New pump big enough?

[Kemist]

So any news? is this pump actually gonna be made? and if so any projected prices (prob too early for that but still)

[Hansfragger]

Originally Posted by Risky

I might suggest that you thread the part at 3/8 instead of 1/4 for the bigger pump (assuming this is physically possible. It may or may not make a material difference, but some people will be attacted by any tiny reduction in restriction.

I just finished comparing ID of various barbs designed for 1/2" ID tubing. Of the brass and poly models tested whether 1/4 NPT or 3/8 NPT, the internal flow path at the most restrictive point(through the barb), were all .37". The poly barbs that come with the HWLabs Black Ice Extreme radiators were a little bit better at .385". Of all I tested, the DD 9/16-18 NPT high flow barb has the largest opening at .40"!

[Bundles]

Originally Posted by kaotic504

damn, watercooling just gets sweeter by the day.

Pretty much what i was thinking.

This pump sounds almost to good to be true. Damn i hope it gets the go ahead.

[shadowing]

Is there any news on the development of this pump? It sounds very interesting.

[Monnie Rock]

Hello Everyone,

Hopefully we will hear something. Any news?


Thanks
Monnie

[Ls7corvete]

No, go away.

[maxSaleen]

I just thought that I would make it three posts in a row from floridians. What the hell, right? Plust I'd also like to raise my hand for and update from Dave.

[Ls7corvete]

Originally Posted by maxSaleen

I just thought that I would make it three posts in a row from floridians. What the hell, right? Plust I'd also like to raise my hand for and update from Dave.

You were warned, Orlando is within driving distance.

Wait patiently friends.

[Dave]

Sorry guys I keep forgeting about this other topic.

I answered in old topic forum

The big pump has completed testing, meeting all customer requirements.

I still do not know "if" the client intends to offer this pump to the general public.

I played around with the new pump and different cooling blocks yesterday, and the only block that shows any measurable improvement with the large pump, is the lastest Storm.

I got about 1.2C deg difference compared to a MAG-MKII.

[bobo5195]

Ah that'll be becuase modern wb arent restricitive enough to use it properly.

Storm shows improvement as jet impingment is heavily sensitive to flow rate for performance even more than standard convective cooling designs. I have a design for a block call hedgehog aka H1 (as H2 never looked like a hedgehog but was thermodynamically almost the same design) that i never got around to posting which should be able to use that pump as its curve kink is way way over as its got a stupidly high dP. Wide holed jet impingment (Lower g's) should show a big boost as well if there werent so optimised.

[JamesAvery22]

I dont know about any other MD20rz owners but I'd love for this pump to come out. Less heat + less noise =

[Les]

Originally Posted by bobo5195

....
Storm shows improvement as jet impingement is heavily sensitive to flow rate for performance even more than standard convective cooling designs..

I think Jet Impingement is a Myth
Was adopted by Wcooling community in two ways.
Initially used for modeling the White Water when crudely applied Sieder-Tate gave inadequate h(conv) to explain results.
Was used as a "catch phrase" description and found to provide exciting visualisation for non-channel, in particular the Cascade series ,waterblocks. No numerical modeling was ever applied..

The crude modeling community (I) still use it , via Flomerics, as a convenient method of calculating h(conv) when L/d is < 60 .Even though, with the reduction of the accepted value of C/W[TIM](tester's TIM) from 0.1 to 0.06(10x10mm die), this is probably not necessary - adequate h(conv) can be obtained from Sieder-Tate by adjustment of the prefactor, outside the accepted 0.19-0.27 range, to account for "entry turbulation"
The non-modeling community still use, I guess , because it provides a pretty picture.

[Cathar]

Nice cynicism there Les.

[Les]

Or perhaps the adoption of Jet Impingement is the cynicism.
Old-school thermodynamicists, Sieder and Tate, would probably think the adopters are the cynics.

[Cathar]

Would be interested to hear how the old school boys would describe blocks like the Storm as being anything other than "constrained submerged jet impingement".

The Cascade isn't really submerged JI though, and I've even said as much multiple times in the last 18 months. Originally designed it, modelling it with traditional JI theory, which strangely enough seemed to work well enough for estimation purposes, even though I've long since come to an understanding that it's not really JI, and rather it was merely a way to exact/maximise high turbulence mashing within a constrained area, much like its derivative inspiration being that of a faucet at full blast in a tall glass.

[Les]

Originally Posted by Cathar

1)Would be interested to hear how the old school boys would describe blocks like the Storm as being anything other than "constrained submerged jet impingement".

Originally designed it, modelling it with traditional JI theory, which strangely enough seemed to work well enough for estimation purposes, ...

1) It would be interesting, but dunno whether still alive.

2)It would also be interesting to see the "modelling it with traditional JI theory".
I cannot get a handle on the h(conv) to h(eff) relation in these or the Storm wbs

[bobo5195]

storm is jet impingment. Its confined jet impingement. Strickly speaking most of the cooling comes from recirculation zones (based on a paper on this kinda thing) at the edges of the jet but its confined jet impingement all the same. The same design was used in the 1980's for supercomputers and it was called jet impingement then, if thats old skool enoguh for these forums.

Sieder tate is one rule but there are many others, les. Beside common sense says just look at the block think about whats the fluid is doing.

[Les]

Yes that is the pretty picture.
However the proof of the pudding is the h(eff).
Need relation between h(conv) and h(eff) before can consider various possible models.
Only Cathar,in our community, has calculated this relationship.

[Cathar]

Originally Posted by bobo5195

The same design was used in the 1980's for supercomputers and it was called jet impingement then, if thats old skool enoguh for these forums.

Similar, but not the exact same thing.

[bobo5195]

I have a paper which says its the exact same design as a storm (drilled cylinders and jets, confining the flow and creating recirculation) admittedly i think the holes are reamed for a flat base and geometry is different.

oh and the various other imporvements and stuff

[Cathar]

i.e. similar but not the exact same thing.

[bobo5195]

yeah i know meh, should of got my facts straight.

Anyway big pump on the way YAY.

[Les]

Originally Posted by Les

I think Jet Impingement is a Myth

I am probably wrong
As the one who still uses Flomerics Jet Impingement h : r values in all cases and irrespective of r's direction, I should probably call the cooling Jet Impingement.
With both Cathar and Bobo recognising the major role of "high turbulence mashing" and "recirculation zones" there is probably little difference in our approaches.
I yield.

However would still like a handle to h(eff)'s relationship with h(conv).
For Pinned and Finned wbs I am happy to use Fin Efficiency(Fe) Calculator and crudely model as described here{Link)
Am still at head scratching stage for cups/honeycombs.
Access to Cathar's solution would assist considerably

The goal of the exercise is to to join up the models for Waterblocks and Heat Dies, as tentatively considered in the attachments to this stalled analysis(Link)
Combined this should give measured estimations of h(conv)

[Cathar]

My "solutions" were really little more than a set of corrective adjustments to Flomerics style calculations, and applied to a primitive (by proper CFD terms) simulator. I used such to assist in the guiding of what aspects of the designs to change, but ultimately it really came down to a mix of emperical data guided by "closest applicable theory". Seemed to predict which way things would go fairly well, but no, by no means a scrutable model in an correct engineering sense, as frustrating as that may be for some.

I had already explained previously that nutting out the full-on theory for such wasn't my goal, just to use what I could and attempt to "map" it well enough to predict behavior. Ugly as all get out to the real pros, but seemed to work well at the end of the day.

[Les]

These are your solutions? (Link):
White Water x 1mm channels: ~67000 W/m²K
White Water x 0.8mm channels Prototype: ~71000 W/m²K
Free (un-nozzled) Jet Array Against flat base-plate (1mm jets) Prototype: ~55000 W/m²K
Mini-cupped free jet (un-nozzled) Prototype: ~65000 W/m²K
Cascade: ~72000 W/m²K
Cascade SS: ~74000 W/m²K
Cascade XXX Prototype: ~80000 W/m²K
Storm/G1 Prototype: ~50000 W/m²K
Storm/G3 Prototype: ~65000 W/m²K
Storm/G4: ~77000 W/m²K
Storm/G5: ~85000 W/m²K
Swiftech STORM (G4 Rev2): ~83000 W/m²K
Storm/G5 w/ G4 Rev2 optimisations (theoretical): ~91000 W/m²K (projected estimate)
Storm/G5 w/ G7-level optimisations (theoretical): ~94000 W/m²K (projected estimate)
Storm/G7: ~105000 W/m²K