The Thermochill one will be coming soon, we've had some problems due to supply chain issues. :) I will try to add some 7v tests in there as well when I do that one! :D
go lower, some of us like silence ;)
i run 3 of these on my PA120.3 hooked up to a rheobus
i run them at about 5v at night, they're barely spinning but its still enough to keep a 3.5ghz yorkfield and a 3870 plenty cool :)
<3 thermochill :D
I really liked this review. I had not seen one on rads and I have been looking for one for a while.
From what I gathered the rads in the tests are supposed to be the best. Could you throw in a Thermaltake TMG in there? It would be interesting to see if the cheaper rads would hold their ground... for example a system with TMG1 and TMG2 (120+240 rad) vs any 240 on the review. The Big water 745 is an all in one and costs â¬100 and includes the TMGs. Maybe is it worth it for a beginner to start with this and add a good CPU block and work the way up from there as experience and funding increases?
I have been interested in water cooling, but it has always been too expensive :S
Originally Posted by Article "Roughly translated, this means that if a system can dissipate 75W and 95W of heat respectively and you're only putting 25W of heat into it, both will cool similarly assuming equal flow. However, when you start dumping 65-70W of heat into the system, you'll find a sizeable difference in temperatures between the two."
NO!!!!!!
Trying to do this starts to break the laws of thermodynamics all over the place (not to mention ones like conservation of energy!). The rule
power = mass flow rate x Cp x temperature difference
will always apply no matter how much heat you're dumping into the system. Since Cp is a physical constant (1010 J/kg for dry air) and mass flow rate is set by your pump or fan (depending on whether you're looking at the water side or air side - the formula applies to both) that means temperature difference must be directly proportional to power *.
If you have two systems, one of which can dissipate 75W and the other 95W for an identical set of mass flows and temperatures, then for that same mass flow/temperature combination the temperature difference between the water and atmospheric temperature will be 27% higher at 25W heat load for the 75W radiator. If you aren't seeing this then you have problems with your temperature measurement system.
It is entirely possible that this temperature difference may not be a problem in practice, but you can't just wave your hands and pretend it doesn't exist.
* Note: I'm assuming we're operating at the low temperatures of a typical watercooling setup here - as the temperature rises radiation cooling becomes a lot more important, and this is proportional to (temperature)^4.
Originally Posted by Article "Roughly translated, this means that if a system can dissipate 75W and 95W of heat respectively and you're only putting 25W of heat into it, both will cool similarly assuming equal flow. However, when you start dumping 65-70W of heat into the system, you'll find a sizeable difference in temperatures between the two."
NO!!!!!!
Trying to do this starts to break the laws of thermodynamics all over the place (not to mention ones like conservation of energy!). The rule
power = mass flow rate x Cp x temperature difference
will always apply no matter how much heat you're dumping into the system. Since Cp is a physical constant (1010 J/kg for dry air) and mass flow rate is set by your pump or fan (depending on whether you're looking at the water side or air side - the formula applies to both) that means temperature difference must be directly proportional to power *.
If you have two systems, one of which can dissipate 75W and the other 95W for an identical set of mass flows and temperatures, then for that same mass flow/temperature combination the temperature difference between the water and atmospheric temperature will be 27% higher at 25W heat load for the 75W radiator. If you aren't seeing this then you have problems with your temperature measurement system.
It is entirely possible that this temperature difference may not be a problem in practice, but you can't just wave your hands and pretend it doesn't exist.
* Note: I'm assuming we're operating at the low temperatures of a typical watercooling setup here - as the temperature rises radiation cooling becomes a lot more important, and this is proportional to (temperature)^4.
Hey PDF27,
You are DEAD on that this is the proper thermodynamic theory. The problem comes in theory != practice. In reality, the closer you get to the outer thermal dissipation, the more that the difference you explained becomes apparent.
It's much like a power supply. Theoretically, at 80% load and 20% load there really shouldn't be a difference in efficiency. However, that's patently untrue, because the basic theory does not account for current leak, minimums held by the capacitors, and various other inefficiencies that come from an imperfect system.
If you test a 240mm radiator and a 360mm radiator both with X heat where X is less than about 40% of load, you'll find they're fairly constant even though the 360mm rad can dissipate considerably more heat with the same fan speed and flow rate. That's cause you'll only be able to get the temperature down to an equilibrium, which at load there will be X amount of heat being dumped in so you can dissipate Y amount of it. It's the part where theory doesn't quite mesh with practice, and if you have a small enough spare system you can see that the same applies to a 120 vs. a 240mm rad.
Basically, yes, a 95w dissipating system should be more efficient across the board than a 75w dissipating system, no matter what the input power. However, in practice there's a point where things just stay cool. :) If both systems can keep the fluid at ambient, nothing is going to make the 360mm radiator cool any "better" than the 240mm. You'll just be able to volt down your fans more and still maintain the same temperature.
Where the theory breaks off, basically, is the allowance for ambient temps. The fluid simply cannot go below that in any passive system. So if you can cool 75W of heat, the system requires about 33% of its overall ability to hit ambient, whereas with 95W it only needs 26%. It is 27% more EFFICIENT at reaching the SAME temperature. Bottom line, your temps don't change until the system is under enough load that one system can't dissipate it all and the other still can, which is what I was trying to illustrate.
Don't forget, we're also talking about CPU temps - which are an equilibrium between CPU and block and fluid. So if fluid is being kept at roughly ambient (which it can't go lower) then CPU temps just plain don't notice a difference. :) Which is the point I'm trying to get to. :D
Originally Posted by Da Dego However, in practice there's a point where things just stay cool. If both systems can keep the fluid at ambient, nothing is going to make the 360mm radiator cool any "better" than the 240mm.
Neither system will ever be able to keep the fluid at ambient unless they have a heat pump fitted. For there to be any heat transfer, there must be a temperature difference - for it to be otherwise (absent Rankine's Daemon or similar) is a violation of the Zeroth Law of Thermodynamics.
I'm quite willing to accept that the performance difference between two radiators is too small to measure with the equipment you're currently using. I sometimes do thermal testing for a living, and know exactly how hard it is to get decent data (and how bad manufacturer's data often is).
Quote:
Originally Posted by Da Dego If both systems can keep the fluid at ambient, nothing is going to make the 360mm radiator cool any "better" than the 240mm. You'll just be able to volt down your fans more and still maintain the same temperature.
You what? If the 360mm radiator can use a lower mass flow of air per fan for the same air-water temperature difference then by definition it is cooling better and hence for the same mass flow of air per fan will keep the water cooler (closer to the ambient air temperature).
Quote:
Originally Posted by Da Dego It is 27% more EFFICIENT at reaching the SAME temperature.
This is the one that I really object to. If the max thermal dissipations are defined for the same mass flows of air and water, and the same temperature difference between air and water inlets then they simply cannot reach the same temperature - that would be a violation of the zeroth law again.
Quote:
Originally Posted by Da Dego then CPU temps just plain don't notice a difference. Which is the point I'm trying to get to.
Now we're getting somewhere. If you're going by CPU diode temperatures then the chances of you noticing temperature differences of the magnitude we're dealing with here are minute.
Incidentally, how long did you run each test before you took temperatures?
Originally Posted by Da Dego CPU temps just plain don't notice a difference. :) Which is the point I'm trying to get to. :D
Now we're getting somewhere. If you're going by CPU diode temperatures then the chances of you noticing temperature differences of the magnitude we're dealing with here are minute.
Incidentally, how long did you run each test before you took temperatures?
Quote:
Originally Posted by Article
Testing Methodology
Understanding the test methods is vital to understanding the results of a test, as well as allowing maximum repeatability. Here at bit-tech we strive to make our tests fair, unbiased, and repeatable by anyone with the time or inclination to do so. The following methods and assumptions were used in this test:
The test setup is built in a climate controlled room free of unnecessary clutter. Once assembled, the system was started on air cooling and the operating system (Windows Vista SP1) was patched fully. Speedfan 4.34 was used to measure the temperature of the CPU cores before, during and after testing as prescribed below. Ambient temperatures were also taken from a digital thermometer on the wall approximately one metre from the test setup.
For this test, Ambient ambient room temperature was 22 degrees Celsius.
The system was booted with CPU and GPU blocks along with the radiator being tested, and left to idle at the desktop (no screen saver) for one hour to allow the fluid temperatures to come to equilibrium, and temperature was then recorded.
Prime95 was then started on each core using Small FFT to assure maximum CPU usage, which was also monitored via a sideboard widget to assure proper load. Temperatures were recorded at 10, 20, 30, and 60 minutes and averaged. No cores showed any significant change in temperature (greater than one degree) after 20 minutes. Once 60 minutes passed, Prime95 was stopped on all cores and the system was allowed to return to idle. Temperatures were recorded at 10, 20 and 30 minutes though no readings showed deviations past 10 minutes.
The system was then shut down and drained. The radiator was removed, and replaced with the next in series. This sequence was continued until all radiators had been tested. For reference purposes (if anyone truly cares), the tests went in the following order: Black ice 360, X-Changer 360, Black Ice 240, X-Changer 240, Black Ice 120, X-Changer 120.
Upon completion of this, the test was restarted from scratch. Results were averaged amongst the two runs to develop the reported findings on the next page.
As an additional note, the CPU and GPU were not moved or altered to maintain testing integrity. The radiator path is designed such that removing the radiator would not cause any additional movement or reseating of either block. By carefully managing subtle issues, all changes in temperature will be accountable to the radiator only.
Anyway PDF27, thanks for opening up an interesting discussion about something which had confused me. What is your Penguin writing?
Comments 26 to 31 of 31
i run 3 of these on my PA120.3 hooked up to a rheobus
i run them at about 5v at night, they're barely spinning but its still enough to keep a 3.5ghz yorkfield and a 3870 plenty cool :)
<3 thermochill :D
From what I gathered the rads in the tests are supposed to be the best. Could you throw in a Thermaltake TMG in there? It would be interesting to see if the cheaper rads would hold their ground... for example a system with TMG1 and TMG2 (120+240 rad) vs any 240 on the review. The Big water 745 is an all in one and costs â¬100 and includes the TMGs. Maybe is it worth it for a beginner to start with this and add a good CPU block and work the way up from there as experience and funding increases?
I have been interested in water cooling, but it has always been too expensive :S
Trying to do this starts to break the laws of thermodynamics all over the place (not to mention ones like conservation of energy!). The rule
will always apply no matter how much heat you're dumping into the system. Since Cp is a physical constant (1010 J/kg for dry air) and mass flow rate is set by your pump or fan (depending on whether you're looking at the water side or air side - the formula applies to both) that means temperature difference must be directly proportional to power *.
If you have two systems, one of which can dissipate 75W and the other 95W for an identical set of mass flows and temperatures, then for that same mass flow/temperature combination the temperature difference between the water and atmospheric temperature will be 27% higher at 25W heat load for the 75W radiator. If you aren't seeing this then you have problems with your temperature measurement system.
It is entirely possible that this temperature difference may not be a problem in practice, but you can't just wave your hands and pretend it doesn't exist.
* Note: I'm assuming we're operating at the low temperatures of a typical watercooling setup here - as the temperature rises radiation cooling becomes a lot more important, and this is proportional to (temperature)^4.
Hey PDF27,
You are DEAD on that this is the proper thermodynamic theory. The problem comes in theory != practice. In reality, the closer you get to the outer thermal dissipation, the more that the difference you explained becomes apparent.
It's much like a power supply. Theoretically, at 80% load and 20% load there really shouldn't be a difference in efficiency. However, that's patently untrue, because the basic theory does not account for current leak, minimums held by the capacitors, and various other inefficiencies that come from an imperfect system.
If you test a 240mm radiator and a 360mm radiator both with X heat where X is less than about 40% of load, you'll find they're fairly constant even though the 360mm rad can dissipate considerably more heat with the same fan speed and flow rate. That's cause you'll only be able to get the temperature down to an equilibrium, which at load there will be X amount of heat being dumped in so you can dissipate Y amount of it. It's the part where theory doesn't quite mesh with practice, and if you have a small enough spare system you can see that the same applies to a 120 vs. a 240mm rad.
Basically, yes, a 95w dissipating system should be more efficient across the board than a 75w dissipating system, no matter what the input power. However, in practice there's a point where things just stay cool. :) If both systems can keep the fluid at ambient, nothing is going to make the 360mm radiator cool any "better" than the 240mm. You'll just be able to volt down your fans more and still maintain the same temperature.
Where the theory breaks off, basically, is the allowance for ambient temps. The fluid simply cannot go below that in any passive system. So if you can cool 75W of heat, the system requires about 33% of its overall ability to hit ambient, whereas with 95W it only needs 26%. It is 27% more EFFICIENT at reaching the SAME temperature. Bottom line, your temps don't change until the system is under enough load that one system can't dissipate it all and the other still can, which is what I was trying to illustrate.
Don't forget, we're also talking about CPU temps - which are an equilibrium between CPU and block and fluid. So if fluid is being kept at roughly ambient (which it can't go lower) then CPU temps just plain don't notice a difference. :) Which is the point I'm trying to get to. :D
I'm quite willing to accept that the performance difference between two radiators is too small to measure with the equipment you're currently using. I sometimes do thermal testing for a living, and know exactly how hard it is to get decent data (and how bad manufacturer's data often is).
Incidentally, how long did you run each test before you took temperatures?
Anyway PDF27, thanks for opening up an interesting discussion about something which had confused me. What is your Penguin writing?