Overclock.net › Forums › Cooling › Water Cooling › Water loop, never ending air pockets? SOLVED!
New Posts  All Forums:Forum Nav:

Water loop, never ending air pockets? SOLVED! - Page 3

post #21 of 68
Thread Starter 
More disappointment.

When I gutted the H60's pump, I also removed a little rubber piece that forced water into the micro channels on the cold plate. Thinking there would be no real need for it, because of the increased pressure and flow rate of the larger pump. Since nothing else was working, I though I might try putting that piece back to see if it help. And it didn't.

My temps are now higher than before and now the exhaust hose (that returns to the reservoir) seems to have lost quite a bit of pressure. As usual I'm seeing no bubbles at all, and I'm using the H60's original 120 radiator. Sitting here typing this, Its running at 40c with hardly any load.

Just in case anyone was wondering, each removal and installation of the water block includes cleaning the water-block and CPU with 91% alcohol and fresh AS5. Ambient temps are around 21c.

If anyone has any suggestions or things to try, please feel free to post them and I'll give it a shot. I'm open to all ideas at this point.
post #22 of 68
Sounds like you've sunk a lot of money into something that just isn't working. Just buying a rasa 120 kit would probably have cost you less, and got you better temps.
Ryzen 5 1600
(12 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 5 1600 MSI B350 Gaming Plus Gigabyte GeForce GTX 1060 WINDFORCE2 OC 16GB (2x8GB) Corsair DDR4 Vengeance LED, PC4-24... 
Hard DriveHard DriveCoolingOS
256GB Samsung PM961 Polaris M.2 NVMe  1TB Toshiba DT01ACA100 3.5" HDD, SATA III  Cooler Master Hyper 212 Evo Windows 10 64 
KeyboardPowerCaseMouse
Unicomp Model M 650W EVGA SuperNOVA G1, 80PLUS Gold, Full Modular Kolink Luminosity Cooler Master Reaper Aluminium 
  hide details  
Reply
Ryzen 5 1600
(12 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 5 1600 MSI B350 Gaming Plus Gigabyte GeForce GTX 1060 WINDFORCE2 OC 16GB (2x8GB) Corsair DDR4 Vengeance LED, PC4-24... 
Hard DriveHard DriveCoolingOS
256GB Samsung PM961 Polaris M.2 NVMe  1TB Toshiba DT01ACA100 3.5" HDD, SATA III  Cooler Master Hyper 212 Evo Windows 10 64 
KeyboardPowerCaseMouse
Unicomp Model M 650W EVGA SuperNOVA G1, 80PLUS Gold, Full Modular Kolink Luminosity Cooler Master Reaper Aluminium 
  hide details  
Reply
post #23 of 68
Did u try to let AS5 to cure? I think it has a rather long curing time(few days i think) compared to MX 4(none)... Why not use MX 4 and be done with it... If if doesnt work with another paste u tried everything with that parts... Than u can move to some other way of cooling ur CPU.
Zeus
(16 items)
 
  
CPUMotherboardGraphicsRAM
i5 2500k 5 Ghz @ 1.488 V 24/7 Asrock p67 extreme 4 B3 Gainward GTX 570 GS GLH @ 850/2100 @ 1.00V 2x4 Gb G:SKILL Sniper @ 1600 9-9-9-24 t1 @ 1.260 V 
Hard DriveOptical DriveCoolingCooling
RAID-2xWD Sata3 Blue 500 Gb + 1xWD Sata3 Green 2TB none RAD: EK XTC 420 CPU Block: EK-Supreme HF - Full Nickel-EN 
CoolingOSMonitorKeyboard
GPU Block: EK Universal EN Win7 Professional 64Bit LG Flatron M2362D 23"[1920x1080] Steelseries 6GV2 
PowerCaseMouseMouse Pad
XFX PRO 550W CoolerMaster CM690 II Nvidia Edition Razer DeathAdder 3500Dpi Left hand edition Razer Goliathus Speed 
  hide details  
Reply
Zeus
(16 items)
 
  
CPUMotherboardGraphicsRAM
i5 2500k 5 Ghz @ 1.488 V 24/7 Asrock p67 extreme 4 B3 Gainward GTX 570 GS GLH @ 850/2100 @ 1.00V 2x4 Gb G:SKILL Sniper @ 1600 9-9-9-24 t1 @ 1.260 V 
Hard DriveOptical DriveCoolingCooling
RAID-2xWD Sata3 Blue 500 Gb + 1xWD Sata3 Green 2TB none RAD: EK XTC 420 CPU Block: EK-Supreme HF - Full Nickel-EN 
CoolingOSMonitorKeyboard
GPU Block: EK Universal EN Win7 Professional 64Bit LG Flatron M2362D 23"[1920x1080] Steelseries 6GV2 
PowerCaseMouseMouse Pad
XFX PRO 550W CoolerMaster CM690 II Nvidia Edition Razer DeathAdder 3500Dpi Left hand edition Razer Goliathus Speed 
  hide details  
Reply
post #24 of 68
Quote:
Originally Posted by Rops84 View Post

Did u try to let AS5 to cure? I think it has a rather long curing time(few days i think) compared to MX 4(none)... Why not use MX 4 and be done with it... If if doesnt work with another paste u tried everything with that parts... Than u can move to some other way of cooling ur CPU.

AS5 does have a longer curing time, but a few runs with IBT should deal with that, and post curing it's only a couple of degrees better than when freshly applied. Plus, please try to use proper English/American in your posts, thanks.
Ryzen 5 1600
(12 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 5 1600 MSI B350 Gaming Plus Gigabyte GeForce GTX 1060 WINDFORCE2 OC 16GB (2x8GB) Corsair DDR4 Vengeance LED, PC4-24... 
Hard DriveHard DriveCoolingOS
256GB Samsung PM961 Polaris M.2 NVMe  1TB Toshiba DT01ACA100 3.5" HDD, SATA III  Cooler Master Hyper 212 Evo Windows 10 64 
KeyboardPowerCaseMouse
Unicomp Model M 650W EVGA SuperNOVA G1, 80PLUS Gold, Full Modular Kolink Luminosity Cooler Master Reaper Aluminium 
  hide details  
Reply
Ryzen 5 1600
(12 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 5 1600 MSI B350 Gaming Plus Gigabyte GeForce GTX 1060 WINDFORCE2 OC 16GB (2x8GB) Corsair DDR4 Vengeance LED, PC4-24... 
Hard DriveHard DriveCoolingOS
256GB Samsung PM961 Polaris M.2 NVMe  1TB Toshiba DT01ACA100 3.5" HDD, SATA III  Cooler Master Hyper 212 Evo Windows 10 64 
KeyboardPowerCaseMouse
Unicomp Model M 650W EVGA SuperNOVA G1, 80PLUS Gold, Full Modular Kolink Luminosity Cooler Master Reaper Aluminium 
  hide details  
Reply
post #25 of 68
Quote:
Originally Posted by allikat View Post

AS5 does have a longer curing time, but a few runs with IBT should deal with that, and post curing it's only a couple of degrees better than when freshly applied. Plus, please try to use proper English/American in your posts, thanks.

Lets not turn this grammar school, im just trying to help him solve a problem... And btw. my posts might not be in perfect English/American but at least im trying to be helpful.

Also i havent seen any rule anywhere on these forums that say that my English needs to be perfect.
Zeus
(16 items)
 
  
CPUMotherboardGraphicsRAM
i5 2500k 5 Ghz @ 1.488 V 24/7 Asrock p67 extreme 4 B3 Gainward GTX 570 GS GLH @ 850/2100 @ 1.00V 2x4 Gb G:SKILL Sniper @ 1600 9-9-9-24 t1 @ 1.260 V 
Hard DriveOptical DriveCoolingCooling
RAID-2xWD Sata3 Blue 500 Gb + 1xWD Sata3 Green 2TB none RAD: EK XTC 420 CPU Block: EK-Supreme HF - Full Nickel-EN 
CoolingOSMonitorKeyboard
GPU Block: EK Universal EN Win7 Professional 64Bit LG Flatron M2362D 23"[1920x1080] Steelseries 6GV2 
PowerCaseMouseMouse Pad
XFX PRO 550W CoolerMaster CM690 II Nvidia Edition Razer DeathAdder 3500Dpi Left hand edition Razer Goliathus Speed 
  hide details  
Reply
Zeus
(16 items)
 
  
CPUMotherboardGraphicsRAM
i5 2500k 5 Ghz @ 1.488 V 24/7 Asrock p67 extreme 4 B3 Gainward GTX 570 GS GLH @ 850/2100 @ 1.00V 2x4 Gb G:SKILL Sniper @ 1600 9-9-9-24 t1 @ 1.260 V 
Hard DriveOptical DriveCoolingCooling
RAID-2xWD Sata3 Blue 500 Gb + 1xWD Sata3 Green 2TB none RAD: EK XTC 420 CPU Block: EK-Supreme HF - Full Nickel-EN 
CoolingOSMonitorKeyboard
GPU Block: EK Universal EN Win7 Professional 64Bit LG Flatron M2362D 23"[1920x1080] Steelseries 6GV2 
PowerCaseMouseMouse Pad
XFX PRO 550W CoolerMaster CM690 II Nvidia Edition Razer DeathAdder 3500Dpi Left hand edition Razer Goliathus Speed 
  hide details  
Reply
post #26 of 68
There are so many things to nitpick at here.

Too much pressure results in turbulent flow that can cause uneven flow inside the waterblock.
Overclocking a Brisbane is kinda like teaching bricks to fly...
The pump-with-plate systems are limited by the SURFACE AREA of contact with the water and have a maximum thermal transmission no matter how fast the water is moving.
Store bought distilled water often has material added back in.
The longer the hose the more back-pressure on the system.
Your first set of fans should PUSH INTO the radiator... pulling through the radiator doesn't force air contact with the surface of the metal.
DO NOT run metal radiator systems without water additives, it will destroy the thermal transfer capacity of the radiator and CPU/GPU waterblock by altering the surface metal(s) exposed to water, in order to recover from this damage you must remove the altered metal by hand, thus reducing the design efficiency of the waterblock itself.

If you've got water cooling and you're still seeing a massive difference between the temperature inside the CPU and the water system, the problem is at the interface of the two.

Is there oxidation on EITHER the cpu spreader OR the waterblock spreader plates where they contact using paste?
How MUCH compound are you using between the CPU and the waterblock?


Do keep in mind that the range of capacity designed into these closed-loop kits is simply "to exceed standard air cooling options". I ran my Brisbanes with Zalman 105s on them and kept temperatures down just fine. I agree that you should keep them cool when running but your target zone is "anything under 62°C".

You won't see much improvement in capacity, if any at all, with a two core brisbane overclock since the IO of it is still locked at hypertransport speeds.

In the long run you're looking at needing to buy a new CPU plate, the old one is damaged.
Edited by prjindigo - 3/15/12 at 2:38pm
post #27 of 68
Quote:
Originally Posted by prjindigo View Post

There are so many things to nitpick at here.

Well, if we are into nitpicking:
Quote:
Originally Posted by prjindigo View Post

Too much pressure results in turbulent flow that can cause uneven flow inside the waterblock.

Actually pressure in itself does not have a huge influence on turbulence. Flows are characterised as turbulent, transitional or laminar based on the Reynolds number. The Reynolds number is defiend as:

Re = rho * v * L / mu

Where:
rho is the density
v is the mean velocity of the fluid
L is the characteristic length (i.e. hydraulic diameter)
mu is the dynamic viscosity.

Whilst pressure does have a marginal effect on rho and mu it is very much a seccond order effect compared to the velocity. Fluid velocity is driven by the pressure difference, not the absolute pressure.

So no, high pressure in itself does not cause turbulent flow, high velocity does.

Besides which, we want turbulence. In laminar flow a large boundary layer forms. This boundary layer is a layer of slow moving fluid which can effectively insulate the hot surface from the free stream. Turbulent flow reduces this boundary layer and promotes fluid mixing, helping to move the heat away from the hot surface.
Quote:
Originally Posted by prjindigo View Post

Your first set of fans should PUSH INTO the radiator... pulling through the radiator doesn't force air contact with the surface of the metal.

No, provided there is flow the air contacts the metal regardless of the position of the fan.

Actually push vs pull is more to do with your particular fan and rad combination. Pull has even been shown to be more effective than push for slow speed fans: proof.gif

This varies from setup to setup, so there are no hard and fast rules really.
Main
(21 items)
 
HTPC
(10 items)
 
 
CPUMotherboardGraphicsRAM
i5 2550k P8P67 Pro Sapphire HD 7950 G.Skill RipJaws X 1600 Cas 9 
Hard DriveHard DriveHard DriveCooling
Corsair Force 120 WD Blue 500GB WD Caviar Green 1TB XSPC RayStorm 
CoolingCoolingCoolingCooling
RX240 MCR 220 EK 7950 Copper Acetal  DDC-1T 
OSMonitorMonitorKeyboard
Windows 7 64-bit Dell U2311H Oculus Rift DK2 Ducky Shine 3 MX Brown 
PowerCaseMouseAudio
Corsair TX 750W CoolerMaster CM690 II G500 Klipsch ProMedia 2.1 
Audio
Asus Xonar DX 
CPUMotherboardRAMHard Drive
A10-6800K Gigabyte GA-F2A85XN-WIFI G Skill 1600 CAS9 Kingston SSD Now 60GB 
Hard DriveOptical DriveCoolingOS
WD Caviar Blue 1TB LG Slim Blu-Ray player Silverstone NT06-PRO  Widows 7 Home Premium 
PowerCase
Silverstone Sfx Series ST45SF 450W Silverstone SG05 
  hide details  
Reply
Main
(21 items)
 
HTPC
(10 items)
 
 
CPUMotherboardGraphicsRAM
i5 2550k P8P67 Pro Sapphire HD 7950 G.Skill RipJaws X 1600 Cas 9 
Hard DriveHard DriveHard DriveCooling
Corsair Force 120 WD Blue 500GB WD Caviar Green 1TB XSPC RayStorm 
CoolingCoolingCoolingCooling
RX240 MCR 220 EK 7950 Copper Acetal  DDC-1T 
OSMonitorMonitorKeyboard
Windows 7 64-bit Dell U2311H Oculus Rift DK2 Ducky Shine 3 MX Brown 
PowerCaseMouseAudio
Corsair TX 750W CoolerMaster CM690 II G500 Klipsch ProMedia 2.1 
Audio
Asus Xonar DX 
CPUMotherboardRAMHard Drive
A10-6800K Gigabyte GA-F2A85XN-WIFI G Skill 1600 CAS9 Kingston SSD Now 60GB 
Hard DriveOptical DriveCoolingOS
WD Caviar Blue 1TB LG Slim Blu-Ray player Silverstone NT06-PRO  Widows 7 Home Premium 
PowerCase
Silverstone Sfx Series ST45SF 450W Silverstone SG05 
  hide details  
Reply
post #28 of 68
Thread Starter 
Thanks for the suggestions guys.

I'm not sure why I never tried this, but I just hooked up the loop without a radiator and....... IT ACTUALLY WORKS! I'm sitting at 27c (idle temp) with 1.52v @ 3.3ghz. Amazing. That tells me one of the radiators is clogged up. And judging by how well the H60 worked before I hacked it up, I'm going to say the Black Ice GT 240 is what caused all the problems to begin with.

More than likely that..gunk I cleaned out of the CPU block was from that GT 240, and I'm sure it's found its way into the stock Corsair 120 by now because I've been switching back and forth.
post #29 of 68
Thread Starter 
Sure enough, it was without a doubt the radiator.

I put some 91% alcohol in the GT 240 and let it set for a little while. After that I hooked the pump up Res > Rad > Res to force anything out, and reconnected it using Res > CPU block > Radiator > Res. I guess the alcohol didn't do a very good job, because temps quickly jump up under load just like they did before.


Btw, I emptied the reservoir and refilled it with cold water. The water got warm using no radiator, so I used cold water to help radiate the heat. With the radiator connected and under Prime95 the res water was still cold. The tubing and radiator were cold also, but the CPU temp was climbing quickly into the high 40's low 50's within seconds. So either there is a massive build up in the radiator which 91% alcohol and a 330GPH pump couldn't clear, or there is some sort of internal damage to the rad itself

I'll try the stock Corsair H60 120 rad and report back.
post #30 of 68
Thread Starter 
Tried the H60 and got the same result.

What I don't get it, I have good pressure. If I take the hose that goes back to the reservoir out of the water, there is a nice flow of water coming out. The pump has an adjustment to control pressure also. If I turn the knob to reduce pressure, the you can see the water pressure lower on the exhaust hose.

What I've figured out so far:

- Using no radiator at all = Excellent temps. That is until the reservoir water heats up.
- Installing either radiator in the loop = High temps, yet I still have excellent pressure.
- Using NO radiator I can feel the res water get hotter, as well as the hose temps.
- Using a radiator, NOTHING gets warm. Everything stays cool but CPU temp continues to rise.
- When using either radiator, fans make no difference at all. No fans or two fans running @ full speed = No change in temps.

So it has to be a problem with the radiators, yes? 91% rubbing alcohol didn't seem to work for cleaning. Anything else I should try?
Edited by *ka24e* - 3/15/12 at 6:19pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Water loop, never ending air pockets? SOLVED!