FlowBench for 150 to 210 cfm heads - FlowBench Materials

[bruce]

You need to connect the inclined manometer inline like you did before with the vertical, without moving it from the inclined position. Draw 12" on the vertical and you should see 12" on the inclined manometer since that is the height of your scale. Do the same for 10", 8", 6" etc . . . this will prove the inclined is following along correctly. You can then connect it back up the way it should be and do your testing.

Hope I got that all straight its been a long time since I messed with water gages!

[larrycavan]

[200cfm]

Ok, looks like I spent a cold day on a cold bench making an incorrect comparison test. Let me study again what you are recommending and try it again tomorrow. If my understanding is correct, I need to get 12" rise on the incline at the same time I get 12" rise on the Dwyer test manometer but keep the flow manometer on an incline instead of a vertical. Thanks.

[bruce]

Yes, thats whatca need to do to "proof" your incline.

Trust me . . . you're not the first person to spend a cold day on this "disease" called flowbench building!



Edited By bruce on 1169865775

[200cfm]

Ok, Larry, Bruce, others, Spent the entire day making changes, testing, and taking data notes. I believe I followed what you recommended. I ended up adjusting the incline rise so that at 12" wc incline depression the fluid would reach the 30" mark. So I nailed her down at that point. Both manometers were in parallel to the top chamber. And I remarked the zero inch base starting line point.

The results gave better numbers but I came away believing I still have leakage.

Here is how I did the leak test. Please evaluate it and advise.

Blocked off the entrance to the top chamber.
Selected the 1.300" orifice plate. Others blocked.

Start motor A and reach 5.4 " test depression. Flow incline moves from zero inch up to 3/8" mark.

Start motor B and reach 3.0" test depression. Flow incline moves 1/8" from zero base line.

Start A + B C motors and reach 27" test depression. Flow incline pulls to 2 7/8 to 3.0" from base line.

Start A + D motors and reach 26" depression. Flow incline reaches about 2 7/8".

How much leakage would this be from a bench percent or cfm lost?

The rest of the day I tested selecting the three orifices and flowing them with the calibration plate.

Calibration plate is 1.266" dia for 115 cfm @ 28" cfm.

Test 1: selected the 1.950 orifice. Set the vertical test incline to 28" wc. Flow manometer goes to 18 6/8". That should be about 78.9% of orifice flow or 140.9 cfm. Leakage calculates to 25.9 cfm or about 14.5 %.

Test 2: selected the 1.650" orifcie. Others blocked. Flow manometer reaches full 30" mark or 129.9 cfm flow. Vertical test manometer reads 23" wc. That calculates out to about 23.8 cfm if I am doing this correctly. I assume this is leakage cfm gained across the orifice.

Test 3: select the 1.300" orifice. Others blocked. Flow manometer reaches full 30" mark for 100% incline flow. That calculated to 79.4 cfm. Vertical reads 11.1 inches wc.

11.1 wc on the calibration plate is 95.1 % flow or 72.4 cfm. Error now is much much less. Only 7 cfm gap. Why? I assume the gain in the incline reading is coming from the 7 cfm bench leak.

Conclusion: Not sure other than to say I have some leakage.
Why is test 3 close and the others further apart? Am I doing anything incorrectly. To avoid back flow through any motors I have all 4 motors running for the above tests but have the test inches wc under 28 to prevent overflow on test 2 and 3.

Comments, suggetions, thanks. Whats the best leak detect method?

[larrycavan]

I really don't understand your test method completely so it's difficult to assist.

Set the bench cabinet leaks aside for the moment and let's focus on the inclined manometer because if it's not setup right then no readings can be trusted. You can't calculate leakage amounts if you can't trust your guage that is measuring them....start at the start....one thing at a time.

It would be easier to help if you use the spreadsheet to create a % scale for the inclined but if you post the details of the construction dimensions of the inclined then at least I can calculate the corresponding vertical vs inclined readings. At this point the data you've posted can't be verified.

What we're after here is the following.

If for example, you used the following dimensions to make your inclined:

Tube I.D. 0.125
Reservoir I.D. 3
Scale Length 20.000
Scale Vetical Height 12.0
Fluid density 1.000

The spreadsheet would calculate everything you need to know. Your 25, 50, 75 & 100 Percent readings would correspond as follows.

25.0% 0.75
50.0% 3.01
75.0% 6.77
100.0% 12.03

That information is what you are looking to cross reference when you connect the two manometers together. If the readings match up, your inclined is setup properly and you can then trust what it reads.

When you perform that test, you don't even need a plate mounted to the top of the bench. Simply select a range that you can pull 100% on and perform the test. Start by dialing up pressure to reach 100% on the linclined manometer, then look at the verticle...it should read the corresponding value that fits your inclined manometer design spec.

Then back it down to 75%....again it should match....then 50% and lastly 25%....all should match their verticle equivelants...

IF they do, your inclined is good to go and you can THEN begin leak testing and calibration.

So the process order is like this.

1. Construct bench - buy or build manometers.
2. Verify the inclined manometer
3. Leak test and cure leaks
4. Calibrate

You can't mix the process order up and get anyplace but "frustration land"....

NOTE: When you cross connect the guages, it's possible to do it 2 different ways....it depends on how you place the T fitting...one way compares the verticle to the incline and the other way verifies the incline to the verticle....they are not the same thing....I know it sounds kind of bogus but it's not.

[200cfm]

Larry, I can not get the spread sheet to calculate your above % values. For example. Length 20 inches. 25% flow. .25 X .25 X 20" = 1.250" mark.

50% flow: .5 X .5 X 20 = 5.0" mark

75% flow: .75 X .75 X20 = 11.25 " mark

100% flow: 1 X 1 X 20 =20" mark

I just reloaded the spreadsheet and saved it. The only way I can get your calculations is to use the 12" scale height number.

25%: .25 X.25 X 12 = .75

I am confused. I thought the % flow calcualations were obtained from the scale length 20" and not scale vertical height 12". At least the spreadsheet is processing the data that way.

What am I missing here! I wonder if the spreadsheet has become contaminated when filed and saved or changes made an "save as" is executed. Double check my work and see if I am incorrect. I may be.

Thanks for your help.

[86rocco]

[larrycavan]

Rocco's sheet works fine.

I have no idea what your inclined manometer design specs are so I just included and example to show you how it all correlates from % to inches of water...

You have to enter "your" manometer dimensions into it.

Use your scale length, rise, tubing id, flulid sg...etc...Also if you're not going to make a % scale, then at least post what data you entered into the spreadsheet so we can examine your results and assist you...

[200cfm]

Ok, I understand now. Let me do some more homework and study this and I will post back. I am using 1/4" tubing on the incline (.250). Total scale length chosen was 30 inches based on readings that a longer scale is more accurate but if not necessary I can shorten it. And the heigth of the incline rise choosen is currently still 12 inches. And the current reservoir ID is 3.5". But that is an estimate. As for the orifice calculations in the same spreadsheet I go by the posted .62 cd valve and 12.0 wc value, changing the orifice diameter value as needed by the selected orifice from the bench.

I see the need to redesign and improve the current incline design. I need to supplement it with a flow % scale as you suggested. And I need to get a valid reservoir ID instead of an approximation.

Thanks again.

[larrycavan]

[200cfm]

Ok, Larry, Bruce, others. I have made some design upgrades. Have enlarged the well surface area by turning the jar on its side. Replaced the inch scale with mm scale. I can read it much better in cm with the mm as the decimal point. Marked off carefully the percent flow marks. And added a sliding point marker to mark the water rise stop points in the upper range. Borrowed the wifes hem line marker which has its own imported mm scale. The new and improved Version 2.0 so to speak has mechanical memory now. And switched over from 1/4 tube to 1/8 diameter tube.

Made the comparison test manometer to incline manometer readings and I must say, once you get the end angle dead on the pressure readings track with the cm/mm % reading markers for just about any percent. I was impressed at how sensitive the upper range mm change is from a test .05 or .1 inch test water change.

I gave up using the motors to source the depression for calibrating the incline. What worked in the end was Minivacumn hand pump. Allows the test depression to be set carefully and holds the depression while you study the cm/mm incline rise. And gives me time to think better. And served as a double check on the tubing stuff. Some were not as leak proof as I thought. So the tube plumbing got upgrade too.

I think I am set up for the bench leak test. Let me know the best 1,2, 3, step approach for bench leak test and what I should be observing. I think I know but want to be double checked by others. Thanks.

[Tony]

Pretty simple really, use a really small flow measurement orifice, like a 1/4 inch hole.

Completely seal the test hole in the bench top, and if you believe in miracles, you should be able to read zero pressure differential (zero flow) across that extremely small and sensitive flow measurement orifice, at full indicated test pressure.

Do not be surprised if there is zero leakage under suction, and a slight measurable leakage under pressure. It is not unknown for the bench joints to spring open and leak slightly under internal pressure, but close up and seal under internal vacuum.

[200cfm]

Tony and others. Made up a 1/4 inch orifice plate as suggested. Sealed the other three plates. Sealed the entry hole at top of bench. Got full 100% incline (76.2 mm) on a test depression 5.2" wc. (two motors.) One single motor pulled
1.7" wc test depression and incline pulled to 39 mm (around 23% mark). I assume this means a "bad bad leak."

Question: With a sealed bench no air should be entering anywhere from motor suction. So any motor or combination of motors should pull their max high rpm and highest inch water depression. So in a perfect sealed bench all the air gets evacuated out of the suction chamber. Now assuming the chamber above the suction chamber is also a perfect seal, then all the air in that chamber should get evacuated too via the .250 inch orifice hole I placed across the two chambers. Would both manometers see this and the sum of the two manometer provide a total motor stall (no air flow) depression reading?

I am not understanding how a zero leak designed bench with the top entry sealed tight is capable of offering 0" wc incline rise. Shouldn't the incline sense the "evacuation" of all chambers and hence rise to the motor stall depression value?

Can anyone clarify this for me? Thanks.

[bruce]

Both ends of your inclined are reading the pressure difference from one side of the orifice to the other so if both sides are the same (sealed pressure) then the inclined will not move. The vertical manometer is open to atmospheric and reads this difference between inside the bench and outside pressure.