A challenge!

[Bignuts]

I think I am gonna volunteer for making a "low flow" system, using the theory that the lower the flow, the more time the water spends in the radiator, the better the system performs. I think I read that somewhere....

[dutchcedar]

Originally Posted by jaydee116

It seems to have been seriously edited. http://www.wizdforums.co.uk/showthread.php?t=427 I missed it aswell. :shrug:

It was... but you didn't miss much. Just a few people airing opinions... :shrug:

[jaydee]

Originally Posted by dutchcedar

It was... but you didn't miss much. Just a few people airing opinions... :shrug:

Much like the AMBMC/PCPerspctive Lightning round eh? I assume your the same DutchCedar. Lost my intrest in that site.

[Cathar]

I have some questions that I believe need to be answered before anything like this proceeds:

1) What is defined as low-flow?
2) What is defined as high-flow?
3) What is defined as thin tubing?
4) What is defined as fat tubing?
5) Is there any blurred middle-ground that we want to avoid purely for the sake of providing a clear distinction between the two?
6) If there is a blurred middle-ground, what do we do about it?

[Chew_Toy]

Good questions and with each I would guess there would be some blurring.
Maybe if enough people would define each as they see it we could reach some common labels for each.

!. I would say low-flo would be pretty much anything with a pump under 3~4' head and maybe small tubes also.
2. System with pumps with 5' head and up and large tubes.
3. Tubes 8mm and under.
4. Tubes 10mm and larger.
5. I dont think that any middle-ground should be avoided, since I feel both sides will overlap each other just because the tech is getting better and will see each side moving closer to the middle.
6. With the middle-ground I would say that would be systems that fit either mold and would be a win/loss for either.

To me its just like anything else in that things are very seldom black and white, there is usually a lot of grey area and its nothing to get upset over.

[dutchcedar]

Originally Posted by jaydee116

Much like the AMBMC/PCPerspctive Lightning round eh? I assume your the same DutchCedar. Lost my intrest in that site.

Same-same... its still a nice place to visit if you work by yourself and need to yack during a coffee break...

I think I take too many coffee breaks...

[jaydee]

Originally Posted by Chew_Toy

3. Tubes 8mm and under.
4. Tubes 10mm and larger.

Well there will have to be some compromise here as in the US we don't use metric tubing.

10 millimeter = 0.3937008 inch
8 millimeter = 0.3149606 inch

.625" is 5/8".
.5 is 1/2".
.375" is 3/8".
.25 is 1/4".


Those are common tubing sizes available in the US. Might throw a little curve in the tests being metric tubing has different ID's than standard tubing.

[Cathar]

For most circumstances:

13mm ~= 1/2" or 12.70mm
10mm ~= 3/8" or 9.53mm
8mm ~= 5/16" or 7.94mm
6mm ~= 1/4" or 6.35mm
4mm ~= 5/32" or 3.97mm

For ID the above can be used quite interchangeably and freely for most things, especially for barb style fittings. The only time where it may become an issue is where the OD is of critical importance for certain push-fit fittings. eg. where 3/8" OD won't seal well in a 10mm OD push-fitting, or 1/4" OD won't fit inside the collet of a 6mm push-fitting.

[Cathar]

Originally Posted by Chew_Toy

1. I would say low-flo would be pretty much anything with a pump under 3~4' head and maybe small tubes also.
2. System with pumps with 5' head and up and large tubes.

I see you included the pumps and tubing. What do we do then call, say, an Eheim 1046 system with 1/2" ID tubing and a low-resistance block and radiator pushing 1GPM, or perhaps a 1048 in the same system pushing over 1.5GPM?

Or for #2), what about, say an MCP600 with a high resistance block, fat tubing, and pushing just 0.5GPM?

I would say that when many people say "low-flow" what they are really trying to say is low-pumping pressure, or perhaps more rarely, moderate pumping pressure matched with restrictive tubing. To define it in terms of pumps and tubing is rather open-ended.

I was more hoping for specific flow rate ranges. How about the following?

< 3 LPM = low flow
3-6 LPM = moderate flow
> 6 LPM = hi flow

Originally Posted by Chew_Toy

3. Tubes 8mm and under.
4. Tubes 10mm and larger.

What about 6mm ID, or even 4mm ID, which are tiny even in comparison to 8mm ID? Like for flow-rates, I'm more tempted to have 3 categories, which also happen to align themselves with the above flow-rate ranges in terms of the tubing not playing a major factor in the system-wide resistance.

<8mm ID (< 5/16") = thin tubing
8-11mm ID (5/16" to 7/16") = moderately sized tubing
>11mm ID (>7/16") = fat tubing

Originally Posted by Chew_Toy

5. I dont think that any middle-ground should be avoided, since I feel both sides will overlap each other just because the tech is getting better and will see each side moving closer to the middle.
6. With the middle-ground I would say that would be systems that fit either mold and would be a win/loss for either.

To me its just like anything else in that things are very seldom black and white, there is usually a lot of grey area and its nothing to get upset over.

I believe that the middle areas should be defined, since they are areas of convergence and grey, and also to provide a clear distinction that everyone can agree upon when talking about hi/low flow, and thin/fat tubing. Such seems to have been the primary issue of contention for past debates where certain "low-flow" setups definitely crept up into the muddled grey-area between the two sides of the debate.

Personally I love the middle-grounds as defined above - they best reflect my own opinion on where water-cooling should be headed to deliver the best of all worlds.

[Kobuchi]

<HINT> An European might nominate Cathar moderator.

[msv]

Originally Posted by Kobuchi

<HINT> An European might nominate Cathar moderator.

Cathar is a very polite person in all his writings here. Among other virtues of his.

BTW, doesn't Rittall use 4 mm ID tubing for their watercooled 19" computer racks?
regards
Mikael S.

[pHaestus]

Quite agree with this:

< 3 LPM = low flow
3-6 LPM = moderate flow
> 6 LPM = hi flow

Don't know why people would define "high and low" flow without listing a flow rate. I would also say 8mm and under would be "small tubing" and I would also put 11mm and up as "big tubing".

I am not sure Cathar is so universally liked; that's one of the things that spawned this cluster**** of a thread after all

[Les]

My definition of low flow :-
< 0.6 LPM for 6mm ID
< 0.9 LPM for 9mm ID
< 1.2 LPM for 12mm ID
These are the LPM below which the flow becomes laminar (Re< 2200)

[AngryAlpaca]

So we're all with high flow systems, Les? Then there should be no argument as we are all equal. Useless numbers.

I don't think this will be that big of a challenge...

Usually requires large heatecore... Can't add HDD, NB, GPU, Mosfet blocks without lowering flow (not inaccurate, just stupid)... More powerful pump needed...

Can add multiple blocks without damaging flow (see above)... Less powerful pump needed...

If that's the best they can come up with, this isn't worth the time.

Also, aren't they focussing too much on the tubing size? I thought this was about the block design and the pumps?

[Les]

Originally Posted by AngryAlpaca

So we're all with high flow systems, Les? Then there should be no argument as we are all equal.

No.
Few(if any) use, what I would describe as, low flow.
High flow I would define as when the flow becomes Fully Turbulent in the tubing.
However I find this boundary more difficult to define.
Some define fully turbulent with Re>4000, yet Wolverine Re>10000 .

Originally Posted by AngryAlpaca

Useless numbers.

If you say so.

[Myth]

Why bother about turbolent and laminar flow in tubing? In rads and blocks i get it.. but why in the tubing?

BTW: check this tool out: http://home.hccnet.nl/m.dijk/pressure_drop_calculator/

[pHaestus]

Not useless. Rather than defining flow as "low" or "high" based on arbitrary numbers, Les is using a definition that one can calculate from physical properties.

//edit: Very cool myth!

[Les]

Originally Posted by Myth

Why bother about turbolent and laminar flow in tubing? In rads and blocks i get it.. but why in the tubing?

Check out the difference between the MCW6002-A and the MCW6000-A
Maybe it is relevant there.

[BillA]

are we not actually interested in the transition between tubing sizes ?
at what point do the line friction losses indicate that performance would be improved by increasing the line size ? (all, LOL)
obviously this is a consequence of the flow rate, which is that resulting from the particular assembly of components in the system

following (more or less) the sizing from the above posts:
- for a 2m length of tubing, the flow resistance is (in mH2O):
size, mm . flow . . . 1x . . . . 2x . . . . . 4x . . where "x" is the initial flow rate for that size
6 . . . . . . . 0.6 . . 0.075 . . 0.263 . . 0.949
8 . . . . . . . 0.8 . . 0.031 . . 0.111 . . 0.389
10-3/8 . . . 1.0 . . 0.016 . . 0.056 . . 0.203
13-1/2 . . . 1.3 . . 0.007 . . 0.026 . . 0.092

now an observant reader might note that a diagonal from upper left to lower right will hold the line losses constant while the flow rate increases, which says . . . . .

simple stuff right ?
and that is quite all there is to it

the "German engineers" are making pretty wbs, and ignoring that which does not fit their conclusions
small is quite ok, if one accepts the compromises that follow
no biggie

easy friction losses: http://www.tasonline.co.za/toolbox/pipe/velfirc.htm
and other stuff

[BillA]

Originally Posted by AngryAlpaca

So we're all with high flow systems, Les? Then there should be no argument as we are all equal. Useless numbers.

I don't think this will be that big of a challenge... If that's the best they can come up with, this isn't worth the time.

Also, aren't they focussing too much on the tubing size? I thought this was about the block design and the pumps?

AA
your post is useless
what solution(s) did you propose ?
when you critique, also propose the/a solution

is everything you do not understand (Les' proposal in this instance) useless ?
the flow rate is the crux of the 'problem', Les understood, you did not

cool your jets

[pHaestus]

now an observant reader might note that a diagonal from upper left to lower right will hold the line losses constant while the flow rate increases, which says . . . . .

simple stuff right ?
and that is quite all there is to it

Exactly Bill

What is units on flow rate in your little table above? Took me a moment to digest that in its current form there heh

And yet the misconceptions persist; my hope was that people would be AWARE of the compromises being made (or being made for them).

//edit: Also of interest (maybe) is the upper limit to hose size for our systems. I don't know if there's an equation to determine the point where you can't clear the lines of air, but it's somewhere between 1/2" and 3/4" I'd say from my test bench. Also the hose routing (and flat mounting of wb with all that torque) becomes a big issue.

[Incoherent]

Originally Posted by Les

Check out the difference between the MCW6002-A and the MCW6000-A
Maybe it is relevant there.

Interesting observation Les.
I am trying to stay out of this, being a "low-flow" advocate because it's more practical in my case, not for performance reasons. The suggestion that smaller tubes could be beneficial at a given flowrate is appealing.
I ran a system with a windscreen/headlight washer pump for six months almost continuosly using copper brake lines. I.D 2mm. Radiator and pump outside through the winter with coolant as low as -20C. Excellent performance as far as I could tell at that time, if you don't count noise in the equation. When the pump, oddly enough, burned up, the performance drop with a more conventional centrifugal pump and summer temperatures was strangely dramatic.
The point being, limitations to performance can be compensated for in other ways, Making the whole argument a little moot.
In an SSF case a 6mm I.D. system could easily outperform an otherwise identical 1/2" I.D. system simply due to airflow restriction.
All other things being equal, high flow can't not beat a low flow system. It's an end, not a means and not really what the debate should be about.

[BillA]

lpm, sorry (started from Les' post)

and your edit is why low flow systems with BIG tubing won't work (though simple numbers would suggest its optimum)
we seem to be a total of 2 who understand this

[WAJ_UK]

I don't know if this will be of interest to anyone. I was just interested to see what kind of a difference tube diameter would make so I thought I would mess around with some numbers

It is a theoretical system with:
1m of tubing, using http://www.tasonline.co.za/toolbox/pipe/velfirc.htm to calculate pressure drop

a ww, using figures from http://thermal-management-testing.com/white%20water.htm

a FEDCO 2-274 using figures from http://forums.procooling.com/vbb/att...achmentid=1786

and pump info from http://forums.procooling.com/vbb/showthread.php?t=5068

The picture shows the effect on the WB C/W caused by the different flow rates acheived in each setup

[BillA]

this is all as obvious as the tail on a donkey

the low flow/small dia guys simply don't wish to accept the fact that there are performance losses associated with small dia systems

-> and are terrified that a "high flow" US wb may beat them also at low flow
the proof is in the pudding

(but I STILL cannot e-mail WCP !, grrr, a conspiracy I tell ya)