EKWB D5 Pump - Impeller Dead?

Smffy

Smffy

Smffy

Soldato
Joined
4 Dec 2015
Posts
3,221
Location
London
Howdy,

So i've done quite a few rebuilds, replacing and changing hardware over the past 12 months and lately noticed much lower flow to the point that its now quite poor even full tilt. I've got 2 x EK D5s linked in a Bitspower pump top, what I've managed to deduce is that only one seems to be driving flow but the other I can hear an (reasonably faint) higher pitch noise of it spinning up but zero flow.

Presuming this means that its probably free spinning inside the impeller and not driving it? Thinking of a complete build change soon so will be able to investigate properly but has anyone had a similar issue?

1 x D5 is not currently strong enough to develop much flow even under full power but my rig is huge with a lot of fluid to move up long verticals. Yes people say 1 is enough but honestly moving 3 litres of fluid around a huge system its not adequate in my case:

ushxLjF.jpg

Pumps connected in one block, disconnecting the power to one or the other reveals the problem:

pwvhV1c.jpg

1. Am I right in my assumption?
2. Is it possible to replace/fix impellers?

Cheers!
 
The impeller in a D5 doesnt work like you are describing, there is no drive pin / shaft going into an impeller that has worked loose, its magnetic, this is one of the unique designs of the D5's, it eliminates shaft slippage, the impeller basically levitates inside the unit and the electro magnet in the unit causes it to spin.

The way I test mine and even clean it out is to get a jug of distilled water, attach 2 pieces of tube to the in and out, then switch it on, let it draw water from one jug and expel it another, maybe it has an air lock.
 
Jamin280672 said:
The way I test mine and even clean it out is to get a jug of distilled water, attach 2 pieces of tube to the in and out, then switch it on, let it draw water from one jug and expel it another, maybe it has an air lock.
Click to expand...

Ah ok, well electrically it sounds like something is happening. I don’t really need to test outside as it’s in the loop next to the other pump, if I unplug the one which is working there is zero flow and very low noise which is why I thought maybe somethings died... if I run the working pump the water is flowing over the dead one so again shouldn’t be an airlock in theory. (?)

Potentially something is stuck who knows - would have to drain and dismantle it all so might well do that in a week or so if I can be bothered ;)

Thanks for the response!
 
Jamin's description of the D5 is spot on. There's a ceramic bearing that can wear down the socket of the impeller and the floating impeller can then seize, but you'd hear that for sure. Are you're D5's variable or fixed speed? Might also be worth checking they are getting the correct voltage from the power connector. How are you monitoring the flow?

I'm surprised that one D5 won't run you're set up. I have more blocks than you in a partial series and parallel set up and run it to an external Mo-Ra3 radiator (1m runs back and forth) and I run the pump at 30 to 45% which is more than adequate (USB D5).
 
Mysterae_ said:
Jamin's description of the D5 is spot on. There's a ceramic bearing that can wear down the socket of the impeller and the floating impeller can then seize, but you'd hear that for sure. Are you're D5's variable or fixed speed? Might also be worth checking they are getting the correct voltage from the power connector. How are you monitoring the flow?

I'm surprised that one D5 won't run you're set up. I have more blocks than you in a partial series and parallel set up and run it to an external Mo-Ra3 radiator (1m runs back and forth) and I run the pump at 30 to 45% which is more than adequate (USB D5).
Click to expand...

Its previously run fine so not voltage i'd wager, seems like its just died but one of them (not sure which) is probably getting on a bit. I've tried various speeds but with the two at full tilt the flow was immense - visibly flow meter and the fluid in to the res.

I'm not sure if your blocks would explain the need for two, but this case is absolutely huge and running 4 x 60mm radiators (480, 360, 240, 240) with 16mm copper pipe around it. The evidence is there to see as with the single D5 flow can be 'ok' but on high speed vs. two I could run very low speed (like yours) and achieve the same flow. its pushing up so pretty large verticals and quite a large distance through it all. Worth saying I had two GPUs in blocks until reasonably recently again better flow.

One thing of note on flow is reversing my GPU inlet (at the moment) which isn't exactly recommended and restricts it a little.

I'll maybe take it apart next week or week after to try and sort it... Just disconnected the working pump again and only the dud, you can definitely hear the motor spin up but sounds like its not moving the impeller so its either stuck or dead or both ;)
 
Last edited:
The ceramic ball bearing takes years to wear down to an unusable state.

Problem is a pump in that kind of top will suffer massively if a pump fails or is running at a lesser speed. One pump pumps into the next pump which pumps it around the system, to give twice the pressure at the same flow rate essentially. If the first pump dies/running slow then the second one is essentially pulling liquid through the other pump which will cause a massive flow drop depending on the clearance on the impellers. If the second pump dies/running slow then the first has to pump through the second pumps impeller just to give you some flow. So its quite vital to have them running at the same speed.

I would as above and check all rpm's. Possible strip down otherwise and IF you do, might aswell clean out the rads as I've had them block up before. Usually stick mine on a hosepipe for 10 mins, flush with distilled, dry out with compressor, rebuild and flush again a few times.
 
Last edited:
tyler_jrb said:
The ceramic ball bearing takes years to wear down to an unusable state.

Problem is a pump in that kind of top will suffer massively if a pump fails or is running at a lesser speed. One pump pumps into the next pump which pumps it around the system, to give twice the pressure at the same flow rate essentially. If the first pump dies/running slow then the second one is essentially pulling liquid through the other pump which will cause a massive flow drop depending on the clearance on the impellers. If the second pump dies/running slow then the first has to pump through the second pumps impeller just to give you some flow. So its quite vital to have them running at the same speed.

I would as above and check all rpm's. Possible strip down otherwise and IF you do, might aswell clean out the rads as I've had them block up before. Usually stick mine on a hosepipe for 10 mins, flush with distilled, dry out with compressor, rebuild and flush again a few times.
Click to expand...

Yeah unfortunately it’s pump 1 that seems to have failed... I will probably drain and sort it soon when I have time, rads and whole system cleaned out two months ago including respraying the radiators so pretty thorough ;)
 
tyler_jrb said:
Problem is a pump in that kind of top will suffer massively if a pump fails or is running at a lesser speed. One pump pumps into the next pump which pumps it around the system, to give twice the pressure at the same flow rate essentially. If the first pump dies/running slow then the second one is essentially pulling liquid through the other pump which will cause a massive flow drop depending on the clearance on the impellers. If the second pump dies/running slow then the first has to pump through the second pumps impeller just to give you some flow. So its quite vital to have them running at the same speed.
Click to expand...

According to this test the drop in pressure when a pump fails in serial configuration is minimal (2%): https://martinsliquidlab.wordpress.com/2011/04/26/pump-setup-series-vs-parallel/2/. Parallel on the other hand, huge pressure drop (83%).
 
Bugbait said:
According to this test the drop in pressure when a pump fails in serial configuration is minimal (2%): https://martinsliquidlab.wordpress.com/2011/04/26/pump-setup-series-vs-parallel/2/. Parallel on the other hand, huge pressure drop (83%).
Click to expand...

Like I said it depends on the clearance on the impellers I was expecting it to be a lot tighter to be honest like most pumps hense my post.

However would be interesting to see how they react if one pump is running at 50% speed of the next/previous.

Parallel is used quite widely but usually there is non return valves to stop flow issues if one dies.
 
You must log in or register to reply here.
Back
Top Bottom