but another noob question from an old man

why is it that there is so much importance put on gallons per hour in wc systems????in my experiance in the automotive field there is such a thing as to much flow,look at a modern extreme performance engine and you will find that instead of using a thermastat to control coolant temps ,it has been replaced with a flow restricter (basicly a big washer) to slow down the coolant to allow it to collect as much heat as possible while it is inside the engine(cpu cooling block).kinda makes me wonder

yes it'll collect more heat, but a cpu block is different then an engine. Cpu blocks differ among each other as well.
Some blocks are restrictive and some are not.
a car engine isn't restrictive (or atleast its not comparible with a WWR so to say).
So why make waterblocks restrictive. Well, it turns out that restriction at the water block actually allows better convection. More friction with water/copper or aluminum allows the water to absorb more heat.

Grampa said:

look at a modern extreme performance engine and you will find that instead of using a thermastat to control coolant temps ,it has been replaced with a flow restricter (basicly a big washer) to slow down the coolant to allow it to collect as much heat as possible while it is inside the engine(cpu cooling block).kinda makes me wonder

Click to expand...

There is a temp range in which engines work best. The thermostat tries to keep them in that range.

Cpus just like to be cold.

thats about what I was saying,however you start with an engine that has intended rpm range,once you excede that range the design of the cooling passages become either your best friend or your worst enemy,cooling some areas more of less than others.That is why they put several temp sensors in different spots so they can make adjustments in flow regulation to the areas that need either more or less flow(nb block,cpu block, graphics card block ,hdd)...the flow rate going to different blocks is not (in my opinion)a" one flow rate does the best job"situation.does this sound right(maybe I need to make sure I took all my meds this morning)

Re: Re: but another noob question from an old man

Giblet Plus! said:

There is a temp range in which engines work best. The thermostat tries to keep them in that range.

Cpus just like to be cold.

Click to expand...

...I agree.but when you run the water through a block too fast it will not absorb(or you could say that its ability to make the most of the convection process) all that it would be capable of if it was slower,or perhaps it needs to be faster because it has absorbed as much as it can and instead of carrying it off it keeps it in the block too long,,,,hhhuummmmmm....I would like to hear anything any of you have to say on this,I am not one to just do as others have done before me, there is always something that someone has missed and can be improved upon.

From what I understand, if the water is moving quickly, the water will not absorb all of the heat it can carry, but the same amount of heat will be removed as compared to if the water was moving at the right speed to use all of its thermal capacity. Too slow would be a problem, as the amount of heat would be in excess of the water's thermal capacity, but too fast should not be a problem.

with all due respect,that is a good theory,but if it is going to fast through you radiator then it has lost it ability to to convect that heat it has absorbed into the radiator,,(again these are just the ramblings of an old man with hardly any practical comp experiance,as it applies to sys building,or water-cooling)I do realise that once the coolant reaches the rad. it will slow a bit because it has reached an area that has more volume than a hose or a block,but is it the optimum speed of flow...take what you have and make the absolute most of it...I guess its time for me to buy a couple of thermal probes and get my wc rig built so I can start playing with these thoughts

Very good reading here on this very subject. Flowrate summary

bad link,try it again,,,I need all the help I can get,,,thanks

This is a subject that has been the topic of MANY discussions, some of them quite heated, even if the cpu wasn't.

It seems that to fully understand all the variables requires an education in thermodynamics and maybe a few others; or at least some intense study and a bunch of math that I just don't have.

I'll attempt a VERY simplistic explanation of my understanding:

Higher flow rates work better (not necessarily gph rating because that is a flawed spec.) because it allows better heat transfer in the water block. The radiator can only cool the water so much, but with a higher flow rate, it effectively allows it to cool MORE water.

There a so many variables and differences of opinion on the subject that it gives me a headache, but my own experience has shown me that this invariably holds true- different systems will show varying degrees of change between high-flow and low-flow, but a higher flow system allways has more cooling potential THAN THE SAME SYSTEM AT A LOWER FLOW RATE.

Seriously- I have been involved in more than a few discussions on the subject and still do not understand the science of it, but there are at least a couple of threads here you can check out-
http://www.ocforums.com/showthread.php?s=&threadid=78055
http://www.ocforums.com/showthread.php?s=&threadid=210844

When first I began planning my water cooled system I struggled with these same questions and notions as you Grampa. As it turns out, my first intuitions, similar to yours on these matters, were wrong.

I'm a bit short on time (need to get back to work), so I'll leave a couple links to some threads that really helped me along here. Simple put, all other things being equal, increasing the flow rate will increase the rate of heat transfer.

The links below are in the Water Cooling Stickies section of the forums, under Water Cooling. The first is the original thread, and the second a summary thread of the original. The posts on the first page of the original thread by RhoXS really helped me to understand where my intuitions let me down.

Links:
http://www.ocforums.com/showthread.php?s=&threadid=78055
http://www.ocforums.com/showthread.php?s=&threadid=210844

Hope you find some useful information here, some good reading at least.

Heheh, beat me to it rogerdugans

I thank you all for your help in understanding a new(to me)type of hot rod(like it or not that is what these systems that we build indeed are )give me a little time and I will have pics of how I apply the theorys I have to water cooling and case mods...thanks,and if anyone wants to add more to this thread I would be gratefull.....

One of the easiest, best things to keep cpu temps down is to use the largest heatercore you can stuff into the case (or external box) and move a reasonable amount of air through it. A heatercore happens to be about the best combination of surface area and low flow resistance for a given space. Trying to compensate by using a smaller radiator with very high airflow (and fan noise) doesn't work as well.

As far as gallons-per-hour pump rating, that's not the important spec unless the system has a very low flow resistance. Computer water systems are relatively high-resistance to the pump so the important spec is head pressure measured in feet or meters. Low pressure aquarium pumps have about 4 feet of head. Typical aquarium pumps have about 6 feet of head which is enough for a reasonable setup IMHO. Among the best (and popular) aquarium pumps is the Danner mag 3 with 10 feet of head. Higher head pressure generates higher flow rates which helps some water blocks more than others, but higher flow certainly doesn't hurt.