear you got it sorted out :). Let us know how things are running(temps, ect)once you finish your rig.
 
I have the machine all wired up, just waiting for my PSU to get back from RMA
 
Back
Top