Overclock.net banner
1 - 14 of 14 Posts

·
Twin Turbski
Joined
·
301 Posts
Discussion Starter · #1 · (Edited)
From my research, at least with my own custom loop, it seems to be the primary factor in bringing temps down and evening out temp spikes in general. In experimenting with my loop, which contains a dual core 120mm + one 360mm radiator with a total of 4 PWM fans (one on the 120mm, 3 on the 360mm) and about 1.5 quarts of reserve water contained in two reservoirs. People have told me there is no significant gains to be had by using a larger reservoir, but I figured why not, the more volume in the system the better, right? So I stacked two alphacool 5.25" reservoirs on top of each other, with the freezemod pump hidden in the back of the drive bay area.

Question, if my calculations are correct, and the speed at which the water circulates is the primary element in loop efficiency, I should be investing in a high flow water pump, no? Can anyone back me up on this? It's pretty crazy, because if I leave the pump at about 40% speed, in torture testing, temps level off around 190*F, however, if I go full throttle on the pump, which works out to about 3400rpm, temps level out at about 150*F under the same torture test.

My question is, is this testing accurate, in other words, can anyone confirm I am on the right track here? And secondly, whats the best way to further improve loop circulation? I think obviously I need a second pump, but I'd like to get something with substantially more flow. I have this pump currently installed. It's a freezemod unit:

Liquid Product Black Font Material property
.


I understand that right angles are also not so great for flow, but I only have two of them anyway, one coming out of the pump itself and then one on the intake line leading to the larger 360mm radiator.

Questions:
This pump seems pretty powerful, but is there anything else on the market that I can replace it with that will substantially boost circulation in the system? Or should I just add a second, identical freezemod pump? I like the idea of two pumps for a number of reasons, primarily for best loop performance but also aesthetics, more tubing and more complexity in the system appeals to me and IMO looks better than just a basic setup with a couple water lines here. Any other tips here? For improving water flow? I was amazed at the temp differences from 50% vs 100% pump speed.

I was looking at this pump as well, which has a higher flow rate of 1200L/H but will not fit in the 5.25" drive bays due to space constraints. However, I think I can mount it under the CDROM cage area, and behind the 360mm radiator. What do you guys think?


Also looking at this one: 1200L/H
 

·
Twin Turbski
Joined
·
301 Posts
Discussion Starter · #2 ·
System layout:
Two reservoirs in front, pump housed behind lower reservoir (800L/H)
But I think my system with lots of tubing and two radiators, the motor is struggling a bit...

Computer Computer case Gadget Computer hardware CPU

Hood Automotive tire Bumper Automotive lighting Motor vehicle
 

Attachments

·
Registered
Joined
·
2,259 Posts
System layout:
Two reservoirs in front, pump housed behind lower reservoir (800L/H)
But I think my system with lots of tubing and two radiators, the motor is struggling a bit...

View attachment 2524769
View attachment 2524768

Ideally you want 1GPM. But even if you’re at 0.8GPM it’s not the end of the world. 1GPM is 227L/h. You’ll never have 800L/h. It wouldn’t be very beneficial even if you could. Fans play a much bigger role than flow rate as long as you have somewhat decent flow.

As a frame of reference, in my system I have 4 radiators and a cpu block. Probably 15-20x 90 degree connectors and 10 Quick disconnect fittings. With a single D5 pump at max speed I get around 195L/h (0.85GPM). With 3x D5 pumps it goes up to 315L/h (1.38GPM). Realistically a single pump at 195L/h with all the restriction in my loop would be fine if I weren’t OCD.

If you’re overly concerned about it, get a flow meter. But generally speaking 2x D5 pumps is more than enough for the majority of builds. And you could probably get away with even just 1.
 

·
Cheesebumps!!
Joined
·
2,848 Posts
optimize your loop, get rid of your 90 degree bends, that will help increase the flow a bit and lessen restriction, also either you use dual D5 or opt for a DDC3.2 pump, I use 2 DDC3.2 pumps from Barrow and they run silent with 1000lph and around 6-7m head pressure on 2x360 rads with 2 QDC's, from my previous 6.7lpm with 2x360's w/o QDC's now I am down to 4.4-4.6lpm, I preferred the easy to maintain route rather than frequently drain and disassemble..(though I take the reading on flow speed with a grain of salt)
 

·
Registered
Joined
·
624 Posts
Are you saying your water temps changes by that much or cpu/gpu temps? Either way, those are very high temps.
 

·
Twin Turbski
Joined
·
301 Posts
Discussion Starter · #6 · (Edited)
Are you saying your water temps changes by that much or cpu/gpu temps? Either way, those are very high temps.
I have no measurement for actual water temp, I am relying on actual CPU temps to tell me what is going on here when I change other parameters. I run the basic CPU-z torture test with all 4 fans at a static 1500rpm, for example. The ONLY change made is the actual pump rotational speed. with the pump at maximum speed (DC mode), or 3400rpm, I see peak temps of about 155*F.

If I then go back in and ONLY make one change, that is, switch to PWM mode on the pump and put it at like 2000rpm, the peak temps climb much more quickly and level off at about 190*F. So you can see why I am obsessed with flow rate, since that is the only thing changing here and I am seeing such a huge difference between the two settings.

I also have a feeling that the flow rate has declined substantially since two years ago when I put the system together. Meaning, it used to sound like a waterfall in the lower reservoir, with the water hitting the front cover of the lower reservoir. Not I do not hear it at all. It's hard to measure since I have no flow rate meter or anything, but just looking at the reservoir, you can tell, it is nowhere near the power it used to be, despite the pump RPM being right where it should be. It's possible this is due to the fact that I added the second 360mm radiator? Perhaps it just doesnt have enough power to push all that water through two radiators and two reservoirs? In any event, any changes will have to wait until I get paid next month. But it just seems the pump is not flowing anywhere near 800L/H. The other day I flushed the system and started over with clean distilled water and no changes were evident. So it's not gummed up or anything like that. I'm pretty certain that with the addition of a second pump I will see much better flow.

System also has a decent water block, its a heatkiller IV all copper water block that seems to get pretty good reviews as a high performance block.
 

·
Registered
Joined
·
3,437 Posts
I have no measurement for actual water temp, I am relying on actual CPU temps to tell me what is going on here when I change other parameters. I run the basic CPU-z torture test with all 4 fans at a static 1500rpm, for example. The ONLY change made is the actual pump rotational speed. with the pump at maximum speed (DC mode), or 3400rpm, I see peak temps of about 155*F.

If I then go back in and ONLY make one change, that is, switch to PWM mode on the pump and put it at like 2000rpm, the peak temps climb much more quickly and level off at about 190*F. So you can see why I am obsessed with flow rate, since that is the only thing changing here and I am seeing such a huge difference between the two settings.

I also have a feeling that the flow rate has declined substantially since two years ago when I put the system together. Meaning, it used to sound like a waterfall in the lower reservoir, with the water hitting the front cover of the lower reservoir. Not I do not hear it at all. It's hard to measure since I have no flow rate meter or anything, but just looking at the reservoir, you can tell, it is nowhere near the power it used to be, despite the pump RPM being right where it should be. It's possible this is due to the fact that I added the second 360mm radiator? Perhaps it just doesnt have enough power to push all that water through two radiators and two reservoirs? In any event, any changes will have to wait until I get paid next month. But it just seems the pump is not flowing anywhere near 800L/H. The other day I flushed the system and started over with clean distilled water and no changes were evident. So it's not gummed up or anything like that. I'm pretty certain that with the addition of a second pump I will see much better flow.

System also has a decent water block, its a heatkiller IV all copper water block that seems to get pretty good reviews as a high performance block.
It doesn’t cost you much to tear down and clean the loop. Chances are either the pump or CPU block is obstructed.
 

·
Twin Turbski
Joined
·
301 Posts
Discussion Starter · #8 ·
It doesn’t cost you much to tear down and clean the loop. Chances are either the pump or CPU block is obstructed.
I just cleaned the lines and the pump and changed the water out, decreased temps by 1*F, lol I think just because I've got two radiators and two reservoirs its struggling to keep up...
 

·
Registered
Joined
·
640 Posts
From my research, at least with my own custom loop, it seems to be the primary factor in bringing temps down and evening out temp spikes in general. In experimenting with my loop, which contains a dual core 120mm + one 360mm radiator with a total of 4 PWM fans (one on the 120mm, 3 on the 360mm) and about 1.5 quarts of reserve water contained in two reservoirs. People have told me there is no significant gains to be had by using a larger reservoir, but I figured why not, the more volume in the system the better, right? So I stacked two alphacool 5.25" reservoirs on top of each other, with the freezemod pump hidden in the back of the drive bay area.

Question, if my calculations are correct, and the speed at which the water circulates is the primary element in loop efficiency, I should be investing in a high flow water pump, no? Can anyone back me up on this? It's pretty crazy, because if I leave the pump at about 40% speed, in torture testing, temps level off around 190*F, however, if I go full throttle on the pump, which works out to about 3400rpm, temps level out at about 150*F under the same torture test.

My question is, is this testing accurate, in other words, can anyone confirm I am on the right track here? And secondly, whats the best way to further improve loop circulation? I think obviously I need a second pump, but I'd like to get something with substantially more flow. I have this pump currently installed. It's a freezemod unit:

View attachment 2524762 .


I understand that right angles are also not so great for flow, but I only have two of them anyway, one coming out of the pump itself and then one on the intake line leading to the larger 360mm radiator.

Questions:
This pump seems pretty powerful, but is there anything else on the market that I can replace it with that will substantially boost circulation in the system? Or should I just add a second, identical freezemod pump? I like the idea of two pumps for a number of reasons, primarily for best loop performance but also aesthetics, more tubing and more complexity in the system appeals to me and IMO looks better than just a basic setup with a couple water lines here. Any other tips here? For improving water flow? I was amazed at the temp differences from 50% vs 100% pump speed.

I was looking at this pump as well, which has a higher flow rate of 1200L/H but will not fit in the 5.25" drive bays due to space constraints. However, I think I can mount it under the CDROM cage area, and behind the 360mm radiator. What do you guys think?


Also looking at this one: 1200L/H
What are these pumps ? Are you trying to cut cost ?

Just go on aliexpress, D5 clone from freezemod is 35$ or less. D5 top from freezemod about 15$. Freezemod res another 15$ or so.

For 75$ or less you'll be running a D5 clone which is very close to EK D5 performance and silent.
 

·
Registered
Joined
·
266 Posts
Your biggest problem that will be effecting flow is the hard 90 straight from the pump, that's a no no. The 800lph for the pump is the theoretical speed without anything in the loop to add restriction.
 
  • Rep+
Reactions: storm-chaser

·
Premium Member
Joined
·
4,195 Posts
Get a temp sensor for the water. You're running blind without one
 
  • Rep+
Reactions: storm-chaser

·
Registered
Joined
·
266 Posts
Also a flow meter would help, you could go THIS route to add 2x pumps in that case
 
  • Rep+
Reactions: storm-chaser

·
Twin Turbski
Joined
·
301 Posts
Discussion Starter · #13 ·
Your biggest problem that will be effecting flow is the hard 90 straight from the pump, that's a no no. The 800lph for the pump is the theoretical speed without anything in the loop to add restriction.
I think I had to do that hard 99* turn out of the pump since it was a direct 3-4" away from the inlet on the first reservoir. I suppose I can go back in, remove that and then ad a few more inches of clear tubing in a loop from pump to reservoir so it doesn't kink. I will also be getting rid of the other 90* bend by using a straight through compression fitting since the 90* turn even then will not kink the line.
 

·
Twin Turbski
Joined
·
301 Posts
Discussion Starter · #14 ·
The inline filter I installed when I put the loop together was part of the problem. I just cleaned that out, it was almost completely covered with gunk and very restricted.

I thought for sure I'd be back in business, but apparently there is also something else obstructing the flow. Because I'm not hearing the water fall yet (as I mentioned, when first set up the flow was so powerful it would impact the front of the reservoir with a noise that resembles a small waterfall)

I took some vinegar out last night and I'm going to run about 10% vinegar in my loop for the next couple days, just to clean everything out.

I have to disassemble the loop so I will find the culprit. The reason is because I have another 800L/h pump and 120mm radiator on the way. This is crazy, for a case that was designed for nothing more than air cooling, I have an alphacool monsta 120mm rad, a fractal designs 360mm radiator and soon another 120mm near the front. Amazingly, everything fits in the case! I will post pics when complete.
 
1 - 14 of 14 Posts
Top