Flowrate...

[Penguin4x4]

Around 150-300

 

[wildfrogman]

to GreenmanWD-40

I for the most part understand your response about neo86 shortening his tubing and then getting higher temp. With shortening the tubing you would not have an increase in tube flow resistance as long as all tubing is situated the same and not like mostly vertical intead to fit in the case. As you already knew shorting the tubing wasnt the reason for his temps going up like it did. It is my opinion that a few people read what you said about reduced water in the system from shortening the tubing and figured ehh 30 inches? of 3/8ths or so isnt that much water loss and amount of water only matters how fast the system will stabilize thing. Not considering that the extra loss of water was more than just volume loss of the shorted tubing. Introduced bubbles as it reduced the water enough to not fill the closed loop water cooling system which as you maybe already knew but didnt say so people took it out of context. Also yes IF you have a tube of water that is REALLY LONG it would slow down but...Not taking into acount that neo86 shortened not only the tube that were horizontal BUT also vertical as well if i read the post correctly so the tube sidewall resistance would be a bit lower at most around half rather than higher than before right? That is my opinion and probably right but wont be upset if somebody disagrees with me. Just i would like more than "because i say so" response. I just would like a good reason so i will be able to say ok that makes sense, then maybe test it out myself~understand exactly where i went wrong if and when i do. Give me some facts that i may have overlooked or such. I have been a long time lurker of the forums but this finally got me to register. Ohwell i think it just boils down to a misunderstanding of what really would help neo~what really is the real problem, how long, how much head~how the tubing is situated, what kind of pump, fan cfm on radiator, airflow restriction from stamped case fan grilles, airflow for case intake case maybe really bad~added heat etc which could have influenced results and answers for help would have been effected as well.

 

[FDUSMC]

I reduced the length of my hose by about 4 feet and so no change in temp whatsoever.

 

[h2sammo]

well, thats one thing.
how about change in flow velocity?
thats what were arguing about

 

[FDUSMC]

I have no way in determining that. I figured it would be higher being that the tubing is shorter.

 

[wildfrogman]

Ah thanks for pointing that out again, i didnt put that in my post though i should have. Short hose runs do help water flow better, as long as the hose has no sharp bends~lots of right angles etc to make it shorter. If your case needs an extra 5~10 inches etc to have a sweeping turn rather that sharp the smooth turn would be better. One thing that i remember from somewhere reading was that not only does the waterblock get more heat taken out and put into the water at high flow but also the higher water velocity would also spread the heat accross the radiator~heat exchanger better. But if you do not have enough airflow through the radiator it doesnt really help much. You do have to have a certain amount of airflow through certain radiators for them to hit their peak performance. Take the Serck radiator, BIG but needs alot more airflow than the "Big mamma" heater core to perform as well. Unless you like to run two 172mm widebody nidec 230ish cfm fans on a radiator the big mamma heatercore would actually perform better if both were used with just one of those big fans. I could be mistaken but the low pressure drop ~less restrictive and the construction of the Serck was for super high airflow while the heater core was more tuned for a lot less cfm and still perform really well.

 

[johnny007]

Simple Physics

I am new to this forum but interested in this subject. I have a dual xeon p4 system and the fan noise is a damned nuisance. I am keen to find a noisless (and practical) solution to cooling this thing.

I think the emphasis on flow rate is somewhat less important than many seem to be implying. The negative effect of greatly increased flow rates is increased friction at surfaces the water is flowing over thus creating heat on these surfaces. I see high flow rates causing turbulent flow, also negative, as opposed to laminar flow for efficient dynamics.
It isn't a closed system, there are too many variables. The critical cooling system devices are the heat transfer efficiency of the interface between the heat source and the coolant and of course the efficiency of the heat dissipation unit.

The key is that the temperature differential between the heat source and the coolant be as high as possible and the heat transfer between them as efficient as possible. To achieve this would perhaps require a huge external radiator type heat sink or an electrical refrigeration unit. I reckon great importance should be attached to the design of the heat exchanger attached to the various heat sources also.

Insulated pipes on the cool side of the system may have a significant effect also.

It is a shame there aren't mechano type heat sinks and plumbing pieces to attach to devices and construct a complete customised pipe network from preformed sections, hmmm i wonder....

I have a clear idea of what I require but see no easy and practical way of achieving it.


John

 

[johnny007]

I reduced the length of my hose by about 4 feet and so no change in temp whatsoever.

The volume of coolant (between the devices) is not really a significant factor - the efficiency of the heat transfer to the coolant and the heat dissipation unit itself is where the cooling work is done. Try insulating the cool side of the system. The liquid in the system is simply a conduit between the devices (the pipes are simply containers)

 

[h2sammo]

Welcome to the Forums

Why would you wanna insulate anything in the system?

We want to ease heat travel to cold sides. If we insulate the cold side, theh heat transfer would be prohibited...no?

 

[wildfrogman]

The reason why you should insulate the water pipes etc inside the case is to keep the case as cool as possible and take care of the heat outsidethe case. A radiator has air blown onto it from inside case and if air in case is warm it wont perform as well as if it was cooler. So pretty much you would still have roughly the same temps for your cpu...but your case temp would possibly be higher if you didnt insulate the tubes, thats all.

 

[johnny007]

Insulation keeps in cold as well as heat

Perhaps I wasn't clear enough in saying what I meant - the pipes carrying cold coolant FROM the radiator TO the heat exchange device - if insulated - will help to maintain the low temperature of the coolant slightly - especially if the ambient air temperature is quite high. Why do you think household refrigerators, or any other type for that matter, have thick insulated casings - it is to keep the heat out and maintain the cooler temperature within.

What we are trying to acheive is a maximum differential between the temperatures of the coolant entering the heat exchanger and that leaving it.


John

 

[wildfrogman]

Ok that makes sense now, insulating the exit tubes would help also, making the case temp less because of less heat loss inside the case even a tiny bit would result in air a bit cooler air hitting the radiator rather than the air in the case get warm.

 

[h2sammo]

the pipes carrying cold coolant FROM the radiator TO the heat exchange device - if insulated - will help to maintain the low temperature of the coolant

aha

 

[FizzledFiend]

OK I don't liquid cool yet, but I got a silly question that may or may not have been answered. I don't know everything about the science of liquid cooling, but I am a mechanic and I do deal with heat and water cooling ...even air cooled (VW, some motorcycles, ETC.) anyways I know we aren't dealing with 280+ degree F temps here, but...from what I did read here is, that if we could get constant temps to the CPU the better we are off. Correct? if yes then, if the temps rise and fall here then we need to regulate the flow to find that sweet spot. OK is this sweet spot for full loaded CPU temps or average temps? or does that matter?. What I am leading up to here is a thermostat...cars use them for various reasons...basically to keep the engine at around 280 degrees F (your conditions may vary ) I know the CPU dosen't have an optimal temp other than cooler is better, but if temps were a constant on a CPU wouldn't it make it better? Just a thought...and I probably don't even belong in here sense I don't liquid cool yet ( I can dream LOL) just an Idea I was bounceing around one day and I know what problems I have come to in my mind......the main one is 'Where in gods name am I going to find a thermostat to meet my requirements', and 'would it be worth it?'

 

[johnny007]

My primary reason for looking into water cooling is to try and silence the damned CPU fans. The dual cpu means these fans are noisy (high pitched whistle). I am still not convinced this sweetspot flow rate argument is of any great importance.

The main impotant factors are the thermodynamic properties of the coolant, the efficiency of the heat exchangers and the efficiency of the radiator.

The coolant within the system obviously needs to flow but the absorbtion of heat at the CPU and the dissipation of heat at the radiator has more to do with thermodynamic properties of the coolant and the size of the surface area it is moving over than the rate at which it does so (within reason obviously). I just don't think flow rate is a big significant factor compared to the efficiency of the devices it is flowing through.

The thing to look at is in energy terms, the cpu temperature is putting heat energy into the coolant at a relatively fixed rate determined by the efficiency of the heat sink device and the radiator is taking it out of the coolant at a relatively fixed rate determined by surface area, ambient temperature and any assisted cooling from fans.

I don't think the thermostat argument holds any weight. In a car it is designed to kick in at a given temperature to prevent engine over-heating and keep the engine running at an optimum temperature range. In a PC the ideal cpu temperature is absolute zero so really there can never be enough cooling, hence a thermostat will really defeat the object. The desired effect is to achieve maximum cooling at all times.

 

[bigben2k]

Here's a crazy idea:

since we're all concerned with getting a pump to achieve a higher flow rate, to get more flow speed, why don't we try to find out what the height of a channel needs to be, in order to transfer heat from (say) a copper surface, to different speeds of coolant?

I mean, wether the channel is .25" high, or 1" high, we can push for a higher flow speed, but the problem is that the latter would require a much larger pump, which would in turn induce more heat.

Anyone?

 

[garasaki]

Re: Simple Physics

It isn't a closed system, there are too many variables.

Um, the definition of closed system is a system that does not allow mass transfer across its boundary. Most WC systems fit that description and are closed systems. It has NOTHING to do with the number of variables involved.

I think the emphasis on flow rate is somewhat less important than many seem to be implying...The critical cooling system devices are the heat transfer efficiency of the interface between the heat source and the coolant and of course the efficiency of the heat dissipation unit.

You're contridicting yourself here. Heat transfer efficiency is influenced greatly by flow rate. If you never have, it would be a good idea for you to go pick up a heat transfer book at your local library and read up on convection...

The key is that the temperature differential between the heat source and the coolant be as high as possible and the heat transfer between them as efficient as possible

You're certainly right about both these points, but you should note that the heat transfer is affected by flow rate. Therefore, the arguement about the optimum flow rate is indeed important!

Insulated pipes on the cool side of the system may have a significant effect also.

I don't personally believe that to be true. How much heat do you really think escapes the tubing, which is made of plastic (an insulator) and is of relatively short length. The main reason you see insulation on pipes in buildings (wether they be plumbing or HVAC related) is to eliminate condensation, NOT to keep the material inside at a certain temperature. Water in the walls of a building is a very bad thing.

It is a shame there aren't mechano type heat sinks and plumbing pieces to attach to devices and construct a complete customised pipe network from preformed sections, hmmm i wonder....

What do you mean exactly? How hard is it to use plastic tubing to connect the devices?? I haven't heard of anyone using copper piping, but it could be done. You'd have to be pretty patient though....

 

[johnny007]

Re: Re: Simple Physics

Um, the definition of closed system is a system that does not allow mass transfer across its boundary. Most WC systems fit that description and are closed systems. It has NOTHING to do with the number of variables involved.

OK a little less pedantics wouldn't hurt. The only true closed system is the universe. It has nothing to do with mass and everything to do with energy. Energy in the form of matter maybe. A closed system is completely independent of it's environment, are you saying a water cooled system inside a PC is that?
Just because the coolant is completely contained doesn't mean it is closed in the energy system sense. Now there are many variables and all have a varying effect on the cooling capacity of the system. Yes?

You're contridicting yourself here. Heat transfer efficiency is influenced greatly by flow rate. If you never have, it would be a good idea for you to go pick up a heat transfer book at your local library and read up on convection...

Pedantics again, I said flow rate is not as critical as some are suggesting. If you've got the bit between you're teeth and aren't going to let go then the discussion is over.
The thermodynamic properties of the coolant itself are critical, the design of the heat sink is also. It will determine the uptake of heat into the coolant largely by maximising the surface area of the coolant channels within the heatsink.

You're certainly right about both these points, but you should note that the heat transfer is affected by flow rate. Therefore, the arguement about the optimum flow rate is indeed important!

I never disputed the fact that a flow rate is important, I said it wasn't as critical as some are making out. The thermodynamic properties of the coolant and the design of the devices doing the cooling are much more critical.

I don't personally believe that to be true. How much heat do you really think escapes the tubing, which is made of plastic (an insulator) and is of relatively short length. The main reason you see insulation on pipes in buildings (wether they be plumbing or HVAC related) is to eliminate condensation, NOT to keep the material inside at a certain temperature. Water in the walls of a building is a very bad thing.

No heat escapes from the tubing. On the cool side the heat of the pipe wall from the ambient air temperature is being drawn in and will have a slighty detremental effect on the coolant temperature, it may well be very small. Insulation was just a suggestion. If the cooling efficiency of the system were too great then condensation would be a huge problem. The pipes and heatsinks would require insulating on all exposed surfaces.

 

[garasaki]

Re: Re: Re: Simple Physics

A system that does not exchange mass or energy with its surroundings is an isolated system (this is what you are refering to). A closed system is a system which does not exchange mass with its surroundings, but is free to exchange energy to it's heart desire.

The thermodynamic properties of the coolant itself are critical, the design of the heat sink is also. It will determine the uptake of heat into the coolant largely by maximising the surface area of the coolant channels within the heatsink.

I never disputed the fact that a flow rate is important, I said it wasn't as critical as some are making out. The thermodynamic properties of the coolant and the design of the devices doing the cooling are much more critical.

Ok, great, you know how to use the phrase "thermodynamic properties". Do you know what it means??? Do you realize that the flow rate also plays a critical part in determining the "uptake of heat" into the cooling liquid?? I hate to sound like Bill A here, but it seems to me that you're missing that rather important point. Seriously, go do some reading about convection. And more to the point, there are very few coolants for us to choose from. I mean, really, are there options besides water?? Not reasonable ones. So you can't go around saying that "the thermodynamic properties" of the coolant are critical, because they just don't vary from one system to another. In fact, if you compare most coolants of WC systems used by OCers, the only difference in the "thermodynamic properties" is caused solely by flowrate!!

As for the design of the devices, yes, that is very important. No disputing you there. I just don't see where you are seeing a huge variation in "the thermodynamic properties" of the coolant...

 

[BillA]

This post deleted by its "derogatory and insulting" author so that it will not be (so) necessary
"to defend the little guys against the tyranny of the know-it-alls."