EKWB Revo D5 problems

[rembomy]

So, I have an issue where my system does not properly bleed from air, It feels that this pump is weak for my system, but I had 0 issues with another pump prior to this.

 

I was using http://www.xs-pc.com/water-pumps/x2o-420-single-bayrespump

Now I am using https://www.ekwb.com/shop/ek-xres-140-revo-d5-pwm-incl-pump

 

My question, how is it possible that my old XSPC pump was able to bleed the system mostly withing hours and new EKWB pump is not able to do so with 2 days and very extensive case shaking/turning. I can see water moving in the rez, but there are air bubbles (i don't mean the micro ones) in the tubes and they are not moving or are stuck in one place, did not see this with the old one. 

 

Picture of my loop attached. Running 360 rad, 240 rad, 2 GPU and 1 CPU, this was my setup when using the XSPC pump as well, even kept the inlet and outlet on radiators same. 

Main points where larger bubble colonies form are right between CPU and GPU and right after GPU outlet. 

If I would use 90 degree fittings somewhere and shorten the tubes could this help? 

 

Any ideas, suggestions would be helpful.

[Limecat86]

41 minutes ago, rembomy said:

So, I have an issue where my system does not properly bleed from air, It feels that this pump is weak for my system, but I had 0 issues with another pump prior to this.

 

I was using http://www.xs-pc.com/water-pumps/x2o-420-single-bayrespump

Now I am using https://www.ekwb.com/shop/ek-xres-140-revo-d5-pwm-incl-pump

 

My question, how is it possible that my old XSPC pump was able to bleed the system mostly withing hours and new EKWB pump is not able to do so with 2 days and very extensive case shaking/turning. I can see water moving in the rez, but there are air bubbles (i don't mean the micro ones) in the tubes and they are not moving or are stuck in one place, did not see this with the old one. 

 

Picture of my loop attached. Running 360 rad, 240 rad, 2 GPU and 1 CPU, this was my setup when using the XSPC pump as well, even kept the inlet and outlet on radiators same. 

Main points where larger bubble colonies form are right between CPU and GPU and right after GPU outlet. 

If I would use 90 degree fittings somewhere and shorten the tubes could this help? 

 

Any ideas, suggestions would be helpful.

90 degree fittings would only add extra restriction.

 

Does the pump run at full speed? Even with no PWM signal?

 

These pumps used to had an issue running at 40% when no PWM signal was provided but I believe they fixed this a while ago.

 

I'm also very interested what's up with your Revo D5. I have experienced the exact same with that pump. I replaced it with a dual DDC bay res as it was barely capable moving the liquid around even running on 100%. I had my CPU + 1 GPU, a 360 and a 240 rad in my loop then.

 

Now the dual DDC's are running on 25% PWM and moving the liquid arround without any effort. The low PWM setting makes them silent and they don't heat up that much.

 

 

[rembomy]

I have tried PWM at 100%, under 100% it looks like nothing is moving. Even tried to connect 5V to the PWM and nothing changed, 100% was as fast as it was on MB. 

I wrote to EKWB support, but haven't received any reply yet, maybe they have some ideas, funny part is that when I compare it to my old pump it performs worse for double the price.

 

Also, i thought of those few 90 degree fittings because I could shorten the loop about 30 cm.  

[Limecat86]

40 minutes ago, rembomy said:

I have tried PWM at 100%, under 100% it looks like nothing is moving. Even tried to connect 5V to the PWM and nothing changed, 100% was as fast as it was on MB. 

I wrote to EKWB support, but haven't received any reply yet, maybe they have some ideas, funny part is that when I compare it to my old pump it performs worse for double the price.

 

Also, i thought of those few 90 degree fittings because I could shorten the loop about 30 cm.  

Yeah, I was pretty disappointed with the pump's performance. They're indeed expensive, and with such bad performance, it doesn't feel right at all. Maybe we are just unlucky and have a bad unit? Well, the pump does run. So maybe a bad impeller? A bad top? Who knows...

 

Heck, now I'm itching to get my XRES 140 Revo D5 installed see if I can figure something out. Could also try getting a top and reservoir (not a combo) and see how that works out.

 

About your loop, yes, you could always try a few 90 degree fittings and shorten the loop. It's best practice anyway to keep the loop as short as possible.

 

 

[rembomy]

4 hours ago, Limecat86 said:

Yeah, I was pretty disappointed with the pump's performance. They're indeed expensive, and with such bad performance, it doesn't feel right at all. Maybe we are just unlucky and have a bad unit? Well, the pump does run. So maybe a bad impeller? A bad top? Who knows...

 

Heck, now I'm itching to get my XRES 140 Revo D5 installed see if I can figure something out. Could also try getting a top and reservoir (not a combo) and see how that works out.

 

About your loop, yes, you could always try a few 90 degree fittings and shorten the loop. It's best practice anyway to keep the loop as short as possible.

 

 

I'll probably rebuild the loop next week as this weekend is already booked for a LAN, going to be an interesting one. I'll get back after this and let you know how it turned out. 

[rembomy]

So today I saw that the bubbles between CPU and GPU moved to GPU block, but never left there. As for the water, I can see it moving in reservoir, but my CPU temps are still going nuts compared to previous, I could clock to 4.8ghz using previous pump (temps were 78-82C @ 1.305v), right now I have to run stock with to keep it under 80C. I even tried to change paste and re-seat the CPU block, still bad results post pump change. 

[Mick Naughty]

Pump works flawlessly. Maybe you need to delid that cpu if you cant tame the temps. What are you doing to put the cpu under load?

[EK Luc]

Did you also change something with your CPU block since you swapped pumps?

[Eddie Current]

Test the pump outside of the system without water blocks. That will tell you if the pump is working or not.

[rembomy]

Forgot to update here, found the culprit, somehow there was some gunk stuck between radiator and cpu block, I guess when i was swapping out the pump some paper pieces made their way there... feels like triple facepalm

[Limecat86]

1 hour ago, rembomy said:

Forgot to update here, found the culprit, somehow there was some gunk stuck between radiator and cpu block, I guess when i was swapping out the pump some paper pieces made their way there... feels like triple facepalm

Thanks for the update. I haven't tried my D5 again yet. Not sure when I will though. My system runs fine in it's current state.

[EK Luc]

5 hours ago, rembomy said:

Forgot to update here, found the culprit, somehow there was some gunk stuck between radiator and cpu block, I guess when i was swapping out the pump some paper pieces made their way there... feels like triple facepalm

Cool, i'm glad you did find out what it was.

 

If people don't try, they don't make mistakes so they can't learn from them.  Never feel stupid about this.  We all make mistakes and watercooling is a universe where we can make a lot of mistakes.