How to test a homemade block?

[Gooserider]

bigben2k Here's a quick diagram (below):

Thanks for the primer BB2k. I just wasn't following the earlier comments on this subject, your diagram makes it alot easier to understand what folks were talking about.

Let me just ask a few questions to make sure I've got it right...

1. The tubing size on the verticals doesn't matter?

2. The height of the verticals should be > the max head on the pump, right?

3. The number of interest for the measurement is the DIFFERENCE between the two colums, not the absolute height of either one, correct?

4. Given the difference, does one then look at the P/Q chart for the pump, and get the flow rate for that resistance?

Somethings occurred to me from looking at the picture, are these proper conclusions?

1. If you just have a setup like this (assuming an independent, non restricted source of water (a swimming pool?)
IN => Pump => Manometer(1) => UUT => OUT
then is the second vertical needed at all? - It seems to me that the second vertical would have a height of 0, so the difference would just be the height of the first?

2. Would there be a way to adjust the flow rate so as to figure out resistances at different pressures?

Perhaps a setup like this: (M = Manometer vertical)

IN => PUMP => M1 => Valve => M2 => UUT => M3 => OUT

It would seem to me like (M1 - M2) + (M2 - M3) would give the total drop for the circuit, which would let you get the flow rate from the PQ chart. If you then subtract the drop from the valve, the remainder would have to be the drop due to the UUT. (Unit Under Test)

If this works, then it seems like a fairly easy way to do a PQ chart for any rad, wb, or other item. It wouldn't have 'Bill A' level accuracy, but it seems to me like it might be good enough for the backyard tester.

Certainly it would fit the 'low cost' standard

I would see some minor errors from plumbing losses (Probably negligible if one keeps tube lengths short and reasonably straight, and also consistent for every test) More of an issue is the accuracy of the PQ chart, since the derived numbers are dependent on it. Does this mean we would have to go back to the bucket test to verify / develop our own pump chart?

Gooserider

[bigben2k]

#1: Correct. tubing size is irrelevant. You should use something that's not too small, just because you want the coolant to level off quickly, but pretty much anything will do.

#2: The height difference will be the pressure drop across the device, for a given flow rate. (see #4)

#3: Correct. The height difference is what you're after. The absolute measurement has nothing to do with anything, other than how much water is in the loop, which is irrelevant.

#4: It has nothing to do with the pump, except if you run this test with the pump (and device, aka heatercore) only. If you do, then the pressure drop that you observe can be looked up on the pump's PQ curve, to figure out what the actual flow rate is.


C1: See #2 and #3.

C2: Correct: you can use more than two tubes, and the height difference is still your pressure drop across that device.

To minimize the other pressure losses, you should keep your tubing straight and short, as much as possible, and use the largest tubing you can. If you know how, you can calculate the typical pressure drop for a given hose ID, at different flow rates.

[Gooserider]

I appreciate your confirmation that what I was understanding is correct - It suggests that I understand the concept pretty well.

I certainly plan to use this type of setup for doing testing on my setup as I get things built. It will probably even get used to test settle the series vs. parallel configuration debate I'm involved with on another thread...

Gooserider

[bigben2k]

NP.

You can convert figures easily, by going to www.convert-me.com , between height in water, to PSI, or whatever you see fit.