Overclock.net › Forums › Cooling › Water Cooling › Radiator Thermal Testing In Progress
New Posts  All Forums:Forum Nav:

Radiator Thermal Testing In Progress - Page 3

post #21 of 43
Thread Starter 
Thanks!
You've read my mind on the shrouds Ira, that's why I set up the test bench the way I did, so I can run a few of those types of experiments and figure out some correction factors for shrouds.

I decided to remove my variac from the testing because I didn't trust how much power it was consuming, and it was a good thing to do. With this test, it looks like a straight line was as good as any to fit, and this makes life a little easier.



And while not the same fan, comparing my c/w result is pretty close to the Delta 1212M fan at 12V that Bill Adams ran with thermochills testing.


So I'm feeling pretty good, I think I'll make it my standard to just test at 300 and again at 600 watts and use a linear trendline.

Next up is to try some flow rate effects testing and I'll have my method down pretty well I think.
    
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 #22 of 43
Quote:
You've read my mind on the shrouds Ira, that's why I set up the test bench the way I did, so I can run a few of those types of experiments and figure out some correction factors for shrouds.
Great...It will be interesting to see how lower and higher fin per inch rads respond to shrouds...It only makes sense that the LFPI rads wouldn't see as much gain as the higher, but a lot of this isn't as intuitive as it would seem either...Whats the name of that old TV show "Expect the Un-Expected"..

Take your time with it Martin, don't burn yourself out with it, your just now getting to the fun part......There are just so many variables, of flow and fan and shroud, it could really get overwhelming...Like you were saying 300 and 600 are the most telling wattages anyway for a typical rig, on the 480 I would just bump it up to include 900W ....

Quote:
I decided to remove my variac from the testing because I didn't trust how much power it was consuming, and it was a good thing to do. With this test, it looks like a straight line was as good as any to fit, and this makes life a little easier
Well thats just one more variable you've removed and its still all good...I hope your heaters don't die to fast, that could get expensive...But Variac or not it would be the same...

I just now caught that your running the medium YL's and not the ultra low...I would of thought the PA could do a little better then that with mediums...Still a very nice delta at 300W....

Whats your max GPM to test with?

Looks like fun Martin......Take your time, this is going to be some great info....
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 #23 of 43
Thread Starter 
Quote:
Originally Posted by ira-k View Post
Great...It will be interesting to see how lower and higher fin per inch rads respond to shrouds...It only makes sense that the LFPI rads wouldn't see as much gain as the higher, but a lot of this isn't as intuitive as it would seem either...Whats the name of that old TV show "Expect the Un-Expected"..

Take your time with it Martin, don't burn yourself out with it, your just now getting to the fun part......There are just so many variables, of flow and fan and shroud, it could really get overwhelming...Like you were saying 300 and 600 are the most telling wattages anyway for a typical rig, on the 480 I would just bump it up to include 900W ....



Well thats just one more variable you've removed and its still all good...I hope your heaters don't die to fast, that could get expensive...But Variac or not it would be the same...

I just now caught that your running the medium YL's and not the ultra low...I would of thought the PA could do a little better then that with mediums...Still a very nice delta at 300W....

Whats your max GPM to test with?

Looks like fun Martin......Take your time, this is going to be some great info....

Speaking of which, the one heater I wasn't using cracked already. Not sure how that happened, it must have been from the other heat that ran all night.

And the one heater I bought locally that was a bit more expensive has some sort of additional overheating protection. I was going to try to do some flow rate effects at the 600 watt load, but running .5GPM rose the temperature enough that it started kicking off. I guess these things weren't meant to run that hot.. If they break quickly, I'll just have to buy some long canister heaters and build my own heatsinks for them, those things are bulletproof.

Not sure I'm going to get anything worthwhile out of the flow rate effects, I think the difference is going to be smaller than my accuracy can show. At least my .5, 1.0 and 1.5 runs don't seem to indicate anything.

I may just have to give up on the flow rate effects and stick with a specific flow rate like 1.5GPM. From what I can see flow rate seems to have minimal impact anyway. I'm going to try one more at max flow which is about 3.3GPM with my DDC2 on the PA.

Yes, those are medium speed yates. I'm suprised how quiet they are, really not any more noisy than the slow speed yates from what I can tell, nice fans, they were one of Vapors top picks for flow rate through a radiator in his testing.
    
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 #24 of 43
Quote:
Originally Posted by Martinm210 View Post
Speaking of which, the one heater I wasn't using cracked already. Not sure how that happened, it must have been from the other heat that ran all night.

And the one heater I bought locally that was a bit more expensive has some sort of additional overheating protection. I was going to try to do some flow rate effects at the 600 watt load, but running .5GPM rose the temperature enough that it started kicking off. I guess these things weren't meant to run that hot.. If they break quickly, I'll just have to buy some long canister heaters and build my own heatsinks for them, those things are bulletproof.

Not sure I'm going to get anything worthwhile out of the flow rate effects, I think the difference is going to be smaller than my accuracy can show. At least my .5, 1.0 and 1.5 runs don't seem to indicate anything.

I may just have to give up on the flow rate effects and stick with a specific flow rate like 1.5GPM. From what I can see flow rate seems to have minimal impact anyway. I'm going to try one more at max flow which is about 3.3GPM with my DDC2 on the PA.

Yes, those are medium speed yates. I'm suprised how quiet they are, really not any more noisy than the slow speed yates from what I can tell, nice fans, they were one of Vapors top picks for flow rate through a radiator in his testing.
Ouch...I guess they aren't meant to be guppie fryers.....You might have to go all industrial on them to get some longevity, besides metal is just cool anyway....

I wasn't thinking either, the flow just through a rad probably wont have much to do with overall temp's except maybe at the low end and high end...

What rad was he using with the fans?
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 #25 of 43
Thread Starter 
Quote:
Originally Posted by ira-k View Post
Ouch...I guess they aren't meant to be guppie fryers.....You might have to go all industrial on them to get some longevity, besides metal is just cool anyway....

I wasn't thinking either, the flow just through a rad probably wont have much to do with overall temp's except maybe at the low end and high end...

What rad was he using with the fans?
Yeah, might have to look for something a little more industrial, we'll see how these last two hold up.
He used an MCR120 for his radiator restricted testing, so it would be a whole different ball of wax with more dense and more restrictive radiators where pressure capabilities would shine.

http://www.xtremesystems.org/forums/...d.php?t=171661

This is the graph I had taken an interest in when picking the fans for medium/low speed:

    
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 #26 of 43
Your my hero> this is looking great. BTW nice harware you got there
post #27 of 43
Yeah that medium kept half its cfm through the rad, not bad at all for a 25mm thick fan...That one clear low sped YL was having fits wasn't it, that thing was all over....
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 #28 of 43
What does Petra Curved Mean? The same model number coming with different blades?

There seem to be quite a bit of variation between the different D12SL12 variation. The Yates Loon catalog only lists it at 1350 rpm, 47 cfm, and 28db. The D12BL-12 is a ball bearing version rated at 1200rpm, 45cfm, and 24db. What is interesting is the noise advantage wasn't there in the test.

An actual RPM measurement would have been nice, to see if erratic voltages are the cause. I couldn't tell from the right up, was only one fan of each tested or a few different samples?

The non-linearities in the measurements and they way they over lap would (to me) indicate several are pretty similar.

Using the flows and published PQ curves on the Pansonics and Deltas, parts of the PQ curve are deducible. At 65cfn, the 12v M1 is at about .04 in H2O; at 53cfm, the L1 at 12v is at about .03 in of H2O; the H1 at 83cfm would be .034 in H2O. At 150cfm, the Delta FFB1212EHE at 12v is at around .36 in H2O. A trouble comes with the 120x76mm GFB1212. Even the most extreme VHW is only rated at .30 in H2O at 190cfm. However, the radiator would flow more are at .36" vs .30". Therefore, unless Delta's published measurements between its own fans are way off (unlikely), there may be a fault in the measurement methodology.

The PQ curve of the radiator is important, since the fans tend to have a knee in their PQ curve, and would probably draw more air vs. noise if they were above the knee.

A useful fan test, however maybe not at the granularity indicated.
Number 1
(13 items)
 
  
CPUMotherboardGraphicsRAM
Q6600 B3 EVGA 780i A1 MSI GTX260/192 2x2GB Corsair Dominator PC1066 
Hard DriveOptical DriveOSMonitor
1x500GB Maxtor 1 Asus SATA Vista x64 Ultimate 32" 1080P TV 
KeyboardPowerCaseMouse
Logitech Wave USB 1000w Cooler Master Lian Li V2000 Ideazon Reaper Edge 
Mouse Pad
QPad QT-R 
  hide details  
Reply
Number 1
(13 items)
 
  
CPUMotherboardGraphicsRAM
Q6600 B3 EVGA 780i A1 MSI GTX260/192 2x2GB Corsair Dominator PC1066 
Hard DriveOptical DriveOSMonitor
1x500GB Maxtor 1 Asus SATA Vista x64 Ultimate 32" 1080P TV 
KeyboardPowerCaseMouse
Logitech Wave USB 1000w Cooler Master Lian Li V2000 Ideazon Reaper Edge 
Mouse Pad
QPad QT-R 
  hide details  
Reply
post #29 of 43
Thread Starter 
Yeah, I really don't know too much about fans yet, just scratching the surface, but I do know the "Petra" curved as indicated in Vapors tests was something he found in his own testing. Apparently Petra's orders direct from Yate, where other vendors use a middlman of sorts. Vapor found that the same exact fan from Petra outperformed the same fan from others.

Not sure I understand that, but needless to say my medium speed yates I'm currently using are from Petras that I bought myself...

I'm just going to undervolt as needed to match specific RPM ranges.

1550 native D12SM12
1350 native D12SL12
1000 undervolt

Then I'm hoping over to the 3000 RPM Scythe Ultra Kaze 38mm fans. A quick static pressure test has it producing around .22" H2O pressure. I figured I'd run once at 3000 RPM then 2000RPM to match the 87CFM variety. That should cover a fairly good range of Fan RPMs.
    
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 #30 of 43
Quote:
Originally Posted by Martinm210 View Post
Thanks!
Trial 4, had to throw out number 3 completly after seeing the thermostat kick on and off repeated, this is what I want to see a nice constant delta between all 4 measurements.

One good test down.
Martin, I know it doesn't seem to be the focus of your test. But what struck me is that after your warm up period, the delta between "Water In" and "Water Out" appears to be much less that 1c. Further the delta seems to remain pretty constant at all heat loads.

Again it makes me question the "coolest water is leaving the rad" dogma most adhere to. While it's true in absolute terms, from your test data, it would appear to be very insignificant.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Radiator Thermal Testing In Progress