F5F Stay Refreshed Power Users Overclocking Dual loop compared to 2 pumps in series versus 2 pumps in parallel

Dual loop compared to 2 pumps in series versus 2 pumps in parallel

Dual loop compared to 2 pumps in series versus 2 pumps in parallel

Pages (2): Previous 1 2
V
VinylGuarder
Member
153
10-06-2016, 07:20 PM
#11
you can observe it directly how this operates. i have completed all the settings that were discussed here. my best final setup, that functioned well, is what i refer to as a complete parallel loop—every component and radiator are paired together. however, as rubix has emphasized several times, the actual loop remains a serial arrangement from components to radiators.
V
VinylGuarder
10-06-2016, 07:20 PM #11

you can observe it directly how this operates. i have completed all the settings that were discussed here. my best final setup, that functioned well, is what i refer to as a complete parallel loop—every component and radiator are paired together. however, as rubix has emphasized several times, the actual loop remains a serial arrangement from components to radiators.

D
Dingdongyou
Member
220
10-06-2016, 09:18 PM
#12
Yes, but a single loop remains overall, functioning as a series. Splits could be used to supply the parallel parts of the loop, though these splits eventually reconnect into a unified serial section. A parallel split would lower restriction, potentially speeding up certain sections if pumps are positioned after the split instead of before. However, this means feeding two 1/2"ID pumps with a single 1/2"ID tube, effectively halving your flow capacity. This remains a serial segment of the loop, which will naturally be more restrictive. It seems that while such setups aren't widely available commercially, most companies provide serial pump options. I’m leaning toward a series configuration if I go dual pump, but I’d appreciate any insights from someone who supports parallel operation.
D
Dingdongyou
10-06-2016, 09:18 PM #12

Yes, but a single loop remains overall, functioning as a series. Splits could be used to supply the parallel parts of the loop, though these splits eventually reconnect into a unified serial section. A parallel split would lower restriction, potentially speeding up certain sections if pumps are positioned after the split instead of before. However, this means feeding two 1/2"ID pumps with a single 1/2"ID tube, effectively halving your flow capacity. This remains a serial segment of the loop, which will naturally be more restrictive. It seems that while such setups aren't widely available commercially, most companies provide serial pump options. I’m leaning toward a series configuration if I go dual pump, but I’d appreciate any insights from someone who supports parallel operation.

N
nebel75
Junior Member
41
10-08-2016, 08:38 AM
#13
you can observe the process directly since I have completed all the settings being discussed here. here are the relevant links: my final setup, which performed best, is a complete parallel loop—components and radiators are paired together. However, as rubix has emphasized, the actual loop remains serial from components to radiators. i plan to examine this thoroughly. There’s a lot of information there, so it will take some time to go through it. Thanks for the effort and the resources you provided, Toolmaker.
N
nebel75
10-08-2016, 08:38 AM #13

you can observe the process directly since I have completed all the settings being discussed here. here are the relevant links: my final setup, which performed best, is a complete parallel loop—components and radiators are paired together. However, as rubix has emphasized, the actual loop remains serial from components to radiators. i plan to examine this thoroughly. There’s a lot of information there, so it will take some time to go through it. Thanks for the effort and the resources you provided, Toolmaker.

T
trixie2303
Junior Member
36
10-08-2016, 02:42 PM
#14
I only bring this up since we frequently focus on whether a loop is serial or parallel, but in reality, a perfect parallel loop doesn't exist because it remains a serial setup at some level. You're still limited by the overall flow and constraints of the circuit, which define the lowest common denominator.
T
trixie2303
10-08-2016, 02:42 PM #14

I only bring this up since we frequently focus on whether a loop is serial or parallel, but in reality, a perfect parallel loop doesn't exist because it remains a serial setup at some level. You're still limited by the overall flow and constraints of the circuit, which define the lowest common denominator.

R
Robang592
Senior Member
368
10-09-2016, 01:52 AM
#15
I only mention this because we frequently become overly focused on whether a loop is serial or parallel, but in reality, a perfect parallel loop doesn’t exist. At some stage, it remains a serial setup, constrained by the overall flow and limitations of that part. Rubix - that’s accurate. I believe the GPU/CPU and radiator section should all run in series to maintain proper flow through the blocks. (I’m still curious about a parallel CPU/GPU loop) I’d like to determine the pump arrangement. Maybe removing one pump, as Ryan suggested, would help clarify things. Toolmaker’s data is useful, but I haven’t fully processed it yet.
R
Robang592
10-09-2016, 01:52 AM #15

I only mention this because we frequently become overly focused on whether a loop is serial or parallel, but in reality, a perfect parallel loop doesn’t exist. At some stage, it remains a serial setup, constrained by the overall flow and limitations of that part. Rubix - that’s accurate. I believe the GPU/CPU and radiator section should all run in series to maintain proper flow through the blocks. (I’m still curious about a parallel CPU/GPU loop) I’d like to determine the pump arrangement. Maybe removing one pump, as Ryan suggested, would help clarify things. Toolmaker’s data is useful, but I haven’t fully processed it yet.

R
razorcrafta
Member
164
10-11-2016, 05:05 AM
#16
Ryan here - I'm going to follow your advice since your reasoning makes sense. Also, I wasn't satisfied with the dual pump/res mock-up; it looked too messy. I changed the design so a single pump/res can be mounted, then disassembled the case for powder coating. I ended up blasting and coating all the tiny components last night (that Corsair paint is really hard to blast through).

Thanks to everyone for your feedback and insights.
R
razorcrafta
10-11-2016, 05:05 AM #16

Ryan here - I'm going to follow your advice since your reasoning makes sense. Also, I wasn't satisfied with the dual pump/res mock-up; it looked too messy. I changed the design so a single pump/res can be mounted, then disassembled the case for powder coating. I ended up blasting and coating all the tiny components last night (that Corsair paint is really hard to blast through).

Thanks to everyone for your feedback and insights.

F
fryzer33
Junior Member
15
10-11-2016, 06:46 AM
#17
Thanks Ryan - I'm going to follow your advice because your reasoning makes sense. Also, the dual pump/res mock-up didn't meet my expectations; it was too messy. I redesigned it so a single pump/res can be mounted, then disassembled the case for powder coating. I ended up blasting and coating all the small parts last night (that Corsair paint is really hard to blast through). Thanks to everyone for your feedback and insights.
F
fryzer33
10-11-2016, 06:46 AM #17

Thanks Ryan - I'm going to follow your advice because your reasoning makes sense. Also, the dual pump/res mock-up didn't meet my expectations; it was too messy. I redesigned it so a single pump/res can be mounted, then disassembled the case for powder coating. I ended up blasting and coating all the small parts last night (that Corsair paint is really hard to blast through). Thanks to everyone for your feedback and insights.

T
TheDarkChest
Junior Member
6
10-19-2016, 11:35 AM
#18
Ryan6 thanks Ryan for the advice, noting the logic is solid. They mention the dual pump/res mock-up was too messy and decided to redesign with a single pump/res mounting, then removed the case for powder coating. They spent last night blasting and coating all small parts, especially since the corsair paint proved difficult to blast through. They ask for updates and request photos once completed.
T
TheDarkChest
10-19-2016, 11:35 AM #18

Ryan6 thanks Ryan for the advice, noting the logic is solid. They mention the dual pump/res mock-up was too messy and decided to redesign with a single pump/res mounting, then removed the case for powder coating. They spent last night blasting and coating all small parts, especially since the corsair paint proved difficult to blast through. They ask for updates and request photos once completed.

O
Onghoe
Junior Member
21
10-19-2016, 02:11 PM
#19
I completely grasp why my work took longer than expected. I didn’t want to spend extra time capturing photos while the process was ongoing, since that time was essential for completion. Additionally, some of my efforts were uncertain about their success.
O
Onghoe
10-19-2016, 02:11 PM #19

I completely grasp why my work took longer than expected. I didn’t want to spend extra time capturing photos while the process was ongoing, since that time was essential for completion. Additionally, some of my efforts were uncertain about their success.

S
seomilk77
Member
181
10-19-2016, 06:25 PM
#20
I completely get why some parts took longer. I didn’t want to spend extra time taking photos while working, since that time was important for finishing the task. Plus, a lot of what I did wasn’t 100% certain it would work out.
Agreed! 😀
S
seomilk77
10-19-2016, 06:25 PM #20

I completely get why some parts took longer. I didn’t want to spend extra time taking photos while working, since that time was important for finishing the task. Plus, a lot of what I did wasn’t 100% certain it would work out.
Agreed! 😀

Pages (2): Previous 1 2