-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
How big is too big?
- Thread starter gbaz
- Start date
Excess: thank you.
clocker2, we may have a differing terminology for the same thing in a car, but the main point of this topic is, please read your basic thermodynamics laws. You can't break these. You want the maximum flow, both in your waterblocks and in your rads. And all that with a pump producing the smallest possible heat load. BTW the new Hydor L35 is a very good improvement over the L30 in this regard, as it produces roughly the same flow/head with 30% less heat. Just an example.
clocker2, we may have a differing terminology for the same thing in a car, but the main point of this topic is, please read your basic thermodynamics laws. You can't break these. You want the maximum flow, both in your waterblocks and in your rads. And all that with a pump producing the smallest possible heat load. BTW the new Hydor L35 is a very good improvement over the L30 in this regard, as it produces roughly the same flow/head with 30% less heat. Just an example.
- Joined
- May 8, 2002
In an open system this would be true:
Try running in rain for five minutes and see if you don't get just as wet as if standing in it for five minutes.
Our systems aren't open, though. We aren't dealing with throwaway packets of water that leave the equation once they've passed the component in question. We have one continuous packet, so that "packet" is constantly in the radiator and constantly in the blocks, all at once.clocker2 said:
Try running in rain for five minutes and see if you don't get just as wet as if standing in it for five minutes.
- Joined
- Jun 26, 2004
- Location
- Westcountry, UK
Lol, great analogy.Sean Lindstrom said:
Neither of these "experiments" strike me as being particularly relevant to the conditions we are discussing. The frying pan trick is particularly egregious. Obviously, in a open system as you posit, the water can be converted to steam thus releasing heat more efficiently. My system doesn't get that hot, does yours? If the frying pan example were valid then the obvious first choice for a cooling system would be a waterfall, not a clunky and inelegant pump/radiator device.
Even a casual reading of my posts will reveal that I never said any such thing.
I said it was possible to overdo the flowspeed of the water, not that slower was better.
A nuance perhaps, but a critical one.
You said:
* physics experiments
* all thermodynamics school manuals (and, physics manuals)
* major scientists through the centuries, where do you think the formula Q=MCdT comes from ?
* real world measurements, which i pointed to you, but you failed to even begin to look at. Please do so, and come back with comments.
Now, *even* in an open system more flow will cool better, as more fresh water comes in the system during the same time interval.
PLEASE read:
* ANY kind of thermodynamics / thermics scientific manual
* http://overclockers.com/articles876/ (look at C/W vs flow rate)
* http://overclockers.com/articles1050/
and for rads:
* http://overclockers.com/articles778/ (PLEASE read EVERYTHING, even page 2 and following..)
All the data he's shown is very interesting but to illustrate my point i'll pick one:
And THAT is completely wrong, according to:clocker2 said:
* physics experiments
* all thermodynamics school manuals (and, physics manuals)
* major scientists through the centuries, where do you think the formula Q=MCdT comes from ?
* real world measurements, which i pointed to you, but you failed to even begin to look at. Please do so, and come back with comments.
Now, *even* in an open system more flow will cool better, as more fresh water comes in the system during the same time interval.
PLEASE read:
* ANY kind of thermodynamics / thermics scientific manual
* http://overclockers.com/articles876/ (look at C/W vs flow rate)
* http://overclockers.com/articles1050/
and for rads:
* http://overclockers.com/articles778/ (PLEASE read EVERYTHING, even page 2 and following..)
All the data he's shown is very interesting but to illustrate my point i'll pick one:
- Joined
- Apr 26, 2003
- Location
- Kenmore, NY
The thermostat's main job is to allow the engine to heat up quickly, and then to keep the engine at a constant temperature. It does this by regulating the amount of water that goes through the radiator. At low temperatures, the outlet to the radiator is completely blocked -- all of the coolant is recirculated back through the engine.
Once the temperature of the coolant rises to between 180 and 195 F (82 - 91 C), the thermostat starts to open, allowing fluid to flow through the radiator. By the time the coolant reaches 200 to 218 F (93 - 103 C), the thermostat is open all the way.
http://auto.howstuffworks.com/cooling-system8.htm
Once the temperature of the coolant rises to between 180 and 195 F (82 - 91 C), the thermostat starts to open, allowing fluid to flow through the radiator. By the time the coolant reaches 200 to 218 F (93 - 103 C), the thermostat is open all the way.
http://auto.howstuffworks.com/cooling-system8.htm
- Joined
- Mar 23, 2004
- Location
- Kansas
This thread is heading in the same direction as this one.
Keep this in mind (quoting myself):
In a closed system (edit: at steady state), the energy going in has to be equal to the energy leaving the system. This is true if your WC is at 20° or 2000000°. The cooling media, i.e, the water, block, rad, will adjust temperature until this condition is met. The reason is heat flows faster across a larger temperature differential. Therfore, the radiator and water block are in a balancing act; the radiator can move heat faster with hot water and the block can move heat faster with cool water. Using a low flow pump keeping the water in the radiator longer may very well cool the exiting water better realative to the incoming water, but remember the heat energy leaving the radiator has to be constant. So what you are doing is raising the equalibrium temp. In other words, less water has to remove more heat/time. More heat in less water = hotter water.
SureFoot, although I agree with you your graph does'nt support your conclusion.
Quoting myself again because I'm lazy:
Heat dissipation does not dictate the system temperature direcly. This data shows that this radiator can move more heat at a given temp at higher flow rates. This data does not show that higher flow rates lead to lower temperatures at a fixed dissipation rate (which is the case in a water cooled computer). I agree higher flow rates are better, this is just not the test that proves it.
This is my theory. clocker2 has his theory. My theory is based on physical laws while clocker2's is based on his own experiences. While I disagree with him I'm not going to say he is wrong without experimental proof. There is alot more physics involved in a wc setup then presented in this thread.
An experiment to evaluate temperature vs flow with a wc setup exposed to a constant heat input would be very nice proof of point. I don't know if any such experiment has been done.
Keep this in mind (quoting myself):
In a closed system (edit: at steady state), the energy going in has to be equal to the energy leaving the system. This is true if your WC is at 20° or 2000000°. The cooling media, i.e, the water, block, rad, will adjust temperature until this condition is met. The reason is heat flows faster across a larger temperature differential. Therfore, the radiator and water block are in a balancing act; the radiator can move heat faster with hot water and the block can move heat faster with cool water. Using a low flow pump keeping the water in the radiator longer may very well cool the exiting water better realative to the incoming water, but remember the heat energy leaving the radiator has to be constant. So what you are doing is raising the equalibrium temp. In other words, less water has to remove more heat/time. More heat in less water = hotter water.
SureFoot, although I agree with you your graph does'nt support your conclusion.
Quoting myself again because I'm lazy:
Heat dissipation does not dictate the system temperature direcly. This data shows that this radiator can move more heat at a given temp at higher flow rates. This data does not show that higher flow rates lead to lower temperatures at a fixed dissipation rate (which is the case in a water cooled computer). I agree higher flow rates are better, this is just not the test that proves it.
This is my theory. clocker2 has his theory. My theory is based on physical laws while clocker2's is based on his own experiences. While I disagree with him I'm not going to say he is wrong without experimental proof. There is alot more physics involved in a wc setup then presented in this thread.
An experiment to evaluate temperature vs flow with a wc setup exposed to a constant heat input would be very nice proof of point. I don't know if any such experiment has been done.
Last edited:
Mea culpa.
My statement that the water must linger in the radiator long enough to dissipate heat is wrong.
The overwhelming mass of thermodynamic law says so.
It was an imperfect and incorrect way to explain a phenomonon regularly encountered at the racetrack.
The explanation was wrong, the observed results are not.
Higher flow rates (i.e. unrestricted radiator intake), which intuitively should result in lower temps, do not always do so. there is a definite sweet spot, beyond which temps will again rise.
Surely pump heat can be a factor, but I also suspect other contibutory effects also.
Perhaps turbulance degenerates into cavitation or there is an increase of boundary layer insulation, I don't know.
Book learning quickly becomes irrelevant when the rubber meets the road.
To the original thread question "How big is too big" I now say that I have no idea other than there IS a "too big".
Perhaps in a perfect theoretical water loop this hold true.
AFAIK, no such creature exists.
My statement that the water must linger in the radiator long enough to dissipate heat is wrong.
The overwhelming mass of thermodynamic law says so.
It was an imperfect and incorrect way to explain a phenomonon regularly encountered at the racetrack.
The explanation was wrong, the observed results are not.
Higher flow rates (i.e. unrestricted radiator intake), which intuitively should result in lower temps, do not always do so. there is a definite sweet spot, beyond which temps will again rise.
Surely pump heat can be a factor, but I also suspect other contibutory effects also.
Perhaps turbulance degenerates into cavitation or there is an increase of boundary layer insulation, I don't know.
Book learning quickly becomes irrelevant when the rubber meets the road.
To the original thread question "How big is too big" I now say that I have no idea other than there IS a "too big".
I still disagree with this assertion, Laws of Thermodynamics or no.
Perhaps in a perfect theoretical water loop this hold true.
AFAIK, no such creature exists.
Last edited:
Oh please. You didn't even read my posts. That's EXACTLY what i said, in a watercooling loop the pump heat input will make diminishing returns and even losses at high powers. Oh, read above...clocker2 said:
And i said: "higher flow -> better HEAT TRANSFER", please read exactly.. This is always true, at least with liquids and gases. If not, please disprove the heat transfer equation Q=MCdT (to stay simple)..
As various people told you there: the thermostat is meant to HEAT up the engine to proper temperature. When the engine needs proper cooling, the vanes are full open for maximum water flow. Hey, would they be wrong at howstuffworks, and my car mfger would be wrong too ??? How's that for "real life" Mr. "i don't believe in the laws of Physics and i even call them theories.."clocker2 said:
What about Bill Adams article ? Did you even read it ? I know the graph i posted doesnt fully support my point, but PLEASE read ALL the article, you'll see it does. WITH A CONSTANT HEAT INPUT, that's what the article is all about, a higher flow rate leads to higher heat transfer. There's no "sweet spot", that has been demonstrated already, IIRC it was Bill who found about it, and told it was due to other external factors. Subsequent testing of air<->water heat exchangers proved that point.
(edit) there is indeed a point where flow will lead to increasing temps - when water starts to boil out of friction. What is required ? 10000gph ? 100K gph ? more ? and even so, pressure would push the boiling point upwards. Cavitation can lead to premature boiling too, but again one needs a very very strong pump, and rigid tubing, as the soft tubing we use in our WC setups would collapse long before it happens...
Last edited:
Of course I did. I even quoted it. Which you have deleted.
Your manipulation of my post is somewhat duplicitous, don't you think?
Just so.
Various people who didn't comprehend my posts obviously.
I was talking about restrictor plates in race machine water systems and used the common thermostat ( a device I'm still not sure that you comprehend), as an example.
In a road car yes, it regulates flow so as to optimize warm up and stabilize systemp...it's secondary funtion is act as predetermined restrictor to flow, as, even when fully open, it's orifice is not as large as the passage it obstructs.
This fact requires no charts or scientific foldorol to document...$6 at your local AutoZone will buy you the proof.
What a luxury...having one's cake AND eating it too.
Anyway, is cavitation a phenomonon somehow unable to occur in a radiator or waterblock?
It certainly can happen in an automotive radiator and block.
Contrary to your guess, a typical automobile waterpump flows at the rate of 20-80L per minute which isn't so different than the systems we are using is it?
In the case of a PC, cavitation which leads to boiling is hardly a danger, but cavitation which prevents a good interface between the water and the metal is.
Last edited:
clocker2
its a loop
the water is not staying in the loop any longer than it is with a less restrictive radiator
as soon as the water leaves, there is more water to replace the water that's not out of the system
surface area dissapates heat, and the efficency of the system is inherently dependant on the ability of the heat to leave the water and get dissapated into the air
clocker2, you're right about that the longer the water stays in the radiator, the more cool it will get, as in more heat dissapated, but with less passes in the radiator, it does not mean that the water is exposed to any less surface area, because more water with more excess heat is just replacing the water thats existing the radiator
i don't believe that surefoot is talking about the water and therefore coolant temps in general, but the resulting temp at a certain point, e.g. the waterblock and the resulting lowered temperatures from which more flow and therefore its impact on the *WATERBLOCK'S PERFORMANCE* is
you are right about the coolant temps and its effect on the engine
however though, i think you guys misunderstand each other
1. the engine might not benefit as much from a higher resulting coolant system from as the waterblock
e.g. the engine is restrictive
e.g. the engine is designed for lowerer flow, and therefore can deal with lower flow and still stay cool *enough*
e.g. the coolant pump has come exotic specs for pressure and flow, and it puts out more head than them aquarium pumps we're used to
SureFoot, i don't think he knows or cares about who "bill adams from swiftech" or "cathar and his fluid thermodynamics" is and what they have to say and so on, he lives in the "car" world and could care less
you can't win the argument by citing other people from the small world of computer cooling products and so on, or even graphs from the internet or a chart from thermochill or anything, he doesn't care
but instead you should focus on explaining the loop thing, where if lets say that per gram of water dumps off .005 w of heat at even cm^2, obviously if there's more water or an increase of surface area would make more water dump more heat, right?
so now, either with more flow the surface area has effectively increased or with more surface area the effective flow has increased, right?
so therefore in his car world where the flow is always the same, regardless of restriction, and where there is a point where lower temperature are not better, you can't win the argument by telling him that more flow = better
but you can convince him that aquarium pumps are bad stuff and therefore we NEED single pass radiators and high flow to achieve the lowest temperatures
because as the water dumps less heat by passing through the radiator faster, more water with more heat is now able to enter the radiator and therefore dump more heat
but per mass of water will be able to dump less heat but there will be more water
and THEN go quote cathar or something
don't start by quoting cathar, he might not know or care about cathar
_
its a loop
the water is not staying in the loop any longer than it is with a less restrictive radiator
as soon as the water leaves, there is more water to replace the water that's not out of the system
surface area dissapates heat, and the efficency of the system is inherently dependant on the ability of the heat to leave the water and get dissapated into the air
clocker2, you're right about that the longer the water stays in the radiator, the more cool it will get, as in more heat dissapated, but with less passes in the radiator, it does not mean that the water is exposed to any less surface area, because more water with more excess heat is just replacing the water thats existing the radiator
i don't believe that surefoot is talking about the water and therefore coolant temps in general, but the resulting temp at a certain point, e.g. the waterblock and the resulting lowered temperatures from which more flow and therefore its impact on the *WATERBLOCK'S PERFORMANCE* is
you are right about the coolant temps and its effect on the engine
however though, i think you guys misunderstand each other
1. the engine might not benefit as much from a higher resulting coolant system from as the waterblock
e.g. the engine is restrictive
e.g. the engine is designed for lowerer flow, and therefore can deal with lower flow and still stay cool *enough*
e.g. the coolant pump has come exotic specs for pressure and flow, and it puts out more head than them aquarium pumps we're used to
SureFoot, i don't think he knows or cares about who "bill adams from swiftech" or "cathar and his fluid thermodynamics" is and what they have to say and so on, he lives in the "car" world and could care less
you can't win the argument by citing other people from the small world of computer cooling products and so on, or even graphs from the internet or a chart from thermochill or anything, he doesn't care
but instead you should focus on explaining the loop thing, where if lets say that per gram of water dumps off .005 w of heat at even cm^2, obviously if there's more water or an increase of surface area would make more water dump more heat, right?
so now, either with more flow the surface area has effectively increased or with more surface area the effective flow has increased, right?
so therefore in his car world where the flow is always the same, regardless of restriction, and where there is a point where lower temperature are not better, you can't win the argument by telling him that more flow = better
but you can convince him that aquarium pumps are bad stuff and therefore we NEED single pass radiators and high flow to achieve the lowest temperatures
because as the water dumps less heat by passing through the radiator faster, more water with more heat is now able to enter the radiator and therefore dump more heat
but per mass of water will be able to dump less heat but there will be more water
and THEN go quote cathar or something
don't start by quoting cathar, he might not know or care about cathar
_
a graph is only as meaningless as anything else which is said be anyone else
so for the title of the thread, there is no "TOO BIG"
for the pump or the radiator or the waterblock
BIGGER IS BETTER
just think about it, wouldn't you want a BIG-ger paycheck?
even if it indicates only the physical size of it, wouldn't bigger be better still
i once saw an argument where scientists were able to use all kinds of stuff to prove that bumble bees could not fly, but yet bumble bees DO fly
it doesn't matter if our rules and our laws don't apply to others in oother places, only worry about HERE, and NOW
so if higher flow = better temps for your PC, GOOD, you don't need to convince the people draining the hoover dam that more flow = better temps, they don't care and don't need to know
so for the title of the thread, there is no "TOO BIG"
for the pump or the radiator or the waterblock
BIGGER IS BETTER
just think about it, wouldn't you want a BIG-ger paycheck?
even if it indicates only the physical size of it, wouldn't bigger be better still
i once saw an argument where scientists were able to use all kinds of stuff to prove that bumble bees could not fly, but yet bumble bees DO fly
it doesn't matter if our rules and our laws don't apply to others in oother places, only worry about HERE, and NOW
so if higher flow = better temps for your PC, GOOD, you don't need to convince the people draining the hoover dam that more flow = better temps, they don't care and don't need to know
- Joined
- Apr 21, 2003
- Location
- New Jersey
fafnir said:
Lmfao, that's awesome, and sounds VERY familiar.
Fafnir,
You have made several false assumptions, both about me and the "car world" also.
Of course I care about watercooling as applied to PCs...my PC is watercooled and I'm here to learn and share information/knowledge too.
Simply using a (somewhat imperfect) analogy using automotive technology does not automatically consign me to the ranks of those who are unable to comprehend the nuances of computer cooling.
I see that you are another unrepentant "bigger is better" adherent.
Early on in this exchange SureFoot said:
This is also valid with water.
One trait that I am proud to have carried over from the "car world" (specifically racing vehicles) is the ability to abandon theory ( or, if you prefer, "scientific laws") when it contradicts physical reality.
If my observed results show that higher flow does not necessarily translate to better temps, then no number of charts, reputable experts or science books will convince me otherwise.
You may see this as being stubborn, or even blind.
I would consider it to be open-minded without being gullible.
Such a situation does not mean that one must nullify a particular Law, either.
Couldn't it as easily be explained by another ( equally valid) Law entering the equation? The synergy between various physical laws negates none of them and leads to the practical applications that we actually build and use.
You have made several false assumptions, both about me and the "car world" also.
Of course I care about watercooling as applied to PCs...my PC is watercooled and I'm here to learn and share information/knowledge too.
Simply using a (somewhat imperfect) analogy using automotive technology does not automatically consign me to the ranks of those who are unable to comprehend the nuances of computer cooling.
I see that you are another unrepentant "bigger is better" adherent.
Early on in this exchange SureFoot said:
Having spent the better part of the last year devising and tinkering with maybe a dozen different air configurations I believe that yes, after a certain point, lower flow is better.
This is also valid with water.
One trait that I am proud to have carried over from the "car world" (specifically racing vehicles) is the ability to abandon theory ( or, if you prefer, "scientific laws") when it contradicts physical reality.
If my observed results show that higher flow does not necessarily translate to better temps, then no number of charts, reputable experts or science books will convince me otherwise.
You may see this as being stubborn, or even blind.
I would consider it to be open-minded without being gullible.
Such a situation does not mean that one must nullify a particular Law, either.
Couldn't it as easily be explained by another ( equally valid) Law entering the equation? The synergy between various physical laws negates none of them and leads to the practical applications that we actually build and use.
- Joined
- Dec 20, 2003
- Location
- London, England
fafnir said:
err - no. if it were just the physical size of it then smaller=better (its easier to keep in my wallet that way) - similarly i am very happy with the performance of my wmd-30RZs, but i would prefer the same performance from a smaller pump...
@LV38_Eagle
actually almost the opposite is the case - the flow rates at which diminishing returns are approached for the rad (especially with good airflow) is considerably higher than the kind of flows we could force through most blocks - a 2-pump solution could give good results but by forcing a higher flow through the rad.
see Joe Camel's thread here for some discussion of running with 2 loops
http://www.ocforums.com/showthread.php?t=277159
- Joined
- Nov 30, 2002
- Location
- California
clocker2 said:
no, pump heat is the only factor.
according the laws of thermodynamics, conduction and convection, heat transfer rate increases (with fixed material properties) only by the difference of temps of each surface of each median. If u allow water to absorb more heat, it will heat up, decreasing the difference between the two surfaces, decreasing heat transfer. Thus moving water faster keeps the heat difference high, cooling things better.
And just to settle this, convection laws are similar for water and air. Only difference are the thermal and material properties. If your theory was true, putting to much air through ur heatsink would get bad results, many people would disagree with that.
- Joined
- May 8, 2002
Never heard a straw man say that before.clocker2 said:
Isn't it enough?
