Ultimate block? Theory.

[Les]

Originally posted by Cathar

Les had an interesting series of graphs over at OCAU that plotted pump pressure vs water flow vs performance for some different wall/channel thicknesses/heights and the graphs were roughly showing that the differences are minimal. As we boost the pump pressure for a "tighter" block to bring its performance up, the block doesn't perform that much better if the same pump pressure is applied to a block of "more open" proportions.

Some more cofigurations to those shown in Volenti's thread (link below):



Volenti : http://forums.overclockers.com.au/sh...threadid=93479
Cathar: http://forums.overclockers.com.au/sh...threadid=95955

[bigben2k]

(This is an edited repost)

If someone here wants to try to reproduce Cathar's block, one could take a copper HSF, and add barbs and a lid.

From this list , I found the following were of a skived copper finned design:
(of course the fin spacing is WAY off, but it's not too bad)

Global Win CAK II 38 and CAK II 16

Dynatron DY1206BH-638

CoolerMaster HSC-V62

Dynatron DC1206BM-R

(I shortened this list even further, to include skived copper heatsinks, without a central hole for the mounting bracket)

[Volenti]

The U1 skivved heatsinks are preferable since the fin height is already almost optimum, I used this one to make this block



Since these pics were taken I've modified the block to have a centeral inlet like cathars as well as other minor internal changes.

if you leave more fin area on the base and leave the base at default thickness(my block was taken down to 2mm thick) it will cool quite well even at very low flow rates.

[Cathar]

Some interesting developments on my front with the micro-channel block.

So I've cut out a small plastic disc and inserted it under the central inlet barb. Out of that disc I've cut a 5mm wide x 15mm long rectangle that straddles the channels, giving a reasonably small entrance for each channel to receive water. This effectively boosts the water velocity and creates impingement jet streams directly over the heat source.

I also picked up a Pondmaster 4200 pump today too, which pushes 1000lph at 4m head, and close to 4000lph in wide-open mode.

Okay, now we get to the interesting bits of comparing nozzled and non-nozzled performance with the Eheim 1250 vs the Pondmaster 4200.

An AthlonXP CPU was set to 1925MHz/2.15v using BurnK7 to generate full load.

The following CPU die temperatures were observed above the water temperatures (reading the on-die diode of the CPU).

Eheim 1250 - No nozzle - 8.0lpm: +18.5C
Eheim 1250 - Nozzled - 7.0lpm: +18.0C (really it's more like +18.25C - temps kept flipping)

Pondmaster - No nozzle - ~15lpm: +18.0C
Pondmaster - Nozzled - 12.0lpm: +17.0C

What's interesting here is that the nozzled block improves in performance as the flow rates are picked up more than the un-nozzled block.

The Eheim 1250 seems to be struggling to get enough flow to make the jet impingement principle work properly, but the pondmaster's extra pressure ensures that the water velocity through the nozzle is boosted enough to get real benefits of jet impingement.

What's uncertain here is where there's more to be gained in terms of the nozzled vs non-nozzled configuration as the flow rates are picked up even more, or if the full benefits of the jet impingement is being realised already.

I believe this is on the right track, by mixing the two proven cooling methodologies of micro-channels and jet impingement, coupled with a pump capable of the water pressure and flow rates to make it all come together as a functioning whole.

[BillA]

as ever, more is better
ignore the actual flow rate
it is the impingment velocity at the surface that is of predominant interest

pinch it off some more, see what happens

[Dix Dogfight]

Cathar
Have you tried both pumps in series? Just to see if there are any benefits from still higher pressure with the nozzled version. Im not shure on this but it should increase pressure/flow. Ask myv65 or any of the other pump experts.
cheers

[Cathar]

Originally posted by Dix Dogfight
Cathar
Have you tried both pumps in series? Just to see if there are any benefits from still higher pressure with the nozzled version. Im not shure on this but it should increase pressure/flow. Ask myv65 or any of the other pump experts.
cheers

Hmmm, 12lpm is close to what the Eheim 1250 can do in wide open mode. I don't know that much about pump theory but it would seem to me that the Eheim 1250 would be adding very little extra pressure if it can barely keep up with the flow rate that the Pondmaster is pushing.

It may help as I pinch the nozzle more as BillA says and the flow rates start dropping below the 10lpm mark.

I'll try for a 4mm wide nozzle next.

[myv65]

As you already suspect, putting pumps in series is a bad idea if their respective output curves don't match reasonably well.

For the "pinching off", there are a few things involved. Ultimately what you are doing is converting pressure to velocity. How efficiently you do this is a function of geometry. A sharp edged orifice will have a higher head loss coefficient than a rounded-edge version. If your slots currently have 90° corners, you may see some minor improvement by rounding the inlet edges.

[Volenti]

Pinch that bitch cathar I just tried out my cheap chinese (resun) pump ,that has similar specs to your pond master , through a 4mm dia nozzle, it got 3.1L/minute, I was just about to test it with another smaller pump in series (like what was suggested for you) when my large pump broke it's intake almost clean off!

however before it met it's premature end I did notice the flow through the nozzle was considerably greater.

[Cathar]

Oh my. What do you think happened that caused your pump to do that (so I can avoid the same thing myself)?

[Volenti]

Originally posted by Cathar


Oh my. What do you think happened that caused your pump to do that (so I can avoid the same thing myself)?

I'd say because it's a cheap chinese pump made from cheap chinese plastic, I didn't like the "feel" of the plastic from the start (very brittle, low tensile strength) I wouldn't have thought that even this big pump would have enough pressure capability to bust it's self. I had it running by it's self with the small nozzle for some time, it wasn't untill I hooked up the other pump in series that it broke, the straw that broke the camels back?

I'm going to have a look at my local aquarium shop tomorrow (they can source "some" Iwaki pumps) to see if they can get a MD-30RZT (the high pressure model) looking at the PDF file for it it has a max pressure of 24psi

imagine around 20psi at 900 odd L/H

[BillA]

Volenti
you might want to lurk on eBay for a while to limit the expense
there is a point where the increased pump heat will negate the gain from increased convection
(as in more is better, but too much is pointless)
for a given wb this can only be determined by testing
- but with very accurate data the respective curves could be plotted to predict such

before upsizing the pump, sort out the nozzle configuration
sorry, forgot you're side to side

be cool

[Volenti]

Originally posted by unregistered
Volenti
you might want to lurk on eBay for a while to limit the expense
there is a point where the increased pump heat will negate the gain from increased convection
(as in more is better, but too much is pointless)
for a given wb this can only be determined by testing
- but with very accurate data the respective curves could be plotted to predict such

before upsizing the pump, sort out the nozzle configuration
sorry, forgot you're side to side

be cool

Yea I'll have to see how much a new one will cost, I know their not cheap.

I've changed the design of the block to a centre inlet after seeing how Cathar's performed (monkey see monkey do) anyway I can easily make another block, and expriment, that part isn't an issue for me.

As for heat, since I use evaporative cooling, I'll utilize the pressure capabiltiy of the pump to use a stack of micro irrigation misters in a spray chamber, or something similer, shouldn't be too much of an issue.

While I'm in it for the overclocking, I'm also in it for the hobby, and my hobbys are rarely cheap

[bigben2k]

Alternatively, as Cathar pointed out, the same design with taller fins might perform almost as well, with a lower flow rate.

[Cathar]

Originally posted by bigben2k
Alternatively, as Cathar pointed out, the same design with taller fins might perform almost as well, with a lower flow rate.

That's not quite what I said. I said that at lower flow rates that the higher fins will work better than the lower fin model.

[bigben2k]

Sorry for misquoting you!

I think that there is potential for a low flow design, with the taller fins. The question now is: how low can you go!

[Cathar]

Originally posted by bigben2k
The question now is: how low can you go!

Hmmm, perhaps that's a question best left to a thread not entitled "Ultimate block? Theory".

[Les]

Originally posted by bigben2k


The question now is: how low can you go!

Have some numbers scattered about for number and height of channels.
Next week will try to organise them into some kind of chaos,if anyones interested.?
They ONLY intended to apply to SIDE-INLET/SIDE-OUTLET configurations..Nature being like it is, the best designs for side-inlet/side-outlet are not necesarily considered the best for central inlet..

[Cathar]

With my block I'm seeing substantially better performance with central inlet over the side-to-side mode of operation. Coupled with the ability of the central inlet to do jet impingement down onto a thin base to achieve good effectiveness, I'm almost now totally convinced that side-to-side flows are a dead end for achieving "the Ultimate".

Whatever pressure is applied to boost side-to-side velocity, is better put to use to apply jet impingement directly at the heat source on a thin base, and then branch out to the sides. At least this is now my current working theory.

The question now is: how low can you go!

I was unclear whether this statement was referring to flow rates, or wall heights.

[Volenti]

Originally posted by bigben2k


I think that there is potential for a low flow design, with the taller fins. The question now is: how low can you go!

With the first version of my micro-channel block, link, I could get the flow rate down to as low as 80L/H (21gph) with temps only 4 degrees worse than the highest flow tested.

[godsdice]

I imagine that would be good for cooling a peltier but the flow is limited around the die with all that mass of tall narrow fins, one reason for the short fins or micochannels in Cathar's block. Furthermore, the best flow is coming from too far above the critical part of the system. I couldn't even begine to imagine how hard it would be to make that myself.

Still, Innovatech's design is somewhat similar to that in principle and it works quite well indeed.

EDIT: This was in reply to christoff's drawings. What the hell happened to the server clock???

[Volenti]

this would be better

EDIT: This was in reply to christoff's drawings. What the hell happened to the server clock???

forum clock attacks yet again

[Dix Dogfight]

Hmmm, 12lpm is close to what the Eheim 1250 can do in wide open mode

Isn't the 1250 like 1200lph=20lpm??? (with no head).

[cristoff]

what i was thinking....

[cristoff]

so would that work well?

Originally posted by Volenti
this would be better



forum clock attacks yet again