Why the NanoBlock is better - Page 2

Brad · 01-20-2002, 04:25 PM

how about we have a powerful pump, and a restricting vavle that we can use to restrict the flow rates, anywhere from 0% to 100%.

then we can see what is the best for that block and that radiator.

Of course you have to remember that each setup will have a differant component in it somewhere, or even just a differant layout. So what Una finds, with his radiators, block and pump will be differant for someone else.

UnaClocker · 01-22-2002, 06:59 PM

Yeah, but the thing is, the radiator will always want less flow than a waterblock will, so with your idea you can only find a happy medium between the two device's happy flowrate. I'd rather have 2 pumps, 2 loops, 1 reservoir. One decent pump for the radiator, like a 250gph, with a valve, and then a high speed pump, like Joe's 500gph monster pumping through the waterblock. Just run the block's water as fast as you possibly can, then tune the radiator's flow till your getting the coolest possible temps. I fully intend to setup something like this in one of my upcoming articles.
Should be fun.

Brad · 01-22-2002, 07:07 PM

yeah thats ideal, you want a res with two pumps in it, but it does add complexity.

I think having two rads in parrallel is a more elegant solution, you only need one pump, and it is easier to position two small radiators than one big radiator in a case.

futRtrubL · 01-22-2002, 09:00 PM

Having that two loop system is actually ineficient and will give you worce temps than a single loop. Cool water from your rad will mix with the hot water from your cpu and send warm water right back to the cpu.
Since I'm doing some bio cources for uni. I'm thinking of biological analogies. The circulatory system of an amfibian is the same as what you are sudgesting( rad.=lungs, each side of hart=pumps, main artery out=res, body=cpu) and they can only do that because they can get oxygen through there skin. Our blood system on the other hand is a single loop with 2 pumps. Most technology seems to be finding synthetic versions of natural mechanisms.

Edward

Brad · 01-22-2002, 09:06 PM

depends on how well you organise your resivoir/pumping system.

if you have the outlet of the radiator loop aimed at the inlet for the cpu loop, and the outlet of the cpu pump aiming at the inlet for the pump of the radiator loop, I can't see that being a factor

futRtrubL · 01-22-2002, 09:22 PM

If you have different flow rates then there will be mixing.A large flow into the res. can't turn into a small flow out to the rad,some of that flow will have to go out to the cpu. If both flows are the same then why have them going through the same res. and risk mixing by turbulence?

Edward

Brad · 01-22-2002, 09:37 PM

yeah I know, but water coming out of the block is barely differant in temp to the water going in

futRtrubL · 01-22-2002, 09:47 PM

But it is different and each time it goes through the block it'll get a little bit hotter. Meanwhile the rad is not getting full temp water so its not dumping as much heat (transfer driven by temp delta).

Edward

Lonely Raven · 01-22-2002, 11:27 PM

If this is still a "live" thread, I could donate a "Tidal Pool"
water block that uses the turbulent water theory. But I'd need
it back at tests end.

UnaClocker · 01-23-2002, 02:58 AM

Heh, Don't need a tidal pool, thanks anyways.

And that's a good theory explaining why it won't work. I won't be happy till I try.

Brad · 01-23-2002, 01:17 PM

so una, what about having two rads in parrallel, that will keep the water going slow through the rads

Pinkster · 01-24-2002, 01:01 AM

i agree with futRtrubL, having two loops one for the rad and one for the waterblock is a bad idea, regardless of how you organise it. with havign one loop all the hottest water, straight from the block, goes to the rad, so the rad is cooling the hottest water. alternately, all the coolest water, coming straight from the rad goes to the block. this is what you want, rad getting hottest water, so greater difference between water temp and ambient so greater difference between temp of water going in rad and temp of water going out of rad. and its obvious that you want the block to get the coolest water.
with a two loop system, the water going into the rad (eg the water in the res) is not the hottest water, it is a medium temp water, hotter than the water going to the rad - as there is also water coming straight from the block in that res which makes it hotter. But it is also cooler than the water coming straight from the block, as not all the water in the res has come straight from the block, some has come from the rad. this means that the water going into the block is not the coolest possible water, so worse cooling. It also means that the water going into the rad is not the hottest possible temperature, so the water temp is closer to ambient, so the water cools down less in the rad, so overall worse cooling.

what i am confused about as u said with the nanoblock the secret to it is that the water moves at a high velocity, due to being a narrow tube for the water in the block to run in, but i suppose being so short even with the narrow water channel beign so short it dosent restrict flow too much. U also said that the next block is going to beone with lots of narrow tubes going through the block. I think this is a great idea, as there is more surface area for the water to tough the copper, and i have been wanting to experiment with that kind of waterblock for a long time (i just dont have the equiptment necessary to constuct it). However, with lots of tubes , the velocity will be lower than in the current nanoblock. although i dont dispute that the new block with lots of narrow channels should perform fantastically, this must be due to the fantastic surface area for water touching copper, not as much the velocity the water in the tube is travelling as you said.

final point: if done well i would have thought that this type of block would really be able to handle high power pelts well given a good rad capable of dissapating the heat and so its effectivness could only really be tested with pelts.

Makaijin · 01-27-2002, 07:35 PM

Personally, I don't think running 2 loops in 1 res would be much more of a difference, even if the flowrate is exactly the same as a single loop system.
The only disadvantage I see in this is that the water wouldn't be the coolest entering the block. But the water temp will be the most stable throughout the whole system.
In a single loop system, if the res isn't removing the heat fast enough, the heat in the res would just accumulate just as fast as is would in a 2 loop system given same volume of water in both systems.
So, what I see in a single loop system is a hot part and a cold part as you travel through the loop.
In a 2 loop system, the water will be the same throughout the whole system.
But, having a 2 loop system would make it possible to fine tune both loops to run at their maxium efficiency, making the overall system alot more effective than a single loop using the same rad and block.

Makaijin · 01-27-2002, 08:06 PM

Looking at the 2 loop system it reminds me of the diagram on chip's website. Imagine the res on the diagram is not there, and the chiller is the res itself.
The concept is very similar, but instead of where the control valve is, the res will be there with the 4 pipes going in. Another pump will be running on the res loop.
Lets take chip's diagram, again imagine the res not there and the chiller is the res. If you open the valve, some of the warmer water will skip the res and back into the block.
If by doing this can lower the overall temperature of the whole system, I don't really see a flaw in using 2 seperate loops. The concept is very similar if not the same, but just applied differently.

BrianW · 01-30-2002, 02:41 PM

Quote:

what i am confused about as u said with the nanoblock the secret to it is that the water moves at a high velocity, due to being a narrow tube for the water in the block to run in, but i suppose being so short even with the narrow water channel beign so short it dosent restrict flow too much.

You are confusing velocity and flow. Flow is constant through the whole system in a closed loop. You add all the individual resistances together and get your flow. Velocity however depends on where in the system the water is. If you have a waterblock that has a smaller channel than the tube id, then the water will go through that point faster. The flow stays the same because there is less volume going through. The readiator on the other hand has wide channels (heatercore at least), therefore the velocity is slower. There again the volume is greater so the flow stays the same. The greater difference between the velocity of the water through the block and rad the closer to ambient your temps will be.

Now if you have two radiators in parallel the flow will be reduced by half through each rad. That is even better as the velocity will be twice as slow through the rad system.

Please correct me if I am wrong...

01-22-2002, 06:59 PM	#27
UnaClocker Cooling Savant Join Date: Apr 2000 Location: Bremerton, WA Posts: 514	Yeah, but the thing is, the radiator will always want less flow than a waterblock will, so with your idea you can only find a happy medium between the two device's happy flowrate. I'd rather have 2 pumps, 2 loops, 1 reservoir. One decent pump for the radiator, like a 250gph, with a valve, and then a high speed pump, like Joe's 500gph monster pumping through the waterblock. Just run the block's water as fast as you possibly can, then tune the radiator's flow till your getting the coolest possible temps. I fully intend to setup something like this in one of my upcoming articles. Should be fun. __________________ The UnaClocker Watercooler Extraordinaire! Overclock till it goes BOOM!

01-22-2002, 11:27 PM	#34
Lonely Raven Cooling Neophyte Join Date: Jan 2002 Location: Chicagoland Posts: 69	If this is still a "live" thread, I could donate a "Tidal Pool" water block that uses the turbulent water theory. But I'd need it back at tests end. __________________ The Lonely Raven Jack of All Trades, Master of None IT Samurai

01-23-2002, 02:58 AM	#35
UnaClocker Cooling Savant Join Date: Apr 2000 Location: Bremerton, WA Posts: 514	Heh, Don't need a tidal pool, thanks anyways. And that's a good theory explaining why it won't work. I won't be happy till I try. __________________ The UnaClocker Watercooler Extraordinaire! Overclock till it goes BOOM!

01-24-2002, 01:01 AM	#37
Pinkster Cooling Neophyte Join Date: Dec 2001 Location: adelaide, australia Posts: 61	i agree with futRtrubL, having two loops one for the rad and one for the waterblock is a bad idea, regardless of how you organise it. with havign one loop all the hottest water, straight from the block, goes to the rad, so the rad is cooling the hottest water. alternately, all the coolest water, coming straight from the rad goes to the block. this is what you want, rad getting hottest water, so greater difference between water temp and ambient so greater difference between temp of water going in rad and temp of water going out of rad. and its obvious that you want the block to get the coolest water. with a two loop system, the water going into the rad (eg the water in the res) is not the hottest water, it is a medium temp water, hotter than the water going to the rad - as there is also water coming straight from the block in that res which makes it hotter. But it is also cooler than the water coming straight from the block, as not all the water in the res has come straight from the block, some has come from the rad. this means that the water going into the block is not the coolest possible water, so worse cooling. It also means that the water going into the rad is not the hottest possible temperature, so the water temp is closer to ambient, so the water cools down less in the rad, so overall worse cooling. what i am confused about as u said with the nanoblock the secret to it is that the water moves at a high velocity, due to being a narrow tube for the water in the block to run in, but i suppose being so short even with the narrow water channel beign so short it dosent restrict flow too much. U also said that the next block is going to beone with lots of narrow tubes going through the block. I think this is a great idea, as there is more surface area for the water to tough the copper, and i have been wanting to experiment with that kind of waterblock for a long time (i just dont have the equiptment necessary to constuct it). However, with lots of tubes , the velocity will be lower than in the current nanoblock. although i dont dispute that the new block with lots of narrow channels should perform fantastically, this must be due to the fantastic surface area for water touching copper, not as much the velocity the water in the tube is travelling as you said. final point: if done well i would have thought that this type of block would really be able to handle high power pelts well given a good rad capable of dissapating the heat and so its effectivness could only really be tested with pelts. __________________ whats a sig? Last edited by Pinkster; 01-24-2002 at 01:14 AM.

01-20-2002, 04:25 PM	#26
Brad Thermophile Join Date: Oct 2001 Location: Nuu Zeeelin Posts: 3,175	how about we have a powerful pump, and a restricting vavle that we can use to restrict the flow rates, anywhere from 0% to 100%. then we can see what is the best for that block and that radiator. Of course you have to remember that each setup will have a differant component in it somewhere, or even just a differant layout. So what Una finds, with his radiators, block and pump will be differant for someone else.

01-22-2002, 07:07 PM	#28
Brad Thermophile Join Date: Oct 2001 Location: Nuu Zeeelin Posts: 3,175	yeah thats ideal, you want a res with two pumps in it, but it does add complexity. I think having two rads in parrallel is a more elegant solution, you only need one pump, and it is easier to position two small radiators than one big radiator in a case.

01-22-2002, 09:00 PM	#29
futRtrubL Cooling Savant Join Date: Jul 2001 Location: Kingston, Jamaica Posts: 204	Having that two loop system is actually ineficient and will give you worce temps than a single loop. Cool water from your rad will mix with the hot water from your cpu and send warm water right back to the cpu. Since I'm doing some bio cources for uni. I'm thinking of biological analogies. The circulatory system of an amfibian is the same as what you are sudgesting( rad.=lungs, each side of hart=pumps, main artery out=res, body=cpu) and they can only do that because they can get oxygen through there skin. Our blood system on the other hand is a single loop with 2 pumps. Most technology seems to be finding synthetic versions of natural mechanisms. Edward

01-22-2002, 09:06 PM	#30
Brad Thermophile Join Date: Oct 2001 Location: Nuu Zeeelin Posts: 3,175	depends on how well you organise your resivoir/pumping system. if you have the outlet of the radiator loop aimed at the inlet for the cpu loop, and the outlet of the cpu pump aiming at the inlet for the pump of the radiator loop, I can't see that being a factor

01-22-2002, 09:22 PM	#31
futRtrubL Cooling Savant Join Date: Jul 2001 Location: Kingston, Jamaica Posts: 204	If you have different flow rates then there will be mixing.A large flow into the res. can't turn into a small flow out to the rad,some of that flow will have to go out to the cpu. If both flows are the same then why have them going through the same res. and risk mixing by turbulence? Edward

01-22-2002, 09:37 PM	#32
Brad Thermophile Join Date: Oct 2001 Location: Nuu Zeeelin Posts: 3,175	yeah I know, but water coming out of the block is barely differant in temp to the water going in

01-22-2002, 09:47 PM	#33
futRtrubL Cooling Savant Join Date: Jul 2001 Location: Kingston, Jamaica Posts: 204	But it is different and each time it goes through the block it'll get a little bit hotter. Meanwhile the rad is not getting full temp water so its not dumping as much heat (transfer driven by temp delta). Edward

01-23-2002, 01:17 PM	#36
Brad Thermophile Join Date: Oct 2001 Location: Nuu Zeeelin Posts: 3,175	so una, what about having two rads in parrallel, that will keep the water going slow through the rads

01-27-2002, 07:35 PM	#38
Makaijin Cooling Neophyte Join Date: Jan 2002 Posts: 17	Personally, I don't think running 2 loops in 1 res would be much more of a difference, even if the flowrate is exactly the same as a single loop system. The only disadvantage I see in this is that the water wouldn't be the coolest entering the block. But the water temp will be the most stable throughout the whole system. In a single loop system, if the res isn't removing the heat fast enough, the heat in the res would just accumulate just as fast as is would in a 2 loop system given same volume of water in both systems. So, what I see in a single loop system is a hot part and a cold part as you travel through the loop. In a 2 loop system, the water will be the same throughout the whole system. But, having a 2 loop system would make it possible to fine tune both loops to run at their maxium efficiency, making the overall system alot more effective than a single loop using the same rad and block.

01-27-2002, 08:06 PM	#39
Makaijin Cooling Neophyte Join Date: Jan 2002 Posts: 17	Looking at the 2 loop system it reminds me of the diagram on chip's website. Imagine the res on the diagram is not there, and the chiller is the res itself. The concept is very similar, but instead of where the control valve is, the res will be there with the 4 pipes going in. Another pump will be running on the res loop. Lets take chip's diagram, again imagine the res not there and the chiller is the res. If you open the valve, some of the warmer water will skip the res and back into the block. If by doing this can lower the overall temperature of the whole system, I don't really see a flaw in using 2 seperate loops. The concept is very similar if not the same, but just applied differently.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)