• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Pump Head?

P

PiratePowWow

Limp Gawd
Joined
Apr 19, 2003
Messages
450
What is this head thing all about. Apparently as head increases, flow decreases.

Hmm, I got to thinking about this.

What is head?

Is it vertical feet, or horizontal feet?

Or is it tube length?

It makes sense to me that it should be vertical feet.

I also got to thinking about flow rate in general.
Most people seem to like their high gph pumps like over 300 gph, but I am thinking only about 60 or so gph is necessary and will cool just as well.

WHY?
Well, if the water is moving slower, won't it spend more time cooling down in the radiator before it gets to the block?
And if the water is moving faster, it will spend less time cooling down in the radiator.
I think the end results will be the same.

Low gph pump theory:
You may say "weak pumps bog down with long tubes".
I doubt it. in fact, I think low gph pumps will run effortlessly simply because of the syphon properties of the water. The water that has passed through the block and begins it's decent back to the reservoir is pulled by gravity which causes suction on the water behind it in the tube still traveling upwards to the block.

Gravity acts against the pump on one side and for it on the other, and the two forces balance out leaving the pump to do its job effortlessly.
 
Head is how high the water can be pumped vertically before the pump cant push any higher.

as far as flow, you want fast flow in the block, and slow flow in the rad if at all possible. this ensures the CPU only sees cool water, and the rad gets maximum time to cool off the hot water.

and why we dont go for low flow is that waterblocks and rads, etc add restriction to the flow, so that 80 may be more like 40 when your done. And 40 isnt all that good really.

300gph is a nice suze to go with, it's a bit large, but not overkill. Plus if your loop is restrictive, the added flowrate goes a long way.
 
Originally posted by PiratePowWow
What is this head thing all about. Apparently as head increases, flow decreases.

Hmm, I got to thinking about this.

What is head?

Is it vertical feet, or horizontal feet?

Or is it tube length?

It makes sense to me that it should be vertical feet.

I also got to thinking about flow rate in general.
Most people seem to like their high gph pumps like over 300 gph, but I am thinking only about 60 or so gph is necessary and will cool just as well.

WHY?
Well, if the water is moving slower, won't it spend more time cooling down in the radiator before it gets to the block?
And if the water is moving faster, it will spend less time cooling down in the radiator.
I think the end results will be the same.

Low gph pump theory:
You may say "weak pumps bog down with long tubes".
I doubt it. in fact, I think low gph pumps will run effortlessly simply because of the syphon properties of the water. The water that has passed through the block and begins it's decent back to the reservoir is pulled by gravity which causes suction on the water behind it in the tube still traveling upwards to the block.

Gravity acts against the pump on one side and for it on the other, and the two forces balance out leaving the pump to do its job effortlessly.
Click to expand...


You can find answers to some of your questions here and here .

More flow=better, and generally higher head=better than higher gph. For the purpose of water cooling, the pump needs to be able to deal with restricted flow, so a pump with a high-head rating and lower gph will often outperform a low-head high-gph pump, when used in a watercooling setup.

One also has to consider the wattage rating of the pumps involved, as high-head high gph pumps while providing greater flow will also add more heat to the system.

It's a balancing act, and a subject that is constantly discussed in the two forums I linked. Look around at some of the other threads in the forums and check it out, we'd be happy to see you there!
 
Higher flow is better for heat transfer in all situations. Needing slower flow through the radiator is a myth. Cooling water is the same as heating it, as far as heat transfer modeling is concerned. However, using a huge pump to obtain huge flow will indeed add more heat to your water.
Also, gravity plays no role in a closed loop system, so there is no syphon effect. There is no vertical head that needs to be taken into consideration, only restriction due to tubing, blocks, and radiators.
 
Thing is slower water doesn't mean lower flow. Larger pipes (or more pipes) can flow more water at lower velocities.

So maybe it isn't a myth as much a misconception.

High velocity with fewer/smaller long tubes or low velocity with more/larger short tubes. Both can have the same flow. All that realy determines the amount of heat that is pulled away is how long a given amount of water stays in the rad's cooling area.

Head is realy just a level of resistance that is made. They use pushing water through a pipe as a measurement because pumps are more frequently used to lift water to a certain location. A given fitting could be converted into a height resistance.
 
Originally posted by Pherret
Thing is slower water doesn't mean lower flow. Larger pipes (or more pipes) can flow more water at lower velocities.

So maybe it isn't a myth as much a misconception.

High velocity with fewer/smaller long tubes or low velocity with more/larger short tubes. Both can have the same flow. All that realy determines the amount of heat that is pulled away is how long a given amount of water stays in the rad's cooling area.

Head is realy just a level of resistance that is made. They use pushing water through a pipe as a measurement because pumps are more frequently used to lift water to a certain location. A given fitting could be converted into a height resistance.
Click to expand...

Good points. However, regarding the "All that realy determines the amount of heat that is pulled away is how long a given amount of water stays in the rad's cooling area" part, I respectfully disagree: it's a common misconception. Temperature differential determines the amount and rate of heat exchange, not the amount of time water remains in the radiator.

A quote from rogerdugans over at ocforums sums it up pretty well:

"*Heat transfer works best with the biggest temp differential: ideal would be cold water to cpu and hot water in the radiator. We cannot achieve this because a closed loop will achieve equilibrium at some point. The principle holds true however- the most efficient transfer happens at the greatest temperature differential, therefore higher flow rates will always help with all other variables remaining the same.
* It is the Heat Transfer that we want to maintain as efficiently as possible, and that is best done with a higher flow rate. Rather than thinking that there won't be enough time for heat to move towards the cool water, and therefore compromising heat loss, it is better to think that there is more fresh water moving onto the CPU and therefore, there is increased cooling.

The reason higher flow rates work better in computer water cooling is this:
There is more water with a larger temperature differential moving through the water block- this removes more heat.
There is more water with a larger temperature differential moving through the radiator- again removing more heat.

This is true even though a system with a lower flow rate will have more time to heat the water in the block and also more time in the radiator to cool the water: since the heat exchange works best with the greatest temperature differential, longer “stay time” is counter-productive."
Click to expand...

The post this was clipped from can be read in it's entirety in this thread.

Very good illustration of "head", and true: a fitting's resistance certainly could alternatively be represented as a height measurement.
 
We need the WB to most efficiently take away heat because it is dealing with a pre-determined area for which we can effectively remove heat(The CPU core size). Temp differential is primarily taken advandage there.

Radiators can get as large as you want, so keeping it efficient is not as needed here.

As you said "Temperature differential determines the amount and rate of heat exchange" Yes.. It does, but Rate over Time= total heat dissipated.

Therefore the longer it stays in the rad the closer to ambient, and thus cooler, it becomes.

Though sence this is an exponential type value, there does reach an area where a bigger rad will mean nearly no temp difference. That is where efficiency becomes an issue.
 
Originally posted by Pherret
As you said "Temperature differential determines the amount and rate of heat exchange" Yes.. It does, but Rate over Time= total heat dissipated.

Therefore the longer it stays in the rad the closer to ambient, and thus cooler, it becomes.
Click to expand...

Yes, but the as you increase the amount of time in the rad, you also increase the amount of time in the block, which increases the temp of the liquid.

Just the other side of the coin.
 
I understand where you're coming from, and your statement about the water in the rad would be true if there were a one-time or finite heat-load to be dissipated.

However, that isn't the situation, the cpu is constantly generating heat. The longer water stays in the rad, the longer the water also stays in the waterblock, resulting in higher temperatures at the heat source. The more water molecules that are able to come into contact with the heat source, the more efficiently the heat source can be cooled. Bear in mind that only a very small portion of the water actually comes into contact with the "hot side" of the block: laminar flow (a layer of static water molecules between the surface of the block and the turbulent water) prevents 100% efficiency, by limiting the amount of flowing water molecules that are able to come into contact with the block surface and remove heat energy.

Higher flow rate almost without exception is better than a slower one, due to reduction of total volume of the laminar layer, or "dead spots" in heat exchangers in the system (both rads and blocks). There is, however, a point can be graphed at which higher flow yeilds diminishing returns with any given setup.
 
Originally posted by Cinic
Yes, but the as you increase the amount of time in the rad, you also increase the amount of time in the block, which increases the temp of the liquid.

Just the other side of the coin.
Click to expand...

You beat me to it :D
 
OK, from now on, by faster flow I mean greater volumetric flowrate, which is independent of tubing cross section. And as I said before, YOU DON'T WANT SLOWER FLOW ANYWHERE when you're trying to transfer heat. Having slow or stagnat water in the radiator will eventually cool it, but flowing water will cool MUCH faster. It's a basic principle of heat transfer and fluid flow, faster flow equates to a smaller boundary layer (layer of water between the flowing portion and the block/radiator that is essentially NOT moving at all). And even faster flow leads to turbulence, which is even better, because now the boundary layer is practically gone.
 
Another facter here is how the water flows through the block. Water moving paralell over the hot part of the block is the least effective method. The best way is to have the water flowing perpindicular to the hot point inside the block. This is why blocks like the WW and RBX are the top performers. It also helps to have the water velocity higher at the point of contact with the hot spot, hence the flow impingement designs found in both the blocks I just mentioned. (bernoulli's principal: as flow restriction increases, velocity increases at a porportional rate)
 
Originally posted by cgrant26
Another facter here is how the water flows through the block. Water moving paralell over the hot part of the block is the least effective method. The best way is to have the water flowing perpindicular to the hot point inside the block. This is why blocks like the WW and RBX are the top performers. It also helps to have the water velocity higher at the point of contact with the hot spot, hence the flow impingement designs found in both the blocks I just mentioned. (bernoulli's principal: as flow restriction increases, velocity increases at a porportional rate)
Click to expand...

It doesn't have so much to do with the direction of water flow as the geometry of the surface it's impinging on, which I think is your underlying statement. If you had water flow parallelto the heated surface, but somehow designed it to create just as much turbulence as the impingement method, it would probably work just as well. And about the velocity, that's exactly what I was saying before about higher flowrates to diminish the boundary layer.
 
Originally posted by Cinic
Yes, but the as you increase the amount of time in the rad, you also increase the amount of time in the block, which increases the temp of the liquid.

Just the other side of the coin.
Click to expand...

It seems like everyone is getting flow speed and regular speed/velocity mixed up. There is a difference between Liters/hour and miles/hour. A liter is a measurement of volume and a mile is a measurement of distance. So flow rate and the amount of time any given amount of water is in the same place are not dependent to each other.

In everyone's WC setup the amount of time water spends in the WB and the amount of time it spends in the radiator are drastically different.

I agree with what you'r saying zer0signal667 and I think you actually see what Im saying. :)
 
Originally posted by Pherret
"It seems like everyone is getting flow speed and regular speed/velocity mixed up."
.......

"So flow rate and the amount of time any given amount of water is in the same place are not dependent to each other."
Click to expand...

Flow rate and the amount of time water is in the same place are the exact same thing. A water cooling setup has a fixed volume, so flow rate and speed are the same: x-liters per hour reflects the number of times water passes a fixed point (flowmeter) in the system.

Changing the diameter of your tubing can lower resistance, and the velocity within the larger tubing will be lower, while total volume moved increases, however, once the water hits your block, rad, or anything else and encounters restriction, it changes velocity and speeds up.

You cannot change the number of times a given volume passes a fixed point (radiator) in the same system without affecting the number of times it passes any other point (cpu, pump etc) in a direct proportion.
 
Originally posted by Pherret
It seems like everyone is getting flow speed and regular speed/velocity mixed up. There is a difference between Liters/hour and miles/hour. A liter is a measurement of volume and a mile is a measurement of distance. So flow rate and the amount of time any given amount of water is in the same place are not dependent to each other.

In everyone's WC setup the amount of time water spends in the WB and the amount of time it spends in the radiator are drastically different.

I agree with what you'r saying zer0signal667 and I think you actually see what Im saying. :)
Click to expand...

Volumetric flowrate does depend directly on velocity though. Flow can be measured in volume/time but can also be considered independently of volume in terms of distance/time.
 
Hmm, this has turned into quite a discussion.

I don't think the radiator or fittings will restrict such a low flow pump. I think only the high flow pumps will suffer.

I don't think anyone disputed my syphon statement, which essentially said there was no "head" in a common watercooling setup.

Once again, I think that the extra time the water is in the radiator will make up for the extra time the water is in the block.

Let me restate one thing though. I believe higher flow will increase cooling ability, but only by a tiny margin, like say less than 1 deg C.

Or maybe not? Wouldn't a higher flow pump add more heat to the water because of the larger motor? Probably not so much with an inline pump as with a submersible pump. But as far as high flow inline vs low flow inline I think the high flow pump would add a tiny amount of heat which may be enough to counteract the slight advantage of a high flow loop.

Thanks for your responses though
 
Originally posted by PiratePowWow
Hmm, this has turned into quite a discussion.

I don't think the radiator or fittings will restrict such a low flow pump. I think only the high flow pumps will suffer.

I don't think anyone disputed my syphon statement, which essentially said there was no "head" in a common watercooling setup.

Once again, I think that the extra time the water is in the radiator will make up for the extra time the water is in the block.

Let me restate one thing though. I believe higher flow will increase cooling ability, but only by a tiny margin, like say less than 1 deg C.

Or maybe not? Wouldn't a higher flow pump add more heat to the water because of the larger motor? Probably not so much with an inline pump as with a submersible pump. But as far as high flow inline vs low flow inline I think the high flow pump would add a tiny amount of heat which may be enough to counteract the slight advantage of a high flow loop.

Thanks for your responses though
Click to expand...

As the temperature difference between your cooling medium(water, in the case of the waterblock) and that which you are trying to cool(the waterblock directly, the cpu die indirectly) decreases, the rate at which the medium can absorb heat energy goes down. Basically what this means is the faster you can get the water into and out of the waterblock, the better. The water will absorb less heat(but do it at a higher rate, because it doesn't heat up as much), and cool the block much better.

In regards to the radiator, it's a fairly similar experience, but in reverse. A given volume of water will spend the same amount of time in a given length of tubing assuming that the flow-rate remains the same. Basically, this means that "slowing down" the water(the speed, not the flow-rate) will have no effect. What is MUCH more important is increasing *surface area*. This is why something like a heatercore, which utilises several smaller tubes, travelling between two resevoirs with a high fin density in-between, provides much better heat transfer than a bended-pipe design; the amount of surface area between the water and the radiator is MUCH higher in a design like this.

In a closed-loop system, or even a semi-open loop system, with a resevoir, the only thing that matters is that the point at which the water enters the resevoir and the point at which the water leaves the resevoir are at least at the same height. If the point at which the water enters the resevoir is *lower* than the point at which the pump is pumping the water out of it, as far as I can tell you'll actually end up with *negative* head, and improve the flow-rate of the pump. It probably wouldn't make a huge difference, but regardless, as long as they're around the same height, then your effective head is zero.

That said, there are still things that will alter your flow-rate. Your waterblock will be a restriction, so will your fittings, so will your radiator.

A higher flow pump won't add *that* much more heat to the water, since any pump that you'll be recommended to use will be a magnetic drive pump. Generally as well, this can be combatted by running the water from the pump to the radiator first, for two reasons.
Reason #1: If you run already warm water through the pump, the temperature difference between the pump and the water will be lower, and the water will pick up less heat-energy from the pump in the first place.

Reason #2: Running the water through the radiator, and *then* the waterblock, ensures that you will get the coolest water flowing over the waterblock.

I also think you have it a bit backwards, in regards to how a pump's flow rate will be effected by fittings etc. A higher flow pump will probably be effected less by fittings, bends, elbows etc than a lower flow one; generally, a higher flow pump will also have a higher shutoff point, IE, it produces higher pressure. A lower flow pump will suffer more loss due to these things because it can't handle pressure nearly as well. However, even if I'm wrong, if you take a 60gph pump and pit it against my 500gph danner pump, you can attach all the fittings and elbows you like to the system; your 60gph pump WILL shut off before mine does.
 
No, I don't have it a bit backwards, I just failed to fully explain.

If the pump is fitted for 3/8" tubing and you bump it up to 1/2" (high flow) and make all the rest of the loop, waterblock, radiator, etc 1/2" fittings then the 3/8" low flow pump will be operating at maximum efficiency. what I really need to do is run a low flow pump through a radiator at 0 head and see how long it takes to fill up a gallon milk jug. Then do the same experiment with no radiator and see if the time to fill up the jug is the same.

You would not get negative head as the water being pulled up the the level of the pump is considered head.
 
Originally posted by Spewn

In regards to the radiator, it's a fairly similar experience, but in reverse. A given volume of water will spend the same amount of time in a given length of tubing assuming that the flow-rate remains the same. Basically, this means that "slowing down" the water(the speed, not the flow-rate) will have no effect. What is MUCH more important is increasing *surface area*. This is why something like a heatercore, which utilises several smaller tubes, travelling between two resevoirs with a high fin density in-between, provides much better heat transfer than a bended-pipe design; the amount of surface area between the water and the radiator is MUCH higher in a design like this.
Click to expand...

I wish you people would listen to me... Of course surface area is important, but we're not talking about that here. Like I said before, time is not the only factor that affects heat transfer, and rate of heat flow is not affected by it whatsoever. Slowing down the water flow DOES have an effect, and if you would lean about heat transfer between fluids and surfaces, then you would realize that. Time is not the issue, flow characteristics and surface geometry are issues. Heatercores are are better than straight-pipe radiators not only because of increased surface area, but also because fluid flow through all the small passages is much more turbulent than through a straight pipe.
 
Originally posted by zer0signal667
I wish you people would listen to me... Of course surface area is important, but we're not talking about that here. Like I said before, time is not the only factor that affects heat transfer, and rate of heat flow is not affected by it whatsoever. Slowing down the water flow DOES have an effect, and if you would lean about heat transfer between fluids and surfaces, then you would realize that. Time is not the issue, flow characteristics and surface geometry are issues. Heatercores are are better than straight-pipe radiators not only because of increased surface area, but also because fluid flow through all the small passages is much more turbulent than through a straight pipe.
Click to expand...

I don't recall saying that the rate of heat flow was affected by the velocity of the water flowing accross the hot surface(hot, in reference to the water). I did however state that as the water flowing accross this surface heats up, the temperature difference between the two goes down, which causes the rate at which heat energy can be absorbed by the water to ALSO go down. Given that, flowing the water over the heat source at a faster rate will allow for better heat transfer because you will be maintaining a higher temperature difference between the waterblock and the water.
 
Originally posted by Spewn
I don't recall saying that the rate of heat flow was affected by the velocity of the water flowing accross the hot surface(hot, in reference to the water). I did however state that as the water flowing accross this surface heats up, the temperature difference between the two goes down, which causes the rate at which heat energy can be absorbed by the water to ALSO go down. Given that, flowing the water over the heat source at a faster rate will allow for better heat transfer because you will be maintaining a higher temperature difference between the waterblock and the water.
Click to expand...

Actually, slower moving water will create a larger temperature difference between the fluid and the surface because of the thicker boundary layer.
 
Originally posted by zer0signal667
Actually, slower moving water will create a larger temperature difference between the fluid and the surface because of the thicker boundary layer.
Click to expand...

What happens when that water heats up?
 
Originally posted by Spewn
What happens when that water heats up?
Click to expand...

That's what is supposed to happen, it heats up and then you transfer that heat into the air. It's going to heat up no matter what the rate of transient heat flow is, where do you think the heat is going? And once equilibrium is reach, heat transfer rate is the same no matter what flowrate you have (approximately equal to the heat output of your CPU). However, with slower flow you'll end up with a larger temperature gradient- meaning higher CPU temps.
 
You must log in or register to reply here.
Back
Top