Overclock.net › Forums › Cooling › Water Cooling › Reservoirs are 2X as restrictive as T-lines
New Posts  All Forums:Forum Nav:

Reservoirs are 2X as restrictive as T-lines

post #1 of 13
Thread Starter 
That's 2X more than almost nothing....hehe. I had this idea in my head that reservoirs were some horrible flow eating waste of space, well they are more restrictive, but still don't amount to a whole heck of alot.

Decided to do a pressure drop test on my fancy dancy official "Coke Bottle Reservoir" And no this wasn't a real reservoir, I gave my old one to Apollo and it got lost in customs somewhere, so this was just a simulation of what I thought would represent a tall style reservoir figuring it would be the worst case. I purposely left the 1/2" barbs fully extened into the sidewalls made it a nice tall reservoir where you almost have no movement at the top when running (it actually works pretty well, no whirlpools on this variety...Look out EK ). Next up was my 3 pack of elbows stacked in series, and finally the T-line your standard 1/2" variety.

I actually had to run 4 curves, one for the test setup and tubing (Pressure drop of the T-lines and 3' of 1/2" ID tubing, then the other three curves. I found the test to be fairly difficult because the pressure drop is so low.

But before I share the curve....

This test is brought to you by...my son's John Deer tractor..


And the resulting graphs:


Now for the reality of what this small pressure drop means:
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing = 1.68 GPM
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing + 1 T-line = 1.66 GPM
D-tek Fuzion, MCW60, MCR320, D5, 7' of 1/2" ID tubing + 1 Reservoir = 1.64 GPM

Bottom line...not enough to worry about, you can have your fancy new "Coke bottle reservoir" and not worry about extra flow loss.
    
CPUMotherboardGraphicsRAM
3930K ASrock X79 Ext7 Evga GTX570 CORSAIR Vengeance 16GB 
Hard DriveOptical DriveOSMonitor
Phoenix SSD+2TB Seagate drives ASUS DVD-RW combo Windows 7 28" Hanns-G HZ281 
KeyboardPowerCaseMouse
MS Natural Corsair TX650 Danger Den Torture Rack Logitech MX518 
Mouse Pad
Desk 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
3930K ASrock X79 Ext7 Evga GTX570 CORSAIR Vengeance 16GB 
Hard DriveOptical DriveOSMonitor
Phoenix SSD+2TB Seagate drives ASUS DVD-RW combo Windows 7 28" Hanns-G HZ281 
KeyboardPowerCaseMouse
MS Natural Corsair TX650 Danger Den Torture Rack Logitech MX518 
Mouse Pad
Desk 
  hide details  
Reply
post #2 of 13
hmmm.... good stuff. i still like my res mounted on the top of my case though, makes for easy filling, and air bleeding.... rep for you!
GiNNz!'Z Rig!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Q6600 @4.02 Ghz Asus P5Q-Turbo BFG GTX 295 OCZ 4GB DDR2 Platinum 800 Mghz 
Hard DriveOptical DriveOSMonitor
2- WD 1000 gig sata II /750 gig external LG 16X lightscribe Windows 7 3 - 19" Dell LCD's/Matrox Triple head2go 
KeyboardPowerCaseMouse
Logitech G-15 Corsair 750 W Cooler Master HAF932 Logicteh Wireless 
Mouse Pad
foam! 
  hide details  
Reply
GiNNz!'Z Rig!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Q6600 @4.02 Ghz Asus P5Q-Turbo BFG GTX 295 OCZ 4GB DDR2 Platinum 800 Mghz 
Hard DriveOptical DriveOSMonitor
2- WD 1000 gig sata II /750 gig external LG 16X lightscribe Windows 7 3 - 19" Dell LCD's/Matrox Triple head2go 
KeyboardPowerCaseMouse
Logitech G-15 Corsair 750 W Cooler Master HAF932 Logicteh Wireless 
Mouse Pad
foam! 
  hide details  
Reply
post #3 of 13
But how does the drop translate into a full blown loop, with a radiator and a couple of blocks? I would think that it would show an increase in temps over the T-line...no?
post #4 of 13
Thread Starter 
Quote:
Originally Posted by CL3P20 View Post
But how does the drop translate into a full blown loop, with a radiator and a couple of blocks? I would think that it would show an increase in temps over the T-line...no?
Not anything you could measure, one thing I've been learning about all of this is how well a system can handle reduced flow rates. While more is always better, most blocks and rads don't really start suffering until you get down under 1 GPM.

The difference in temp is really really small even with a change of .5 GPM, so .02 GPM is nothing at all. As long as you're keeping flow rates up in the 1.5 range you're plenty good.
    
CPUMotherboardGraphicsRAM
3930K ASrock X79 Ext7 Evga GTX570 CORSAIR Vengeance 16GB 
Hard DriveOptical DriveOSMonitor
Phoenix SSD+2TB Seagate drives ASUS DVD-RW combo Windows 7 28" Hanns-G HZ281 
KeyboardPowerCaseMouse
MS Natural Corsair TX650 Danger Den Torture Rack Logitech MX518 
Mouse Pad
Desk 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
3930K ASrock X79 Ext7 Evga GTX570 CORSAIR Vengeance 16GB 
Hard DriveOptical DriveOSMonitor
Phoenix SSD+2TB Seagate drives ASUS DVD-RW combo Windows 7 28" Hanns-G HZ281 
KeyboardPowerCaseMouse
MS Natural Corsair TX650 Danger Den Torture Rack Logitech MX518 
Mouse Pad
Desk 
  hide details  
Reply
post #5 of 13
Quote:
Originally Posted by Martinm210 View Post
Not anything you could measure, one thing I've been learning about all of this is how well a system can handle reduced flow rates. While more is always better, most blocks and rads don't really start suffering until you get down under 1 GPM.

The difference in temp is really really small even with a change of .5 GPM, so .02 GPM is nothing at all. As long as you're keeping flow rates up in the 1.5 range you're plenty good.
Yup, looks like diminishing returns starts to really kick in around 1GPH from these tests. 1.5-2GPH should be the minimum flow rate you want to achieve.
http://www.overclockers.com/articles1163/
http://www.overclockers.com/articles1158/
http://www.overclockers.com/articles1140/
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
post #6 of 13
I wish I had a bay res I wasn't using so you could try one of them out...Just guessing but since they have a divider down the center of them they may keep a more positive pressure on the suction side....


Nice testing...Thanks....
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
E8600 4545Mhz*9 w/1.29V DFI DK P45 XFX 8800 GS D9GKX & D9GMH 
Hard DriveOptical DriveOSPower
2x Seagate 160 Perp. RD-1 2x Asus Vista Premium Corsair HX 620 
Case
Open Bench 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
E8600 4545Mhz*9 w/1.29V DFI DK P45 XFX 8800 GS D9GKX & D9GMH 
Hard DriveOptical DriveOSPower
2x Seagate 160 Perp. RD-1 2x Asus Vista Premium Corsair HX 620 
Case
Open Bench 
  hide details  
Reply
post #7 of 13
Thread Starter 
Quote:
Originally Posted by DuckieHo View Post
Yup, looks like diminishing returns starts to really kick in around 1GPH from these tests. 1.5-2GPH should be the minimum flow rate you want to achieve.
http://www.overclockers.com/articles1163/
http://www.overclockers.com/articles1158/
http://www.overclockers.com/articles1140/

Nice articles, and there seems to be some consistency in the difference of how flow rate affects radiators and waterblocks. It appears from what I'm seeing that radiators are more sensetive to flow rate than water blocks for thermal abilities. Seems like in general most radiator thermal curves start dropping off in the 1GPM to 1.5GPM range and water blocks don't drop off until you start getting under 1 GPM. This probably has something to do with the velocity differences and turbidity. Since waterblocks tend to be more restrictive...alot more restrictive in general, the velocity and mixing of water is going to be alot higher there particularly with nozzles and such. Radiators on the other hand are very open, so we're probably getting more laminar type flow and the thermal transfer just ends up dropping off sooner.

So I guess if you overbuild and oversize your radiator capacity your system as a whole will be more resilient to lower flow rates because the water block is still fairly efficient at those lower flow rates. But if you're pushing the capacity of your radiator, lowering flow rates will begin to have an effect earlier on the system.

I'm still pretty weak on understanding the thermal side of things, but after I get a better part of the components everyone wants in the flow rate estimator done, I'm going to start tinkering with tying that flow rate number to thermal curves. It seems like that's the end result were all after, and determining flow rate is only getting you half way home. I just know it's a bit of a monster when you start feeling out all the variables, but a challenge is always good
    
CPUMotherboardGraphicsRAM
3930K ASrock X79 Ext7 Evga GTX570 CORSAIR Vengeance 16GB 
Hard DriveOptical DriveOSMonitor
Phoenix SSD+2TB Seagate drives ASUS DVD-RW combo Windows 7 28" Hanns-G HZ281 
KeyboardPowerCaseMouse
MS Natural Corsair TX650 Danger Den Torture Rack Logitech MX518 
Mouse Pad
Desk 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
3930K ASrock X79 Ext7 Evga GTX570 CORSAIR Vengeance 16GB 
Hard DriveOptical DriveOSMonitor
Phoenix SSD+2TB Seagate drives ASUS DVD-RW combo Windows 7 28" Hanns-G HZ281 
KeyboardPowerCaseMouse
MS Natural Corsair TX650 Danger Den Torture Rack Logitech MX518 
Mouse Pad
Desk 
  hide details  
Reply
post #8 of 13
Martin there is one thing to remember, more flow will sometimes give you higher OC's even though the water temp doesn't drop much...There are microscopic hot spots on the die that the sheer volume of water passing over helps to cool and that will give you better OC's....I lost the links to them when my browser book-marks were corrupted a few weeks ago...Its some of Cathars work I'll try to find them again and I'll link you to them....It sucks I had some obscure links and its hard to find them all again...
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
E8600 4545Mhz*9 w/1.29V DFI DK P45 XFX 8800 GS D9GKX & D9GMH 
Hard DriveOptical DriveOSPower
2x Seagate 160 Perp. RD-1 2x Asus Vista Premium Corsair HX 620 
Case
Open Bench 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
E8600 4545Mhz*9 w/1.29V DFI DK P45 XFX 8800 GS D9GKX & D9GMH 
Hard DriveOptical DriveOSPower
2x Seagate 160 Perp. RD-1 2x Asus Vista Premium Corsair HX 620 
Case
Open Bench 
  hide details  
Reply
post #9 of 13
Thread Starter 
Quote:
Originally Posted by ira-k View Post
I wish I had a bay res I wasn't using so you could try one of them out...Just guessing but since they have a divider down the center of them they may keep a more positive pressure on the suction side....


Nice testing...Thanks....
Oh, I completely agree, a bay reservoir is going to be much less restrictive. not only do you have that divider that keeps the water moving, you also have those barb threads that are external so you don't get nearly the exit loss from a sharp protruding pipe. The simulation I made with the coke bottle is going to be more representative of the Microres or the tall EK reservoir style. I actually had a bay reservoir, but I shipped it to Apollo a while back and he said he never got it, so one of the customs guys is probably sporting my old res..

In culvert design they call this an entrance loss and it can be significant.
http://www.fsl.orst.edu/geowater/FX3...oefficient.htm
K value of projected (this would be like a barb sticking in from the sidewall) is .9.
K value flush = .5
K value beveled = .25
K value rounded = .2

So a flush barb with the reservoir sidewall is going to be almost half as restrictive as one that extened into it, and one that is beveled or tapered would be half of that. Of coarse this is smaller scale stuff, but the theory should be similar.

Although the friction losses are so small they don't really matter, if you did want to make your reservoir flow as absolutely freely as possible, I would cut the threading off so you don't have any extending into the tank, and better yet take a countersink bit and taper the sharp edge. You have enough pumping power and flow rate where the smaller losses start making a little bigger difference....

Same idea behind the D5 mod, and it helped there.
    
CPUMotherboardGraphicsRAM
3930K ASrock X79 Ext7 Evga GTX570 CORSAIR Vengeance 16GB 
Hard DriveOptical DriveOSMonitor
Phoenix SSD+2TB Seagate drives ASUS DVD-RW combo Windows 7 28" Hanns-G HZ281 
KeyboardPowerCaseMouse
MS Natural Corsair TX650 Danger Den Torture Rack Logitech MX518 
Mouse Pad
Desk 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
3930K ASrock X79 Ext7 Evga GTX570 CORSAIR Vengeance 16GB 
Hard DriveOptical DriveOSMonitor
Phoenix SSD+2TB Seagate drives ASUS DVD-RW combo Windows 7 28" Hanns-G HZ281 
KeyboardPowerCaseMouse
MS Natural Corsair TX650 Danger Den Torture Rack Logitech MX518 
Mouse Pad
Desk 
  hide details  
Reply
post #10 of 13
Yeah you want to for sure trim the threads on the barbs flush with the wall, that way you will get every bit of pressure pushing to the out-let...I haven't ever beveled the edges but it a good idea, every little bit counts...I'm glad you like doing all this its very helpful...The flow calc I still cant read......
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
E8600 4545Mhz*9 w/1.29V DFI DK P45 XFX 8800 GS D9GKX & D9GMH 
Hard DriveOptical DriveOSPower
2x Seagate 160 Perp. RD-1 2x Asus Vista Premium Corsair HX 620 
Case
Open Bench 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
E8600 4545Mhz*9 w/1.29V DFI DK P45 XFX 8800 GS D9GKX & D9GMH 
Hard DriveOptical DriveOSPower
2x Seagate 160 Perp. RD-1 2x Asus Vista Premium Corsair HX 620 
Case
Open Bench 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Reservoirs are 2X as restrictive as T-lines