pt. I - The Iwaki has landed. pt. II - Plumbing it?

[dogbait]

Thought I'd show off my new gadget, an Iwaki MD-15R(M), (and ask the wise ones here how to plumb it too!).





It's threaded, with 3/4" BSP, and the inner boare diameter is just over 3/8". What I'm hoping to do to get it setup quickly with two Swiftech MCW5000 blocks is to use a 3/4" BSP >> 3/8" barb so that 1/2" OD tubing can be used.

The rad. can take either 3/8" ID or 1/2" ID tubing, but for simplicity's sake, I'd like to plumb the whole system with 3/8" ID tubing and then upgrade later when I add more blocks.

Does anyone know where the above barbs could be purchased?
I've only been able to find 3/4" >> 3/4" or to 15mm or to 22mm (standard plumbing sizes here in the UK). Or perhaps suggest a quick solution since I don't plan to build my reservoir/manifold just yet.

[Blackeagle]

First, welcome to Pro Cooling!


I have the same pump.

I don't have it set up yet, but I'm going a bit differant route. I want to preserve every drop of flow until it hits the CPU block. My set up will go like this.

Pump close coupled to the res with 3/4" copper tube to insure max low resistence flood of the inlet.

The pump outlet I'm going to turn 90 (Rotated on the pump head so it comes out level, NOT a 90 degree fitting on the outlet.) degrees and then run 3/4" copper to the rad which will also be very close to the pump. By close I mean a few inches (3?).

The rads outlet will go into 5/8" copper flexable line. This 5/8" line then goes to the CPU block.

The CPU outlet will go into 1/2" or 58" line to the GPU. The outlet of the GPU goes back into 5/8" line running back to the res.

The Iwaki's are expensive, so I want all the performance I can squeeze out.

The rad & it's fans, res & pump will all be housed in a small separete box that the main tower will sit on top of. The lines will run up into the tower through holes I'll drill for them. The idea is to keep them both as short and straight as I can, along with removing the heat from the rad, res and pump from the tower.

Should result in very low case temps with a minimum number of fans.

There is no way I'd waste any of the pumps flow with 3/8" ID lines. And if your lines are 3/8" ID think about what the ID of the barbs will be. Not good.

But to each his own. Best of luck with your pump!

[Blackeagle]

It would help all here to help you if you spec the case this is going into, or if it will be outside the case.

Knowing the rad and blocks to be used with the pump would also help. Without enough info giving advise is like taking a shot in the dark. That is why I spec'd my intentions rather than try to answer your questions directly.

The reason for my large lines is that the larger the lines the lower the velocity of the water flowing through them and thus lower head loss.

A good starting point would also be for you to read the aritcle by pHaestus on maximizing flow in fittings and lines in the articles section of the front page. A very good article indeed.

[Got KarmA?]

Damn, didn't realize those iwaki pumps were male-thread ... hopefully the home depot will have the parts I need to convert 'em to my barbed connections.

[BrianW]

Yeah just get a dual female piece, then you can attch the male barbs. Althogh I wish Home Depot or Lowes carried a larger selection of male NPT X Barbs in poly....

BrianW

[nikhsub1]

They come with different barb configs. My MD-15R has fixed 1/2" OD barbs, no threads at all. The threaded version usually has a "T" in the part number, like MD-15RT or RLT.

[dogbait]

nikhsub1, in the UK, oddly enough the threaded Iwaki pumps have an 'M' instead of a T at the end of the model number. I believe this applies to all 240v models.

Blackeagle, thanks for all the info on your setup. My system too will go in a separate box, with rad, pump and fans in there. My radiator (Thermochill 120.3) has 3/8" BSPT threads so I can't really widen that, and at the most could use 1/2" barbs on it. I think I most certainly will plug my waterblocks in parallel which should result in a theoretical 3/4" path (2x 3/8").

Here is the layout and specs on the parts for a serial system which I might use to begin with:



The radiator has a 3/8" threaded connection, and so this as well as the blocks is the limiting point, however unlike the blocks (Swiftech MCW5000) this can accomodate 1/2" ID tubing depending upon the size of barbs I use.

To go with a parallel system, two radiators would be required to open up that 3/8" path to 3/4", which would prove an expensive way to reduce restriction:



Not sure whether to use a reservoir immediately, esp. since John Guest sell pushfits with one input and up to four outputs.

Any input appreciated, esp. regarding the effect one rad. would have on flow.

[cahillr]

With the swiftech waterblocks you'd be much better off sticking with serial for the blocks than going parralel, they have pretty low flow resistance and it would help greatly in keeping your flowrates up.

I only went parralel for the first time with mine on my dual system because I was using cascades which are highly flow resistant.
Before this I have always used serial connections on the blocks and have had very good temps, it also helps to keep the inside much neater.

[Blackeagle]

I'd say go with the 15mm barbs which 1/2" lines can streach over with ease.

One thing I didn't spec properly in my first post was my rad. I'm going to be making use of a fairly large heater core of a single pass design. It has a 3/4" inlet & a 5/8" outlet on it. The rad is 9 1/4" X 6 3/8" X 2" for it's finned area with a total length of 11". I"ll have 4 Delta 120mm fans @130cfm in push=>pull, two per side. All 4 of these fans will be hooked up to a reobus of course so I can turn them down to 7 volt for daily use.

Will you do some system flow testing as you set yours up? I'd be interested to see what your flow rates will be. While your rad is considerably more restrictive than mine, my CPU block is more restrictive (Cascade). I havn't purchased a GPU block yet as I'm waiting as long as possible for Cathar to bring his GPU to market. If it gets to the point that I've all else done I may go DIY or perhaps the same Swifty you have.

[Since87]

As cahillr said:

Originally posted by cahillr
With the swiftech waterblocks you'd be much better off sticking with serial for the blocks than going parralel, they have pretty low flow resistance and it would help greatly in keeping your flowrates up.

[The following is cut and pasted from an earlier post of mine.]

Here's a graphs showing some pressure drop vs flowrate curves.

The 'Sim' curves, are graphs of equations that approximate the PQ curves of Iwaki MD20-R and MD20-RZ pumps.

The other four curves show hypothetical cooling loops consisting of a 2-342 heatercore, (single pass, low restriction) 6 feet of 1/2" ID tubing (resistance of the tubing is based on one continous straight piece) and one of the following:

2 White Waters in series
2 White Waters in parallel
2 MCW-5000's in series
2 MCW-5000's in parallel

No attempt is made to account for the added flow resistance of 'tees' or 'wyes' required for a parallel setup.



Two White Waters in parallel have about the same curve as two MCW-5000's in series, so the two curves nearly overlap.

For the Sim20-R, per block flowrate is:
WWS 7
WWP 6.15
MCWS 12.3
MCWP 8.75

For the Sim20-RZ, per block flowrate is:

WWS 8
WWP 5.4
MCWS 10.8
MCWP 6

In all of these cases, series blocks always yields better flowrates through each block. In the high pressure pump case, (similar to cheap pumps in series?) the flowrate advantage of putting the blocks in series is even greater.

Keep in mind that these numbers don't include the flowrate hit for 'wyes' in the parallel cases. Variations in flow resistance in the tubing is not accounted for either.

Looking at the C/W vs flowrate curves for the blocks, it appears that the gain in flowrate would frequently offset the higher water temperature seen by the second block in a series setup. The first block in the series combination would always gain performance from the higher flowrate of course.

References:
Bill Adams' White Water test data.
Bill Adams' MCW-5000 test data.

[dogbait]

Since87, thanks for the data. I gather from your explanation that the gains to be made from a parallel system are negligible compared to a series setup? ( with low restriction blocks)

With the pressure drop much higher for the serial loops, isn't the pump essentially having to work harder and meeting more resistance?

And so wouldn't the pump (having to work harder against more resistance, and generating more heat) dump more heat into the system when in series than in parallel?

I know the effect on the temps is determined by just how much these factors influence each other, but to prolong the life of the pump and reduce heat dumped into the circuit by it, would the parallel be better?

And by how much?

I'm trying to truly understand what BillA has been banging on about in his reviews

[Since87]

Since87, thanks for the data. I gather from your explanation that the gains to be made from a parallel system are negligible compared to a series setup? ( with low restriction blocks)

I also showed curves for White Waters and that the flowrate is also higher in the series case with what few would call a "low restriction" block.

A pump with an unrestricted outlet is actually doing more work (moving water) than a pump with a restricted outlet.

I did some power consumption measurements on a Danner Mag5 and found that the power consumption dropped as the flow was restricted. This was counterintuitive to me at the time so I brought the issue up in this thread. I think most of your questions aside from pump longevity are answered there.

There may be more stress applied to the impeller in the restricted case, but considering the construction of Iwaki impellers, I have a hard time imagining the impeller itself ever failing due to backpressure. (That there is more stress on the impeller in the restricted case, is purely speculation on my part.)