Smaller/less tubing = higher flow = worse temps?

[MasterplayerC]

Just switched tubing, used less (maybe a foot or two?) and its 3/8 instead of the previous 7/16. Temps seem to be exactly the same at load, or possibly a little worse.

The reason this shocks me is because I went from slow Yates mounted directly to the rad to GT AP-15s at the full 1850RPM mounted to a 25mm shroud in push.

The reason I question higher flow is because with bubbles in the res I could see them moving around (can't without bubbles).

pump: MCP350 with XSPC top
rad: XSPC RX360

Thinking about getting a fan controller for the fans anyway, so would it make sense to hook the pump to it and see if I get improved temps with the pump at a lower speed?

[ggdh]

Tubing length or diameter should not affect performance or flow that much. There needs to be something else wrong (mounting? TIM?)

[Kurz]

I think your delta of water to air temp is pretty tight already. Everything past 10C delta is dwindling returns.

[bmaverick]

You could do a fan controller on the pump, but expect a good amount of heat dump for every 1VDC you drop the pump. The MCP350 can safely run at 7VDC as long as the number of watts is near or at 10W.

I tried resistors, then a fan controller, but the heat load was still hot. Finally settled on a wall-plug of 9VDC@750mA. The temps for my loop in most instances stayed the same and the pump is very silent. I'm all for a silent system and not necessary in this for OCing.

[64dragon]

Tubing size has basically no impact on temps, see here http://www.xtremesystems.org/forums/...d.php?t=147767
However, you should have seen some drop from the increased fan speed. I'm with ggdh, if you removed your block I'd question the amount of TIM used this time or mounting pressure

[BrokenArrow]

3/8" ID tubing area = .1104 square inches

7/16" ID tubing area = .1503 square inches

Area difference ~ 36%

Pressure drop is a function of area (square of flow velocity).

Overall flow rate is a function of total pressure drop.

So you may have shortened the total length, but you have made it a little more restrictive by changing to a smaller diameter.

I know the Cathar graphs indicate negligible differences, but maybe the small differences for the worse you note is from the 3/8" tubing change.

P.S. The fluid velocity would be higher in the 3/8" tubing.

[MasterplayerC]

thanks for the info guys. I've gotta do a reseat or two and see if it makes a difference (i'll admit i did rush it a bit).

temps definitely arent BAD, and i just realized that the change of location (that i failed to mention) probably has something to do with it. i moved it from home to work, where its definitely warmer. ambient at home never went above 70F, now its closer to 80F

idle is 30C (86F), full load never goes above 62-64C.


probably still going to get a fan controller at some point, but won't bother putting the pump on it.

will reseat at some point today and report back

[SlowSVT]

A smaller tube diameter increasing the velocity of the coolant results in more turbulence in the system which works the pump harder. You should always run as large of a tube diameter as possible regardless of the restrictions within the blocks, radiator and reservoir. 90 degree fittings should also be avoided. The slower the coolant moves the better.

[fitipaldi]

A smaller tube diameter increasing the velocity of the coolant results in more turbulence in the system which works the pump harder. You should always run as large of a tube diameter as possible regardless of the restrictions within the blocks, radiator and reservoir. 90 degree fittings should also be avoided. The slower the coolant moves the better.

Hell no! Less time water spend's in the blok the bether.
From my expirence, just by turnig my pump (Laing D5) from min to max i get 5°C lower temps on my GPU at load.

[BrokenArrow]

Hell no! Less time water spend's in the blok the bether.
From my expirence, just by turnig my pump (Laing D5) from min to max i get 5°C lower temps on my GPU at load.

Ugh Oh. Here comes the race track analogy. More flow rate (not velocity) is always better, but has diminishing returns.

[fitipaldi]

Oh come on, enlighten me.

More flow rate (not velocity) is always better, but has diminishing returns.

Btw, more flow=more mass(-water) at a same time amaunt=more speed=greater velocity.(at the same tube diameter)

[SlowSVT]

Hell no! Less time water spend's in the blok the bether.
From my expirence, just by turnig my pump (Laing D5) from min to max i get 5°C lower temps on my GPU at load.

The size of the tubing or fittings has nothing to do with how long the water stays in the block or radiator. It only effects the velocity of the coolant inside the tubing.

Think about it.

[fitipaldi]

The size of the tubing or fittings has nothing to do with how long the water stays in the block or radiator. It only effects the velocity of the coolant inside the tubing.

Think about it.

No, it dozen't. Smaller tubing just obstructs the flow more, like you said. But how long water stay's in the block is importanant and that's where flow comes in. You said;

The slower the coolant moves the better.

and that means that it move's slower thru block.

[SlowSVT]

No, it dozen't. Smaller tubing just obstructs the flow more, like you said. But how long water stay's in the block is importanant and that's where flow comes in. You said;

and that means that it move's slower thru block.

No

The flow rate in the tube has nothing to do with the flow rate in the block unless the tube is more restrictive then the block itself.

Once the slower moving coolant in the large line hits the restriction imposed by the water block it will speed up (think of it like an Hourglass).

AGAIN! The size of the tube has no impact on how long the coolant stays in the block (the pump pressure and volume will determine that). That is unless the tube is small enough where it becomes more restrictive then the block itself which is what I was getting at in the first place.

This is pretty basic hydraulics.

[fitipaldi]

Oh, man! I'm NOT!!! talking about tubes at all!

I'm disagreeing only to part where you said that "The slower the coolant moves the better."
Than stop the flow and see what happens...
Know what i mean now?

[Church]

I wonder why some simply don't get that in closed loop LC water will spend time in rads and waterblocks same amount of time no mater what the flow is? You double flow, but water recirculates and goes through those rads/waterblocks twice, at the end spending in those components same total time. +With higher flow because of extra introduced turbulence instead of slow, laminar flow, you get better accumulating heat in waterblocks and better transfering it to air in rads.
Of course, at some point increasing flow gets diminishing returns and simply is not worth to invest in enhancing it further, but artificially limiting flow to as slow as possible is just that - stupid thing to do.
Only scenario, when higher flow can result in little bit worse cooling - with underrad loops, when pump's own heat dump is noticeable relative to rad cooling capabilities. Then yes, better heat exchange in waterblocks and rads can be somewhat offset with increased pump heat dump at higher performance/higher flow modes. But i don't consider such scenarios at all, as such underrad liquid cooling systems cooling performance-wise are similar to air coolers just costing much much more. Where is sense to buy & build such?

[prava]

No

The flow rate in the tube has nothing to do with the flow rate in the block unless the tube is more restrictive then the block itself.

Once the slower moving coolant in the large line hits the restriction imposed by the water block it will speed up (think of it like an Hourglass).

AGAIN! The size of the tube has no impact on how long the coolant stays in the block (the pump pressure and volume will determine that). That is unless the tube is small enough where it becomes more restrictive then the block itself which is what I was getting at in the first place.

This is pretty basic hydraulics.

Flow rate is constant througout the loop, but the pressure is not, obviously.

The faster the water moves the more turbulent it is and the more heat can take. And, If you think about it, it doesn't matter how fast the water moves it will spend exactly the same time in each part of the loop.

[fitipaldi]

+With higher flow because of extra introduced turbulence instead of slow, laminar flow, you get better accumulating heat in waterblocks and better transfering it to air in rads.

The faster the water moves the more turbulent it is and the more heat can take.

And that's what i've been trying to say.

I wonder why some simply don't get that in closed loop LC water will spend time in rads and waterblocks same amount of time no mater what the flow is?

And I get this.

English isnt my native, maby i used wrong words?

[SlowSVT]

Oh, man! I'm NOT!!! talking about tubes at all!

I'm disagreeing only to part where you said that "The slower the coolant moves the better."
Than stop the flow and see what happens...
Know what i mean now?

That's funny, I was under the impression we were talking about tube size and nothing else but now you saying "your not talking about tube size"

[Church]

fitipaldi: I just wrote that when i read this thread since the first post and stumbled upon bolded The slower the coolant moves the better. And it's not the first time i see such misconceptions (nor first forum i read such ). I wonder where people get such ideas/what makes them think like that? :/

[SlowSVT]

Flow rate is constant througout the loop, but the pressure is not, obviously.

The faster the water moves the more turbulent it is and the more heat can take. And, If you think about it, it doesn't matter how fast the water moves it will spend exactly the same time in each part of the loop.

.........and the higher the velocity the more turbulent it becomes making it harder on the pump to move the coolant......which equated to a lower volume of coolant thru the system.

That all I was getting at. Nothing more.

Put another way: Would you rather have 3/8" ID or 1/2" ID tubing in your loop?

[fitipaldi]

No i was not talking about tube size. I tryed to make that clear by underlining sentence in your post. I'm agreeing with every thing else in that post.
I was trying to tell that i disagree with & why i disagree underlined sentence.
Hope now you understand what i meant.

[SlowSVT]

fitipaldi: I just wrote that when i read this thread since the first post and stumbled upon bolded The slower the coolant moves the better. And it's not the first time i see such misconceptions (nor first forum i read such ). I wonder where people get such ideas/what makes them think like that? :/

But you conveniently left out what was stated prior

"A smaller tube diameter increasing the velocity of the coolant results in more turbulence in the system which works the pump harder. You should always run as large of a tube diameter as possible regardless of the restrictions within the blocks, radiator and reservoir. 90 degree fittings should also be avoided." The slower the coolant moves the better.

Is that a misconception?

It's not a good idea to omit a few details when trying to make a point.

[SlowSVT]

No i was not talking about tube size. I tryed to make that clear by underlining sentence in your post. I'm agreeing with every thing else in that post.
I was trying to tell that i disagree with & why i disagree underlined sentence.
Hope now you understand what i meant.

No "biggie" man. I was very clear on what I was trying to convey but at this point all I'm doing is "bust'N em".

Happy New Year

[Church]

SlowSVT: I have nothing against prior part of statement, it's true to some extent (except that Kathar's test in stickies shown that tubing size doesn't impact flow and temps that much as some might have thought, and in many cases convenience that smaller ID tubing brings for easier tube routing in cramped cases can more then offset slight win that is brought by using "fat sausages"). It's the last sentence (this time not bolded ) about outright non true statement about less flow being better that i tripped at. If you say first truth, then finish with lie, how should i answer you? It's not misconception, it is?