Overclock.net › Forums › Cooling › Water Cooling › Is a stronger pump better?
New Posts  All Forums:Forum Nav:

Is a stronger pump better? - Page 2

post #11 of 25
Thread Starter 
Quote:
Originally Posted by cgrado View Post
yeah, but if it goes slower, it has more time to pick up heat, so each drop picks up more heat as opposed to a lot of water picking up less heat because it moves faster, so the same amount of heat is absorbed. (just my uneducated little theory, i could be completely and absolutely wrong).
If what goes slower? Pick up heat from where? I know you know what you are talking about, but we don't. In a technical conversation, you have to clearly define your subject.
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
post #12 of 25
If the water moves slower as it passes the heat source, wouldn't more heat transfer to the water than if the same amount of water was moving at a faster past the heat source?
Immortal
(13 items)
 
  
CPUMotherboardGraphicsRAM
E6600 44G 3.240GHz Asus P5B-deluxe eVGA 8800GT (650/950) 2GB PQIturbomem 667(800mhzballistix soon) 
Hard DriveOptical DriveOSMonitor
1x 400GB Seagate 7200rpm/16mb cache Some super combo drive Windows XP Home/Ubuntu NEC MultiSync LCD1970GX 
KeyboardPowerCaseMouse
Logitech cordless 610w PCP&C silencer TTArmor w/side fan G7 
Mouse Pad
Black IcematII 
  hide details  
Reply
Immortal
(13 items)
 
  
CPUMotherboardGraphicsRAM
E6600 44G 3.240GHz Asus P5B-deluxe eVGA 8800GT (650/950) 2GB PQIturbomem 667(800mhzballistix soon) 
Hard DriveOptical DriveOSMonitor
1x 400GB Seagate 7200rpm/16mb cache Some super combo drive Windows XP Home/Ubuntu NEC MultiSync LCD1970GX 
KeyboardPowerCaseMouse
Logitech cordless 610w PCP&C silencer TTArmor w/side fan G7 
Mouse Pad
Black IcematII 
  hide details  
Reply
post #13 of 25
Thread Starter 
Quote:
Originally Posted by cgrado View Post
If the water moves slower as it passes the heat source, wouldn't more heat transfer to the water than if the same amount of water was moving at a faster past the heat source?
The only way you could have the same amount of water moving at a faster rate is using a different time interval, and you can't compare things on different time intervals.
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
post #14 of 25
Quote:
Originally Posted by cgrado View Post
(just my uneducated little theory, i could be completely and absolutely wrong).
Quote:
Originally Posted by pauldovi View Post
The only way you could have the same amount of water moving at a faster rate is using a different time interval, and you can't compare things on different time intervals.
^^^
Immortal
(13 items)
 
  
CPUMotherboardGraphicsRAM
E6600 44G 3.240GHz Asus P5B-deluxe eVGA 8800GT (650/950) 2GB PQIturbomem 667(800mhzballistix soon) 
Hard DriveOptical DriveOSMonitor
1x 400GB Seagate 7200rpm/16mb cache Some super combo drive Windows XP Home/Ubuntu NEC MultiSync LCD1970GX 
KeyboardPowerCaseMouse
Logitech cordless 610w PCP&C silencer TTArmor w/side fan G7 
Mouse Pad
Black IcematII 
  hide details  
Reply
Immortal
(13 items)
 
  
CPUMotherboardGraphicsRAM
E6600 44G 3.240GHz Asus P5B-deluxe eVGA 8800GT (650/950) 2GB PQIturbomem 667(800mhzballistix soon) 
Hard DriveOptical DriveOSMonitor
1x 400GB Seagate 7200rpm/16mb cache Some super combo drive Windows XP Home/Ubuntu NEC MultiSync LCD1970GX 
KeyboardPowerCaseMouse
Logitech cordless 610w PCP&C silencer TTArmor w/side fan G7 
Mouse Pad
Black IcematII 
  hide details  
Reply
post #15 of 25
Quote:
Originally Posted by pauldovi View Post
I am not sure how you figure the energy absorbed by the water will be constant.
The processor is consuming electrical energy at a constant rate and generating a constant amount of heat energy. If the processor remains at a constant temperature, then a constant amount of heat energy is being transferred away from it. This heat energy must equal the electrical energy consumed (conservation.) Since the conditions under which the processor was operating under was constant, then it follows that heat output throughout your tests must have been constant.

Just step back for a moment and take a practical, common sense view of the heat output you recorded. If your processor is really putting out at least 206 J/s, then at the lowest flow rate a minimum of 71 J/s must be dissipating through secondary heat paths. That’s more energy than the processor consumes at stock speed and voltage. If you’ve done any soldering you know that 71 watts worth of heat energy will easily melt solder. I seriously doubt that the socket’s solder joints were melting.

Quote:
Originally Posted by pauldovi View Post
That means we could insert a pump that moves the water at a trickle and still have the same cooling performance?
No. What it means is that we could insert a pump that moves the water at a trickle and have the same amount of energy transferred to the water. You will always have the same amount of energy transferred. The only question is at what temperature this transfer will take place. Cooling is something that happens at the other end of the cycle and has nothing to do with what I’m talking about.

Quote:
Originally Posted by pauldovi View Post
Increase flow will increase energy transfer, this is hypothesized by theoretical physics
Well, if we are talking about flow alone without the influence of turbulence, then that is not correct and is most definitely not supported by physics theorems. Fourier’s Law says that as mass increases q will increase, but only if delta T remains the same. In PC watercooling, it’s the q that is constant. q is the heat energy of the processor...it is our source. Changing flow cannot change the heat energy produced by the processor nor can flow affect the amount of heat transferred, as all of it must be transferred to maintain a constant CPU temperature. This means that when flow increases, delta T must change. Many experiments over the years have shown that this is exactly what happens.

Quote:
Originally Posted by pauldovi View Post
The published thermal output of a processor is low at best.
I would suggest you examine Tom’s Hardware review of the Core 2. The power consumption tests show that the new chips have huge reductions. I believe Intel specs are accurate.

Quote:
Originally Posted by pauldovi View Post
I did not find any effect of flow rate on the radiator. The input temperature for the water stayed fairly consistent. This means the radiator was pretty much returning the water to the same temperatures at all times. That actually puts higher heat transfer at low flow rates on the radiator (which makes sense, the longer the water spends in the radiator, the more heat removed).
And this didn’t raise a concern? If you double the flow, the delta T must be cut in half. You tripled your flow. There should have been a large drop in delta T.

Measuring practices state that you must have 10 times the discrimination to measure accurately. That means that if you want to be accurate to the tenth of a degree, then you must have an instrument that is accurate to the hundredth of a degree. The type of measurements you’re trying to make requires accuracy to the hundredth, and that takes expensive equipment. This is why it’s so difficult to do these kinds of tests.
post #16 of 25
Thread Starter 
Quote:
The processor is consuming electrical energy at a constant rate and generating a constant amount of heat energy. If the processor remains at a constant temperature, then a constant amount of heat energy is being transferred away from it. This heat energy must equal the electrical energy consumed (conservation.) Since the conditions under which the processor was operating under was constant, then it follows that heat output throughout your tests must have been constant.
The processor DID NOT remain at a constant temperature. And what is heat energy? Heat is transferred energy. Yes, the heat generated by the processor is constant, but as the test shows, the temperature of the processor was NOT constant, meaning that heat transfer varied.

Quote:
Just step back for a moment and take a practical, common sense view of the heat output you recorded. If your processor is really putting out at least 206 J/s, then at the lowest flow rate a minimum of 71 J/s must be dissipating through secondary heat paths. That’s more energy than the processor consumes at stock speed and voltage. If you’ve done any soldering you know that 71 watts worth of heat energy will easily melt solder. I seriously doubt that the socket’s solder joints were melting.
If you have done any soldering, you would remember that all that heat is focused at the tip of the soldering gun. Try spreading the heat out and see if you can melt the solder....

Quote:
No. What it means is that we could insert a pump that moves the water at a trickle and have the same amount of energy transferred to the water. You will always have the same amount of energy transferred. The only question is at what temperature this transfer will take place. Cooling is something that happens at the other end of the cycle and has nothing to do with what I’m talking about.
This is simply not true. A related rate can be found because the heat transfer and the temperature of the water. (I probably should have included this in the report). You see, as the water's temperature increases (caused by the absorption of thermal energy), the amount of thermal energy transfered will decrease. This is because dT is smaller (the difference in temperature between the processor and the water). If dT is smaller, so is dQ (just look at the equation). Therefore, water spending a longer amount of time in the processor would result in reduced heat transfer. The results of my experiment confirm this.

Quote:
Well, if we are talking about flow alone without the influence of turbulence, then that is not correct and is most definitely not supported by physics theorems. Fourier’s Law says that as mass increases q will increase, but only if delta T remains the same. In PC watercooling, it’s the q that is constant. q is the heat energy of the processor...it is our source. Changing flow cannot change the heat energy produced by the processor nor can flow affect the amount of heat transferred, as all of it must be transferred to maintain a constant CPU temperature. This means that when flow increases, delta T must change. Many experiments over the years have shown that this is exactly what happens.
You keep on claiming that the CPU temperature is a constant, but it isn't. Yes, the Q released from the processor is a constant, however, the rate at which it is removed from the processor by the water is NOT a constant. Look at the variables that cause heat transfer.

1. M - More water goes through, M increases
2. dT - the longer water stays in the block the smaller dT is.
3. c - consant.

You see there are 2 changing variables that affect dQ, this is the mass of the water going through the block, and the difference in temperature of the water and the processor.

Quote:
I would suggest you examine Tom’s Hardware review of the Core 2. The power consumption tests show that the new chips have huge reductions. I believe Intel specs are accurate.
I have never regarded Tom's Hardware as a reliable source....

Quote:
And this didn’t raise a concern? If you double the flow, the delta T must be cut in half. You tripled your flow. There should have been a large drop in delta T.
Doubling flow did not cut dT in half. Why would this be expected to happen? The relationship is not linear.

Quote:
Measuring practices state that you must have 10 times the discrimination to measure accurately. That means that if you want to be accurate to the tenth of a degree, then you must have an instrument that is accurate to the hundredth of a degree. The type of measurements you’re trying to make requires accuracy to the hundredth, and that takes expensive equipment. This is why it’s so difficult to do these kinds of tests.
I understood and noted the limitations caused by lack of accuracy in the measurements. However, accuracy does not effect trends found in data. In my previous post I misused the word precision. The appropriate word to use is accuracy.
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
post #17 of 25
Quote:
Originally Posted by pauldovi View Post
The processor DID NOT remain at a constant temperature.
Sorry, I meant that within each test the processor temperature remained fairly constant, with only 1 C fluctuations from the average. That is within the accuracy of the sensor.

Quote:
Originally Posted by pauldovi View Post
And what is heat energy?
Yes, I know “heat energy†isn’t a proper physics term. I simply use it as a label to refer to the energy that was transferred by heat. For example, if we have a liter of water and use a burner to raise its temperature by 1 C, then I would refer to the 1000 calories as heat energy merely because it was transferred by direct heat, rather than, say electrical resistance to a current. As I say, it’s just a label that I use to refer to the amount of energy transferred to the water from the CPU. People seem to get it so that’s why I use it. But I will refrain from using it in these discussions.

Quote:
Originally Posted by pauldovi View Post
Yes, the heat generated by the processor is constant, but as the test shows, the temperature of the processor was NOT constant, meaning that heat transfer varied.
Sorry, I confused you by using my “heat energy†term. The ENERGY that is transferred by heat is constant. It must be. The heat transfer conditions can vary, but the energy transferred must always be the same. Similar tests conducted by ProCooling, SystemCooling, and others show the energy transfer to be the same at various flow rates. That’s one of the validations methods of the test results. There’s no way around it...for a processor, which is consuming 100 watts, to remain at a steady temperature, 100 watts of energy has to transfer as heat away from the processor.

Quote:
Originally Posted by pauldovi View Post
Try spreading the heat out and see if you can melt the solder....
Well, it isn’t spread out, it’s concentrated at each pin, and it can definitely melt the solder, but forget all that...you’re missing the point. If a processor package could dissipate 70 watts of energy without damage then, at stock speed and voltage you wouldn’t need a heatsink for your Intel Core 2! Obviously that can’t be.

Quote:
Originally Posted by pauldovi View Post
You see, as the water's temperature increases (caused by the absorption of thermal energy), the amount of thermal energy transfered will decrease.
Thermal energy? What's that?

But seriously...your description of transfer is correct for different temperatures, but that is not the situation we have with a processor. The processor doesn’t operate at a given temperature...it operates at a given energy output. Temperature rises and falls as necessary to transfer all the energy as heat. It MUST transfer all of the energy consumed by the processor, or the processor will burn up. That is why testers typically measure the total energy transferred to validate tests.

Quote:
Originally Posted by pauldovi View Post
Doubling flow did not cut dT in half. Why would this be expected to happen? The relationship is not linear.
Please review Fourier’s law. The relationship between mass and dT is linear. For q to remain unchanged, changes in mass require linear changes in dT, and vice versa.

Quote:
Originally Posted by pauldovi View Post
However, accuracy does not effect trends found in data.
Usually not. The problem here, however, is that your trend is completely within the accuracy of the instrument.
post #18 of 25
Thread Starter 
I don't see these discussions going anywhere anytime soon, so I figure its best to agree to disagree.
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
post #19 of 25
To help clarify energy, heat and thermal heat:

Energy comes in several forms: Electricity and thermal (heat) So when we say something dissipates 70W (W = VA, a measure of energy), we mean it dissipates 70W as energy in the form of heat. You honestly can't describe heat in any other matter, because temperature is different. For example:

If I put in 70W of energy as heat into a 1 Gallon container of water, and the same amount of energy as heat into a gallon of oil, theoretically the oil's temperature would be greater than the water due to water's higher specific heat.

Nice results BTW
Burning Phoenix
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Q9550 Gigabyte EP45-UD3P eVGA GTX 260 Core 216 SuperClocked 4GB G.Skill DDR2-1000 5-5-5-15 
Hard DriveOptical DriveOSMonitor
3 WD 80GB RAID0, Seagate 500GB, WD 1TB Caviar BLCK Lite-On LH-20A1S Windoze 7 Professional Dual Dell E207WFP 20.1" Widescreen LCD's 
KeyboardPowerCaseMouse
Logitech G15 Silverstone OP750 Lian-Li PC-A10B Logitech G5 
Mouse Pad
Harley Davidson 
  hide details  
Reply
Burning Phoenix
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Q9550 Gigabyte EP45-UD3P eVGA GTX 260 Core 216 SuperClocked 4GB G.Skill DDR2-1000 5-5-5-15 
Hard DriveOptical DriveOSMonitor
3 WD 80GB RAID0, Seagate 500GB, WD 1TB Caviar BLCK Lite-On LH-20A1S Windoze 7 Professional Dual Dell E207WFP 20.1" Widescreen LCD's 
KeyboardPowerCaseMouse
Logitech G15 Silverstone OP750 Lian-Li PC-A10B Logitech G5 
Mouse Pad
Harley Davidson 
  hide details  
Reply
post #20 of 25
Thread Starter 
Quote:
Originally Posted by Burn View Post
To help clarify energy, heat and thermal heat:

Energy comes in several forms: Electricity and thermal (heat) So when we say something dissipates 70W (W = VA, a measure of energy), we mean it dissipates 70W as energy in the form of heat. You honestly can't describe heat in any other matter, because temperature is different. For example:

If I put in 70W of energy as heat into a 1 Gallon container of water, and the same amount of energy as heat into a gallon of oil, theoretically the oil's temperature would be greater than the water due to water's higher specific heat.

Nice results BTW
Quote:
heat in any other matter
"lachen"

That really made me laugh! Hehe... In many ways that could be a pun!

I think we both understand the difference between heat and temperature.
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Is a stronger pump better?