The impact of tubing sizes

[Otter]

If I a reading the C/W charts correctly I should be able to deduce that it is much better to have a higher flowrate through the water block than the radiator?

Yes. Though as long as you keep above the elbow of the c/w curves (usually around 2 or 3 LPM), more flow won't make all that much difference to either rad or block.

I am asking this because of all of the uproar about Swiftech posting the pics of two Stealths running in parallell. Will that REALLY hurt the temps of the cores on the cards much?

If the pump is decent, I wouldn't think so.

 

[eightballrj]

Thanks a bunch Otter! Any second opinions, conflicting or not?

 

[SureFoot]

I had various 3/8" and 1/2" setups myself, and indeed the difference in flow is not about the tubing size, but more about the fittings, elbows, and waterbocks that are in between (if we consider we keep the same pump).
A good choice of fittings is way more important. If you have to use elbows, try to use round ones such as those made by Eheim, instead of regular "square" ones. Drilling out barbs in order to increase the inner diameter works very well. Getting slightly oversized barbs is also a solution, at the expense of convenience.
Try to avoid any sudden diameter change, pinch, and other "speed bump" and the loop will flow very well in all cases.
In a Lian Li V2000 i have enough room to work a 1/2" loop with no problem, in smaller cases i'll pick 3/8". A watercooled SFF would have to use 1/4"...
For me available parts & barbs are mostly in 1/2" as it's a very common size (aquarium stores carry them typically) and smaller sizes are a bit harder to find.
So all in all tubing size is mostly a matter of convenience, to fit the parts you have together, in the way that is the less painful for you.

 

[natewildes]

Just as a clarification: the DDC pumps will still greatly benefit from the modded tops; the increase in tubing size from 3/8" to 1/2" is not what creates the better results, but the change in inlet position.

On that note, off I go to buy a DDC!

 

[Kyle_vdk]

My head hurts now..

 

[QuietIce]

 

[QuietIce]

Tubes Sizes , if you have a bigger tubes that the water block on the CPU that it will absorbs much more heat , it make your computer cool faster . you will have much more time to the your application and away from early computer breakdown ..

Uhhh, not exactly ...

 

[Conumdrum]

Umm yep. Not exactly that simple. Bigger tubes, better flow. Of course, flow only helps to a point.

 

[ravaneli]

Not to stir anything, but I have seen many professional comparisons of radiators when I was buying mine, and different radiators have different cooling characteristics depending on the water flow and speed of the fans. It is possible one radiator to cool better than other at certain flow (water or air) but worse than them on different flow.

All I am saying if you performed the test with a different radiator you will get different results, where the difference might be bigger.

All that said, I do believe that size of tubing has insignifficant effect on temp, since elbows and water blocks are where the restriction and pressure drop occurs.

Also, did anyone consider that flowin 5 lpm with 3/8 tubing and flowing 5 lpm w 1/2 means that the water in the 3/8 flows significantly FASTER? Volume per minute is not the same like speed per minute..

 

[Frankenchrist]

HAHA! THANK YOU! You Just saved me alot of money. Now I will just upgrade my rad and block instead of my whole system.

 

[QuietIce]

Also, did anyone consider that flowin 5 lpm with 3/8 tubing and flowing 5 lpm w 1/2 means that the water in the 3/8 flows significantly FASTER? Volume per minute is not the same like speed per minute..

I saw water speed make a difference ONCE. When I was testing 1/2" compared to 3/8" tubing the Apogee (original design) did slightly better with the smaller tubing - probably because there was more of a "jet" effect with the smaller barb. But most of today's blocks have that effect built into the design itself, so it's not an issue anymore. Other than that one instance on an old and very plain block I've never seen water speed make any difference. After all, what you're really talking about is water speed in the tubing since the water paths in the blocks are the same regardless of the tubing size used. Volume of water (GPM) is what carries off heat and determines water speed (and turbulence) inside the block ...

 

[Porvalsh]

I'm not convinced of these results as of yet. It could be that with the systems that we are using here, the results represent what we would normally see regardless of these other factors being discussed, but I'm not convinced this is the case.

Our pumps are rated at a certain volume. If the tube is smaller then the speed of the fluid will be faster than in a larger tube. Faster water creates more turbulence through the system, thus creating a larger pressure drop. In our system, this pressure drop is of no consequence because the pump can keep up. Pressure drop across components is usually only of significance when our fluid is being used to do WORK being that Work=Pressure Force*Area. If a pump produces 100psi, we want to use all that to do work, not to move the fluid through elbows and piping.

For us, I think that we're only concerned with flow (assuming the radiator can keep up with the heat transfer required), that flow being laminar. Like mentioned first off, maybe we just have pumps that effectively are able to work across this full range of tubing sizes. If this is the case, its no wonder tubing size does not make a difference in heat transfer, because flow is basically constant, as being determined by the pump.

At some point a smaller tube size will create too high of a pressure drop, which will disrupt flow. In the worst case, it could dead head our pump and stop all flow. Whether or not we'll see that in our application would require a lot of math which I haven't done in a long time, lol.

 

[ddrueding]

Hi (new here, and to WC in general).

I have a related question. For loops that are longer than average (10 feet? 20?) would the larger diameter tube have a more significant impact on pressures?

I plan on locating the rad a significant distance from the PC, and am concerned about the added resistance of the tube itself.

Thanks!

 

[QuietIce]

to OCF!


Yes, if you have longer runs the tubing size is more important. I also have external rads (heater cores, actually) with long runs and have always used 1/2" tubing.

If you feel you need smaller tubing inside the case you can do that with minimal restriction using standard copper pipe adapters instead of barbs. Or you can run the 1/2" tubing to the first component (block, pump, whatever) then come out of that component with the smaller size, going back to 1/2" at the outlet of the last component ...

 

[ddrueding]

Thank you for the quick response! I have 30' of 1/2" on the way, and will just build the whole loop with that.

 

[4GHZ_or_bust]

30 feet that is quite long for one computer. The most I've needed was only 6 feet when I had external reservoir and radiator. Just wondering what did you need 30 feet for?

 

[Spawn-Inc]

30 feet that is quite long for one computer. The most I've needed was only 6 feet when I had external reservoir and radiator. Just wondering what did you need 30 feet for?

i'm assuming it's in a remote location like an air conditioner.

i have a bit more then 15 feet in my setup with 2 runs over 5 feet over.

 

[ddrueding]

30 feet that is quite long for one computer. The most I've needed was only 6 feet when I had external reservoir and radiator. Just wondering what did you need 30 feet for?

Several different configurations I have in mind, actually. My first will be my office machine, and I plan to stick the radiator on the other side of a wall (in another room) so it won't be noisy.

Specs for this system are as follows:

i7EE@4Ghz, 12GB RAM@1600MHz, OCZ Z-Drive m84 512GB, HD5870, Gigabyte GA-EX58-Extreme, Antec P-183, Antec CP-850, 7 Ultimate

As there are no HDDs in the system, it should be pretty darn quiet with the pump and radiator in another room!

 

[Spawn-Inc]

Several different configurations I have in mind, actually. My first will be my office machine, and I plan to stick the radiator on the other side of a wall (in another room) so it won't be noisy.

Specs for this system are as follows:

i7EE@4Ghz, 12GB RAM@1600MHz, OCZ Z-Drive m84 512GB, HD5870, Gigabyte GA-EX58-Extreme, Antec P-183, Antec CP-850, 7 Ultimate

As there are no HDDs in the system, it should be pretty darn quiet with the pump and radiator in another room!

how many blocks/rads will will there be? for the sake of safety and flow i might look into 2 pumps. since it's in the other room then 2 DDC's and a dual top would work great.

 

[Lutfij]

thanks for the info, opened my eyes alot on my WC loop!