Noob questions regarding tubing

[ncantador]

I've been entertaining the idea of changing over to water cooling, and I've read many post here and elsewhere. Just wondering if anyone has any first hand experience with using different diameter tubing in their setups and if that could possible improve performance.

For example, using a smaller diameter from the reservior>pump>rad, then larger diameter to the blocks. Theoretically, I beleive that would increase the flow rate in the blocks, and allow the use of a lowered powered, quieter pump without affecting performance that much. Nothing to prove it though, just wondering.

 

[Tigerbiten]

Almost right.

The largest bore pipe in the system wants to run between the res and the pump.
That will reduce the suction head pressure and cause the pump to run more efficently.

Luck.......

 

[kunsunoke]

For example, using a smaller diameter from the reservior>pump>rad, then larger diameter to the blocks. Theoretically, I beleive that would increase the flow rate in the blocks, and allow the use of a lowered powered, quieter pump without affecting performance that much. Nothing to prove it though, just wondering.

Hate to tell you this, but you're off base, both from a theory (e.g. transport phenomena) standpoint and from an in-practice standpoint.

I know of nobody who uses mixed diameter tubing in their system. There's a good reason for this. Each neckdown, adaptor, elbow, etc. you add to the system introduces frictional loss, such that volumetric flow velocity is decreased as a result. The energy of your pump is thus wasted trying to overcome frictional losses.

Think about it from another standpoint. Know anyone with atherosclerosis? How much harder does the human heart have to work to compensate for narrowed blood vessels with lots of flow restriction? How effficient is oxygen and or nutrient transfer from blood in that circumstance?

Instead of increasing volumetric flow rate (which will remain constant anyway), you'll only be limiting it by using different sized tubing. The system dynamics will be dictated by the smaller diameter tubing.

In water cooled systems, the tubing diameter should always be kept constant, and as large as is practical, so as to minimize frictional losses. 1/2" inside diameter tubing should be fine. You will also want to keep the tubing as straight as possible, and keep the number of sharp bends, twists, etc. to an absolute minimum.

The only place where loss in flow velocity is okay is within the water block or blocks. At these places, the turbulence caused by restrictive flow actually works for you, such that heat transfer is maximized from the CPU to the block to the water. This is why DangerDen now uses flow restrictor plates in their TBX blocks, such that the stream velocity (notice I didn't say volumetric flow rate) is maximized just over the top of the CPU. It also explains the need for high flow rate pumps when pressure drop is high across the block(s).

Rules of thumb:

More length of tubing = more friction = slower flow rate
More head (e.g. difference between lowest and highest points in system) = more power required = slower flow rate
More joints = more friction = slower flow rate
More elbows = more friction = slower flow rate
More surface area in block = greater pressure drop across block = slower flow rate (but also greater heat transfer)
Greater number of passes through radiator = greater pressure drop across radiator = slower flow rate (but also greater heat transfer)
Narrower tubing = more friction / higher pressure drop = slower flow rate

 

[kunsunoke]

Almost right.

The largest bore pipe in the system wants to run between the res and the pump.
That will reduce the suction head pressure and cause the pump to run more efficently.

Luck.......

What you suggest does reduce friction loss between the reservoir and the pump. However, it does not necessarily reduce suction head (NPSH) pressure. The suction head is really dictated by the elevation of the water in the highest part of the system.

Actually, reducing suction head pressure is a no-no. Reduce it too much and you end up with cavitation, which can cause destructive erosion of your pump's impeller.

 

[ncantador]

Hate to tell you this, but you're off base, both from a theory (e.g. transport phenomena) standpoint and from an in-practice standpoint.

I know of nobody who uses mixed diameter tubing in their system. There's a good reason for this. Each neckdown, adaptor, elbow, etc. you add to the system introduces frictional loss, such that volumetric flow velocity is decreased as a result. The energy of your pump is thus wasted trying to overcome frictional losses.

Think about it from another standpoint. Know anyone with atherosclerosis? How much harder does the human heart have to work to compensate for narrowed blood vessels with lots of flow restriction? How effficient is oxygen and or nutrient transfer from blood in that circumstance?

Instead of increasing volumetric flow rate (which will remain constant anyway), you'll only be limiting it by using different sized tubing. The system dynamics will be dictated by the smaller diameter tubing.

In water cooled systems, the tubing diameter should always be kept constant, and as large as is practical, so as to minimize frictional losses. 1/2" inside diameter tubing should be fine. You will also want to keep the tubing as straight as possible, and keep the number of sharp bends, twists, etc. to an absolute minimum.

The only place where loss in flow velocity is okay is within the water block or blocks. At these places, the turbulence caused by restrictive flow actually works for you, such that heat transfer is maximized from the CPU to the block to the water. This is why DangerDen now uses flow restrictor plates in their TBX blocks, such that the stream velocity (notice I didn't say volumetric flow rate) is maximized just over the top of the CPU. It also explains the need for high flow rate pumps when pressure drop is high across the block(s).

Thats ok, your corrections are appreciated, and this is exactly the type of information I'm looking for because as you stated, I haven't seen anyone that uses mixed diameter tubing either, I was just wondering why. There is one more question using your heart analogy above. Lets say you start out with restrictive tubing to begin with. Wouldn't increasing the diameter at the pump as tigerbitten suggested decrease the pressure in that area allowing the pump to work more efficiently. Sort of what happens in an angioplasty? Does Bernouli's law apply here?

Ooops, typing too slow, just saw your answer thanks again for the good information