Overclock.net › Forums › Cooling › Water Cooling › Thoughts on flow and temp changes throughout the loop
New Posts  All Forums:Forum Nav:

Thoughts on flow and temp changes throughout the loop

post #1 of 9
Thread Starter 
I've been thinking about this a lot lately and I think I know the answer but I want to bounce it off the people who know before I continue.

Throughout the course of water going through your loop, does the flow rate actually increase/decrease between various points in the loop? Let's say I have the following loop configuration:

Res > Pump > MOBO1 > CPU > MOBO2 > GPU > GPU > Rad > Rad

Is my flow going to be much different after the last radiator than after the CPU or first GPU? My guess is that my collective flow rate will be determined by the most restrictive component in the loop.

The same question for temps. Now I realize that there will be a difference in temps from the time the water leaves my pump until it enters the first radiator. But after a while wouldn't the water temperature basically level out?

The answers to these will help me better understand where to put my flow and temp sensors. Right now I have one of each but I thought that maybe it would be beneficial to have 2 temp sensors; basically one in the res or coming out of the pump and another just before it enters the first radiator. Still not 100% sure where I want to put any of them but I thought the answers to the above questions may help.

Thanks!
Fusion Reactor
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel 3930K @ 4.8GHz Asus Rampage IV Extreme 2 X EVGA GTX 670 FTW 32GB Mushkin Redline 2133 
Hard DriveCoolingCoolingCooling
Plextor M3 XSPC Raystorm Koolance Motherboard Blocks 2 X HEATKILLER GTX 680 Hole Edition Water Block... 
CoolingCoolingCoolingMonitor
Swiftech MCP35X2 Water Pump 2 X Alphacool NexXxoS Monsta Quad 120mm Radiators 32 X Gentle Typhoon AP-15 1850 RPM  HP w2408 
Case
Caselabs TH10 
  hide details  
Reply
Fusion Reactor
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel 3930K @ 4.8GHz Asus Rampage IV Extreme 2 X EVGA GTX 670 FTW 32GB Mushkin Redline 2133 
Hard DriveCoolingCoolingCooling
Plextor M3 XSPC Raystorm Koolance Motherboard Blocks 2 X HEATKILLER GTX 680 Hole Edition Water Block... 
CoolingCoolingCoolingMonitor
Swiftech MCP35X2 Water Pump 2 X Alphacool NexXxoS Monsta Quad 120mm Radiators 32 X Gentle Typhoon AP-15 1850 RPM  HP w2408 
Case
Caselabs TH10 
  hide details  
Reply
post #2 of 9
Quote:
Originally Posted by ColdFusionWi View Post

I've been thinking about this a lot lately and I think I know the answer but I want to bounce it off the people who know before I continue.
Throughout the course of water going through your loop, does the flow rate actually increase/decrease between various points in the loop? Let's say I have the following loop configuration:
Res > Pump > MOBO1 > CPU > MOBO2 > GPU > GPU > Rad > Rad
Is my flow going to be much different after the last radiator than after the CPU or first GPU? My guess is that my collective flow rate will be determined by the most restrictive component in the loop.

this is mostly true... the flow will change very miniscuoley throughout the loop, but you probably wont even be able to measure it (what does fluctuate is the pressure though)
Quote:
Originally Posted by ColdFusionWi View Post

The same question for temps. Now I realize that there will be a difference in temps from the time the water leaves my pump until it enters the first radiator. But after a while wouldn't the water temperature basically level out?
The answers to these will help me better understand where to put my flow and temp sensors. Right now I have one of each but I thought that maybe it would be beneficial to have 2 temp sensors; basically one in the res or coming out of the pump and another just before it enters the first radiator. Still not 100% sure where I want to put any of them but I thought the answers to the above questions may help.
Thanks!

this is a very big debate currently... what it would seem the answer has come to, is that loop order does matter, but will only provide very small returns without massive heat dumps... the best way for you to set up your loop, would be with the least amount of tubing and fittings as possible... theoreticaly, a RAD between every block will give better MAX temps and cooling of each block, but in reality it is a very small improvement... that said, if you had 4x GTX480s (probably not even 680s), then set up would definitely matter because of the absurd heat dump
Big Bang Theory
(15 items)
 
old meets new
(17 items)
 
 
CPUMotherboardGraphicsGraphics
3930k @ 4.1 GHz Big Bang Xpower II HIS 7970 ICE X2 HIS 7970 ICE X2 
RAMHard DriveHard DriveCooling
G.Skill Ripjaw Z 64GB @ 1833 MHz Vertex 4 512GB Vertex 4 512GB H80i 
OSMonitorMonitorPower
Windows 7 Ultimate Auria 27" (2560x1440) Auria 27" (2560x1440) Lepa G. 1600 
CaseMouseAudio
HAF X 942 Corsair M90 Senheiser S170 
CPUMotherboardGraphicsGraphics
Q6600 EP45-UD3P Sapphire 5870 ASUS 5850-Top Direct CU 
RAMHard DriveCoolingOS
DDR2 Dominators 1x 80GB samsung H70 XP Pro 
MonitorMonitorMonitorKeyboard
Dell 22" (1680x1050) Sony 17" (800x900) Auria 27" (2560x1440) Wolf king 
PowerCase
Ultra X4 1600W CM HAF X 
  hide details  
Reply
Big Bang Theory
(15 items)
 
old meets new
(17 items)
 
 
CPUMotherboardGraphicsGraphics
3930k @ 4.1 GHz Big Bang Xpower II HIS 7970 ICE X2 HIS 7970 ICE X2 
RAMHard DriveHard DriveCooling
G.Skill Ripjaw Z 64GB @ 1833 MHz Vertex 4 512GB Vertex 4 512GB H80i 
OSMonitorMonitorPower
Windows 7 Ultimate Auria 27" (2560x1440) Auria 27" (2560x1440) Lepa G. 1600 
CaseMouseAudio
HAF X 942 Corsair M90 Senheiser S170 
CPUMotherboardGraphicsGraphics
Q6600 EP45-UD3P Sapphire 5870 ASUS 5850-Top Direct CU 
RAMHard DriveCoolingOS
DDR2 Dominators 1x 80GB samsung H70 XP Pro 
MonitorMonitorMonitorKeyboard
Dell 22" (1680x1050) Sony 17" (800x900) Auria 27" (2560x1440) Wolf king 
PowerCase
Ultra X4 1600W CM HAF X 
  hide details  
Reply
post #3 of 9
Thread Starter 
Thanks. I guess I'm really not too concerned about the order of components in the loop. I know there are going to be differences and the amount will vary depending on load and what the component being cooled is. For example the two GPUs in question are 670 FTWs. I know they don't give off a ton of heat but they will spend most of their life folding so I will see a bit of a heat dump after them in my loop thus they are last.

I'm thinking about putting the flow sensor on the return line to my res; it will be hidden and most likely will display the "lowest" number. As long as that lowest number is at or above 1gpm I should be good. Still not sure where to put the temp sensors to get the most useful information though. When everyone is comparing the delta between their water temp and ambient temps where are they measuring? Is there a benefit to measuring water temps at multiple locations?
Fusion Reactor
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel 3930K @ 4.8GHz Asus Rampage IV Extreme 2 X EVGA GTX 670 FTW 32GB Mushkin Redline 2133 
Hard DriveCoolingCoolingCooling
Plextor M3 XSPC Raystorm Koolance Motherboard Blocks 2 X HEATKILLER GTX 680 Hole Edition Water Block... 
CoolingCoolingCoolingMonitor
Swiftech MCP35X2 Water Pump 2 X Alphacool NexXxoS Monsta Quad 120mm Radiators 32 X Gentle Typhoon AP-15 1850 RPM  HP w2408 
Case
Caselabs TH10 
  hide details  
Reply
Fusion Reactor
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel 3930K @ 4.8GHz Asus Rampage IV Extreme 2 X EVGA GTX 670 FTW 32GB Mushkin Redline 2133 
Hard DriveCoolingCoolingCooling
Plextor M3 XSPC Raystorm Koolance Motherboard Blocks 2 X HEATKILLER GTX 680 Hole Edition Water Block... 
CoolingCoolingCoolingMonitor
Swiftech MCP35X2 Water Pump 2 X Alphacool NexXxoS Monsta Quad 120mm Radiators 32 X Gentle Typhoon AP-15 1850 RPM  HP w2408 
Case
Caselabs TH10 
  hide details  
Reply
post #4 of 9
The mass flow rate (i.e. the amount of water passing any point) is constant throughout the loop as water is incompressible. The velocity of the flow will change as the cross section area in the loop changes.
This flow rate will be determined by flow rate of the pump given the total pressure drop across all components of the loop, which is most likely dominated by the one or two most restrictive blocks in the loop,
Order doesn't matter as long as your flow rate is adequate (1 GPM) and/or you have a block (or series of blocks) dumping kW into the loop.
 
HTPC
(8 items)
 
 
CPUMotherboardGraphicsRAM
i7-3930k MSI X79MA-GD45 Powercolor 7950 G Skill 
Hard DriveCoolingCoolingCooling
Samsung 830 Swiftech Apogee XT Rev. 1 EK-FC HWLabs Black Ice GTX 240 and a Koolance HX-CU72... 
CoolingCoolingOSOS
MCP 35X and a MCP 350 Koolance 401x2 Opensuse 12.1 Windows 7 
KeyboardPowerCase
Ducky OCN with cherry blues Antec True Power New 650W Cooler Master CM690 II 
CPUMotherboardGraphicsRAM
i5-750 Gigabyte p55-ud2 GTX 430 G Skill 
Hard DriveOSPowerCase
Samsung HD204UI Fedora 15 Corsair CX-430 Silverstone GC04 
  hide details  
Reply
 
HTPC
(8 items)
 
 
CPUMotherboardGraphicsRAM
i7-3930k MSI X79MA-GD45 Powercolor 7950 G Skill 
Hard DriveCoolingCoolingCooling
Samsung 830 Swiftech Apogee XT Rev. 1 EK-FC HWLabs Black Ice GTX 240 and a Koolance HX-CU72... 
CoolingCoolingOSOS
MCP 35X and a MCP 350 Koolance 401x2 Opensuse 12.1 Windows 7 
KeyboardPowerCase
Ducky OCN with cherry blues Antec True Power New 650W Cooler Master CM690 II 
CPUMotherboardGraphicsRAM
i5-750 Gigabyte p55-ud2 GTX 430 G Skill 
Hard DriveOSPowerCase
Samsung HD204UI Fedora 15 Corsair CX-430 Silverstone GC04 
  hide details  
Reply
post #5 of 9
Quote:
Originally Posted by ColdFusionWi View Post

Thanks. I guess I'm really not too concerned about the order of components in the loop. I know there are going to be differences and the amount will vary depending on load and what the component being cooled is. For example the two GPUs in question are 670 FTWs. I know they don't give off a ton of heat but they will spend most of their life folding so I will see a bit of a heat dump after them in my loop thus they are last.
I'm thinking about putting the flow sensor on the return line to my res; it will be hidden and most likely will display the "lowest" number. As long as that lowest number is at or above 1gpm I should be good. Still not sure where to put the temp sensors to get the most useful information though. When everyone is comparing the delta between their water temp and ambient temps where are they measuring? Is there a benefit to measuring water temps at multiple locations?

i wouldnt realy bother measuring water temps if im honest... keep a flow meter and temps of teh actual components... you would also benefit flow wise from having your GPUs in parallel... the only thing i would not do setup wise is go GPUs>CPU... the heat outfut of your cpu wont change the water temp much, so the other way isnt an issue smile.gif and i like 1-2GPM (or about .5-.75 GPM per card smile.gif
Big Bang Theory
(15 items)
 
old meets new
(17 items)
 
 
CPUMotherboardGraphicsGraphics
3930k @ 4.1 GHz Big Bang Xpower II HIS 7970 ICE X2 HIS 7970 ICE X2 
RAMHard DriveHard DriveCooling
G.Skill Ripjaw Z 64GB @ 1833 MHz Vertex 4 512GB Vertex 4 512GB H80i 
OSMonitorMonitorPower
Windows 7 Ultimate Auria 27" (2560x1440) Auria 27" (2560x1440) Lepa G. 1600 
CaseMouseAudio
HAF X 942 Corsair M90 Senheiser S170 
CPUMotherboardGraphicsGraphics
Q6600 EP45-UD3P Sapphire 5870 ASUS 5850-Top Direct CU 
RAMHard DriveCoolingOS
DDR2 Dominators 1x 80GB samsung H70 XP Pro 
MonitorMonitorMonitorKeyboard
Dell 22" (1680x1050) Sony 17" (800x900) Auria 27" (2560x1440) Wolf king 
PowerCase
Ultra X4 1600W CM HAF X 
  hide details  
Reply
Big Bang Theory
(15 items)
 
old meets new
(17 items)
 
 
CPUMotherboardGraphicsGraphics
3930k @ 4.1 GHz Big Bang Xpower II HIS 7970 ICE X2 HIS 7970 ICE X2 
RAMHard DriveHard DriveCooling
G.Skill Ripjaw Z 64GB @ 1833 MHz Vertex 4 512GB Vertex 4 512GB H80i 
OSMonitorMonitorPower
Windows 7 Ultimate Auria 27" (2560x1440) Auria 27" (2560x1440) Lepa G. 1600 
CaseMouseAudio
HAF X 942 Corsair M90 Senheiser S170 
CPUMotherboardGraphicsGraphics
Q6600 EP45-UD3P Sapphire 5870 ASUS 5850-Top Direct CU 
RAMHard DriveCoolingOS
DDR2 Dominators 1x 80GB samsung H70 XP Pro 
MonitorMonitorMonitorKeyboard
Dell 22" (1680x1050) Sony 17" (800x900) Auria 27" (2560x1440) Wolf king 
PowerCase
Ultra X4 1600W CM HAF X 
  hide details  
Reply
post #6 of 9
Quote:
Originally Posted by eskamobob1 View Post

i wouldnt realy bother measuring water temps if im honest... keep a flow meter and temps of teh actual components... you would also benefit flow wise from having your GPUs in parallel... the only thing i would not do setup wise is go GPUs>CPU... the heat outfut of your cpu wont change the water temp much, so the other way isnt an issue smile.gif and i like 1-2GPM (or about .5-.75 GPM per card smile.gif

Meh, I saw a 3c increase in CPU temps by going from CPU>GPU to GPU>CPU.

Here's an easy one to remember, 264w of heat @ 1gpm bring water up 1c (delta from entry and exit). So even with two GPU's you're looking at a 2c increase in water temps arriving to the CPU.
Synthetic Anatomy
(14 items)
 
NAT-inator
(11 items)
 
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Z68X-UD7-B3 NVIDIA GeForce GTX 580 NVIDIA GeForce GTX 580 
GraphicsRAMRAMRAM
NVIDIA GeForce GTX 580 Corsair  Corsair  Corsair  
RAMHard DriveCoolingPower
Corsair  2x Mushkin 240Gb SSD Raid 0  A lot. ST1500 
CaseAudio
FT02 Xonar ST 
CPUMotherboardGraphicsRAM
i5-2500K P8P67 WS Revolution Sparkle Calibre GTX 560 G.Skill 
Hard DriveHard DriveOptical DriveCooling
Corsair Force SSD WD Caviar Black BD combo Liquid 
OSPowerCase
Win 7 Ultimate AX650 500R 
  hide details  
Reply
Synthetic Anatomy
(14 items)
 
NAT-inator
(11 items)
 
 
CPUMotherboardGraphicsGraphics
Intel Core i7 2700K Z68X-UD7-B3 NVIDIA GeForce GTX 580 NVIDIA GeForce GTX 580 
GraphicsRAMRAMRAM
NVIDIA GeForce GTX 580 Corsair  Corsair  Corsair  
RAMHard DriveCoolingPower
Corsair  2x Mushkin 240Gb SSD Raid 0  A lot. ST1500 
CaseAudio
FT02 Xonar ST 
CPUMotherboardGraphicsRAM
i5-2500K P8P67 WS Revolution Sparkle Calibre GTX 560 G.Skill 
Hard DriveHard DriveOptical DriveCooling
Corsair Force SSD WD Caviar Black BD combo Liquid 
OSPowerCase
Win 7 Ultimate AX650 500R 
  hide details  
Reply
post #7 of 9
Quote:
Originally Posted by mdatmo View Post

The mass flow rate (i.e. the amount of water passing any point) is constant throughout the loop as water is incompressible. The velocity of the flow will change as the cross section area in the loop changes.
This flow rate will be determined by flow rate of the pump given the total pressure drop across all components of the loop, which is most likely dominated by the one or two most restrictive blocks in the loop,
Order doesn't matter as long as your flow rate is adequate (1 GPM) and/or you have a block (or series of blocks) dumping kW into the loop.

^This. Mass flow rate stays the same, velocity increases and pressure decreases as area gets smaller and vice versa
Main Rig
(15 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 4820K Asus Rampage 4 Black EVGA GTX 1080Ti GSkill Trident X 2400 
Hard DriveHard DriveCoolingOS
Western Digital Black 1TB Smasung 850 evo  Custom loop windows 10 
MonitorKeyboardPowerCase
Asus PB278Q Razer Blackwidow 2013 Seasonic X1250 Caselabs SMA8 
MouseMouse PadAudio
Razer Death Adder Chroma Razer gigantus AKG K7XX headphones 
  hide details  
Reply
Main Rig
(15 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 4820K Asus Rampage 4 Black EVGA GTX 1080Ti GSkill Trident X 2400 
Hard DriveHard DriveCoolingOS
Western Digital Black 1TB Smasung 850 evo  Custom loop windows 10 
MonitorKeyboardPowerCase
Asus PB278Q Razer Blackwidow 2013 Seasonic X1250 Caselabs SMA8 
MouseMouse PadAudio
Razer Death Adder Chroma Razer gigantus AKG K7XX headphones 
  hide details  
Reply
post #8 of 9
Thread Starter 
Quote:
Originally Posted by mdatmo View Post

The mass flow rate (i.e. the amount of water passing any point) is constant throughout the loop as water is incompressible. The velocity of the flow will change as the cross section area in the loop changes.
This flow rate will be determined by flow rate of the pump given the total pressure drop across all components of the loop, which is most likely dominated by the one or two most restrictive blocks in the loop,
Order doesn't matter as long as your flow rate is adequate (1 GPM) and/or you have a block (or series of blocks) dumping kW into the loop.

This is what I was looking for when it comes to flow. Thank you.
Fusion Reactor
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel 3930K @ 4.8GHz Asus Rampage IV Extreme 2 X EVGA GTX 670 FTW 32GB Mushkin Redline 2133 
Hard DriveCoolingCoolingCooling
Plextor M3 XSPC Raystorm Koolance Motherboard Blocks 2 X HEATKILLER GTX 680 Hole Edition Water Block... 
CoolingCoolingCoolingMonitor
Swiftech MCP35X2 Water Pump 2 X Alphacool NexXxoS Monsta Quad 120mm Radiators 32 X Gentle Typhoon AP-15 1850 RPM  HP w2408 
Case
Caselabs TH10 
  hide details  
Reply
Fusion Reactor
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel 3930K @ 4.8GHz Asus Rampage IV Extreme 2 X EVGA GTX 670 FTW 32GB Mushkin Redline 2133 
Hard DriveCoolingCoolingCooling
Plextor M3 XSPC Raystorm Koolance Motherboard Blocks 2 X HEATKILLER GTX 680 Hole Edition Water Block... 
CoolingCoolingCoolingMonitor
Swiftech MCP35X2 Water Pump 2 X Alphacool NexXxoS Monsta Quad 120mm Radiators 32 X Gentle Typhoon AP-15 1850 RPM  HP w2408 
Case
Caselabs TH10 
  hide details  
Reply
post #9 of 9
Quote:
Originally Posted by eskamobob1 View Post

i wouldnt realy bother measuring water temps if im honest... keep a flow meter and temps of teh actual components... you would also benefit flow wise from having your GPUs in parallel... the only thing i would not do setup wise is go GPUs>CPU... the heat outfut of your cpu wont change the water temp much, so the other way isnt an issue smile.gif and i like 1-2GPM (or about .5-.75 GPM per card smile.gif

I think it's worth measuring water temp in one place-- gives you a sense for how well your rad's performing, helps you troubleshoot problems if your temps are high, helps you decide how to control your fans, helps you figure out what to upgrade next when you get that itch (i.e. can my rads handle another block?)...

Not much point in having more than one sensor though. (Take it from a guy with 4... )
Kathis
(16 items)
 
  
CPUMotherboardGraphicsRAM
2600k Maximus V Formula EVGA GTX 670 Kingston something 
Hard DriveHard DriveCoolingOS
OCZ Vertex 4 ~ 4 terabytes across 3 drives, not in RAID. 2 ... Water cooled, aquaero 5 LT controller/temp monitor Win7 x64 
MonitorMonitorKeyboardPower
Acer 23.3", 1920x1080 Acer 19", 1440x900 CM Storm Quickfire 700W Cooler Master Silent Pro 
CaseMouseMouse PadAudio
NXZT Switch 810 Logitech G9X Yup.  E-MU 1820m running into a Pyle 120wx2 amp, driv... 
  hide details  
Reply
Kathis
(16 items)
 
  
CPUMotherboardGraphicsRAM
2600k Maximus V Formula EVGA GTX 670 Kingston something 
Hard DriveHard DriveCoolingOS
OCZ Vertex 4 ~ 4 terabytes across 3 drives, not in RAID. 2 ... Water cooled, aquaero 5 LT controller/temp monitor Win7 x64 
MonitorMonitorKeyboardPower
Acer 23.3", 1920x1080 Acer 19", 1440x900 CM Storm Quickfire 700W Cooler Master Silent Pro 
CaseMouseMouse PadAudio
NXZT Switch 810 Logitech G9X Yup.  E-MU 1820m running into a Pyle 120wx2 amp, driv... 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Thoughts on flow and temp changes throughout the loop