Pump Flow test results.

[BrianW]

Not sure if anyone wants this data i logged today.

Test: GPM of pump with various components in the loop.

Methodolgy: Once pump primed, pump from one 5 gallon bucket into another 5 gallon bucket until second 5 gallon bucket is filled. Do test three times for each setup.

Pump: Iwaki MD-15RT - Japanese motor
Tubing: 10' 1/2 ID Tubing
Radiator: 2 Pass Core(7 tubes and 8 Tubes) (12.125"X7.875")
Waterblock: Some old Dtek bllock with 1/2" Barbs

Setup 1: 10' Tubing Only

Test1: 76sec, Test2: 74sec, Test3: 77sec ; Avg: 75.66; AVG GPM: 3.96

Setup 2: Radiator and 10' Tubing

Test1: 85sec, Test2: 86sec, Test3: 86sec ; Avg: 85.66; AVG GPM: 3.50

Setup 3: Waterblock, Radiator and 10' Tubing

Test1: 117sec, Test2: 116sec, Test3: 120sec ; Avg: 117.66; AVG GPM: 2.55

Pump observations: Very low vibration, near silent, and casing gets quite hot(read: make sure you ventilate well), yet adds little heat to water loop.

BrianW

[bigben2k]

Thanks Brian!

Got a Fedco# for that core?

Could I interest you in measuring the pressure drop?

[nikhsub1]

Looking at the MD-15R P/Q curve HERE, It would seem the Iwaki is pushing against just under 9ft of head.

[BrianW]

I do not know how to measure pressure drop. The fedco # is 2-354. 9 feet of head is quite alot no? of course 10' of 1/2" tubing is prolly more than I will use when i set it up in the case. I plan on adding three blocksa in series after that heatercore. Do you think it will be enouph flow?

Update on heat: The casing is really hot. Is this normal? The water is still not being warmed signifigantly, but the casing is real hot(read: can not keep hand on it for more than 5-10sec).

Thanks,

BrianW

[murray13]

Yes, the case getting very warm (hot) is normal.

[BrianW]

Well according to the following two charts, the pressure drop for 10' of 1/2" ID tuging is approx. 4'.

Tubing Graph



So that would jive with what I tested according to the PQ curve for my pump. That would estimate approx 2' of pressure drop for the rad. Does that seem right? Obviously If i can reduce my tubing lengths, It will serve to help flow quite a bit. I will have 5/8" Tubing for a large part of my case setup. This should help a bit as well.

Nik, you have this pump in a closed case. Have you noticed the case warming up at all from the pump? Does it vibrate any noise into the case?

Thanks again,

BrianW

[bigben2k]

To measure the pressure drop, you'd need a manometer (see pic below). of course you'd still need to know the flow rate, but a bucket test might give us a rough idea.

[nikhsub1]

Originally posted by BrianW
Nik, you have this pump in a closed case. Have you noticed the case warming up at all from the pump? Does it vibrate any noise into the case?

Thanks again,

BrianW

I used to run it in a Lite-On FS-020 right next to the lower drive cage with NO direct air on it, it was fine. Now, My case has 13 fans... One of my heatercores is directly infront of the pump blowing air on it, stays MUCH cooler, I can actually hold it and not get burned. No vibration whatsoever, I took some mousepad material and cut it to the shape of the bottom bracket, then superglued it on the bottom so the pump sits on the neoprene. The pump is BOLTED to the bottom of the case, no noise at all.

[Agrius]

I see alot of R-30's on ebay. I was curious if the R-15 would still be a better choice due to the heat increase in going with an R-30 or does the flow rate increase negate the heat increase?

[nikhsub1]

Originally posted by Agrius
I see alot of R-30's on ebay. I was curious if the R-15 would still be a better choice due to the heat increase in going with an R-30 or does the flow rate increase negate the heat increase?

Depends on the model... There are 3, the MD-30RT, RZT and RXT. You do NOT want a WMD at all. I would go with the RZT and avoid the RXT like the plague. Here is the P/Q chart.

[DMOS]

This shows why I'm in electrical engineering, and not mech, but why would you avoid the RXT like the plague? Does that change when you go to a high flow block like a Swiftech MCW-5002?

[Since87]

Originally posted by DMOS
This shows why I'm in electrical engineering, and not mech, but why would you avoid the RXT like the plague? Does that change when you go to a high flow block like a Swiftech MCW-5002?

Ok, bad electronics analogy time.

The RZ is like a 10V battery with an internal resistance of 20 Ohms.

The RX is like a 1V battery with an internal resistance of 1 Ohm.

A state of the art waterblock (plus the rest of the loop) is like a 100 Ohm resistor.

You will get a higher current (flowrate) through the 100 Ohm loop using the 10V/20 Ohm battery. (0.083 Amps)

The 1V/1 Ohm battery will only put 0.01 Amps through the loop.

Now if your loop load was 1 Ohm, you would get 0.5 Amps out of the 1V battery vs the 10V battery's 0.48 Amps.

The MCW5002 is still too high in resistance for the RX pump to be the best choice. The standard version (not RX or RZ) would likely be best with the MCW5002. (Depends on the rest of the loop.)

[DMOS]

Thanks, that explains it in terms I can understand.

I just saw a WMD30RXT for cheap ($20), so I was going to snap it up. Instead I'll find another cheap pump.

[Gooserider]

DMOS:
...I just saw a WMD30RXT for cheap ($20), so I was going to snap it up. Instead I'll find another cheap pump.

Note that you do NOT want any of the "WMD" model pumps - there is a reason why WMD's are cheap...

As I understand it, IWAKI pumps come in two flavors, both use the same pump head for a given model number, but the motors differ.

"MDxx" models use a Japanese motor, "WMDxx" models use an American motor. Aside from Iwaki itself, I have never seen anything positive said about the American WMD motors beyond the 'Made in USA' factor. OTOH, I've heard that the WMD motors don't have the same reliability and quiet operation that the Japanese motors do. Judging by the pictures, I think the WMD motors may be slightly larger as well.

Just remember - WMD is bad, MD is good....

Gooserider