Blowing machine

The Ultimate On-Line Whistle Community. If you find one more ultimater, let us know.
trill
Posts: 685
Joined: Thu Aug 17, 2006 8:44 pm

Re: Blowing machine

Post by trill »

So, I've been trying to visualize the flow processes . . .

Image

I'm wondering now about the role of entrainment.

Reminds me of "added mass" for underwater bodies. It gets dragged along.

I'm wondering if that "dragging" adds resistance.

Speculation, of course.
david_h
Posts: 1735
Joined: Mon Aug 13, 2007 2:04 am
Please enter the next number in sequence: 1
Location: Mercia

Re: Blowing machine

Post by david_h »

Yes, I think something like that. Maybe the energy saved by not stirring up as much atmosphere at the lower velocity exit from the tube was more than that lost to friction along the tube. The 5 degree cone that would be on the windway exit if it was a venturi tube would presumably reduce turbulence even more.

I wonder what Terry would find by probing the air round the end of the longer extensions with a connection to the manomoter. Still locally low because air was being pulled into the turbulence or high because air was being pushed away from the exit?

I also wonder if the asymmetric enclosure of a whistle head means the jet won't go straight for what is in line with the windway exit. Though maybe once resonances in a head (even without a tube) come into play it might not be significant.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Blowing machine

Post by Terry McGee »

trill wrote: Tue Mar 28, 2023 11:57 pm Any chance you could run it out another 100 or 200 mm ? Out to 220 or 320 ? Big increments (25 or 50mm) should be fine.

I ask in the the interest of observing the effect of "skin friction".
OK, I thought to take in the big picture first, 500mm, cutting back in 100 then 50mm increments.

Code: Select all

4 x 30mm Calibrators plus variable lengths of 13.3mm tubing, 30 March 2023			
+ Length	Flow	MM(H20)	Resistance
500	20	69.5	0.42
400	20	66.5	0.41
300	20	67	0.41
200	20	68	0.41
150	20	68	0.41
100	20	67.5	0.41
50	20	68	0.41
0	20	76	0.44
			
Average Resistance			0.41
Bit of an odd droop at 400 and kickup at 500, but they're probably well beyond our interest. Otherwise not much change visible beyond 50, and a sudden change when that 2" piece of tube is removed. Those figures appear largely consistent with the earlier set, although it of course shows more detail as I was taking finer increments. (Could be useful to superimpose those data on the graph you presented.)

Interesting to listen while doing it. With the long tube, not much sound issueing. As the tube is shortened, the sound gets louder and shriller.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Blowing machine

Post by Terry McGee »

david_h wrote: Wed Mar 29, 2023 4:11 am I wonder what Terry would find by probing the air round the end of the longer extensions with a connection to the manomoter. Still locally low because air was being pulled into the turbulence or high because air was being pushed away from the exit?
Had a quick look.

With a 100mm extension, pressure at the open end was about 1mm. Push the manometer tube in about 70mm (so 30 from the end of the calibrator) gives about 2mm.

With the 50mm extension, pressure at open end again about 1mm. Go in 20mm, ie -30 from calibrator end, pressure also rises to 2mm. But it's getting very noisy, and the poor old manometer is struggling to cope.
trill
Posts: 685
Joined: Thu Aug 17, 2006 8:44 pm

Re: Blowing machine

Post by trill »

Terry McGee wrote: Wed Mar 29, 2023 5:49 pm . . . OK, I thought to take in the big picture first, 500mm, cutting back in 100 then 50mm increments. . . (Could be useful to superimpose those data on the graph you presented.)
Here it is:

Image

Well, not what I was expecting. Oh well... was just curious. Thanks for running+posting.

Also: interesting observation about the how the sound changes with length. I wonder what's happening with the flow ?

Also x2: Thinking further about the original (blue) data

I can think of 3 processes at work here:

1) entrainment
2) expansion+mixing of the plume/jet coming from calibrator
3) expansion+mixing of the flow when it emerges from the added-tube.

All 3 are influenced by the presence and length of the added-tube-wall.

I think the reduction in entrainment agrees with the reduction in resistance as the tube length initially grows from 0.

I don't yet have an explanation for the dip + rise/recovery.

Still pondering . . .
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Blowing machine

Post by Terry McGee »

Yeah, me too. I was particularly bothered by the two different values of pressure noted at 100mm overhang. So I went and zoomed in on the 0-150mm area, with a new piece of tubing (spare no expense!), and cutting it off at 10mm increments. Needless to say, that gave me a third value for 100mm! And slightly different results all over. So we might be up against a repeatability issue. Which could be temperature, how long the compressor has been running, flow gauge inaccuracy, operator error, etc, etc. Anyway, see what it tells us.....

Code: Select all

4 x 30mm Calibrators plus variable lengths of 13.3mm tubing, 30 March 2023			
+ L	Flow	MM(H20)	Resistance
150	20	66	0.41
140	20	68	0.41
130	20	68	0.41
120	20	69	0.42
110	20	69.5	0.42
100	20	70	0.42
90	20	68	0.41
80	20	69.5	0.42
70	20	69	0.42
60	20	69.5	0.42
50	20	70	0.42
40	20	71	0.42
30	20	72.00	0.42
20	20	71.50	0.42
10	20	79.50	0.45
0	20	79.50	0.45
			
Average Resistance	0.42
trill
Posts: 685
Joined: Thu Aug 17, 2006 8:44 pm

Re: Blowing machine

Post by trill »

Terry McGee wrote: Thu Mar 30, 2023 12:02 amYeah, me too . . . So I went and zoomed in on the 0-150mm area. . . Anyway, see what it tells us . . .
Yikes !

Image

My head hurts . . . :-?

Well, the "scatter" is on-the-order 1mmH2o. Hmmm . . . a butterfly flapping it's wings ?

I just found the link to the manometer manual provided by Stringbed

Lists accuracy as "+/- 0.3% FSO".
If full-scale is ~1400mm, I get +/- 4.2mm.

Hmmm . . .
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Blowing machine

Post by Terry McGee »

Yeah, we're probably up against it with both the flow meters and the manometer being less than ideal. But there may be more. I'll have a sniff tomorrow to see if I can spot any sources of errors, eg hysteresis, environmental issues, etc. I'll start with calibrator only. We may then need to redo some of our recent tests. Do feel free to suggest tests that are worth running if anything occurs to you.
david_h
Posts: 1735
Joined: Mon Aug 13, 2007 2:04 am
Please enter the next number in sequence: 1
Location: Mercia

Re: Blowing machine

Post by david_h »

Hmm. Interesting!
I am not sure the flow change is important for these small changes in pressure. I wonder if we should regard the regulator as a flow control - through the whole system including itself - rather than being a pressure control. So we are looking at the pressure difference across the calibrator when we are forcing 20l/min (or whatever) through it (which doesn't change the modelling). It's resistance does change the flow but a few mm of H2O due to tubes on the exit isn't much compared to the difference between the compressor reservoir and atmospheric.

But I haven't really though about the physics of that!
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Blowing machine

Post by Terry McGee »

Just ran an initial test, just with the 4 x 30mm to air, to look for any sign of hysteresis and to get a feel for repeatability. The first test is on reaching 10L/Min, the second on reaching 20L/Min, then overrunning that and bringing it back to 20L/Min, then dropping it down to 10L/Min again before dropping to zero. Then repeat 5 times.

Code: Select all

4 x 30mm Calibrator to air					
		Run 1	Run2	Run 3	Run 4	Run 5
Up to 10	19	18	18	18	18
Up to 20	79	75.5	75	78	77.5
Back to 20	78	79	75	80	78
Back to 10	18	17.5	18	18	18
I don't think I'm seeing signs of hysteresis, which would show as a consistently different reading going up and down. The 10L/Min readings are pretty repeatable, but more noise on the 20L/Min readings. You see that in the flowmeter too, with the ball dancing about. I don't know if that's just the flowmeter ball trying to decide where to be, or if its signs of turbulence somewhere. I don't know if some capacity added to the system (eg a 2 litre plastic milk bottle) might slow some of that down....

A little change introduced to the Calibrators - a small ring incised all round 5mm in from each end. "The Insertion Guide", to help standardise insertion to 5mm.

I'll rerun this experiment a bit later, leaving the compressor going to let the system stabilise.
Tunborough
Posts: 1419
Joined: Sun Dec 05, 2010 2:59 pm
antispam: No
Please enter the next number in sequence: 10
Location: Southwestern Ontario

Re: Blowing machine

Post by Tunborough »

Terry McGee wrote: Thu Mar 30, 2023 5:31 pm I don't think I'm seeing signs of hysteresis, which would show as a consistently different reading going up and down. The 10L/Min readings are pretty repeatable, but more noise on the 20L/Min readings. You see that in the flowmeter too, with the ball dancing about. I don't know if that's just the flowmeter ball trying to decide where to be, or if its signs of turbulence somewhere. I don't know if some capacity added to the system (eg a 2 litre plastic milk bottle) might slow some of that down....
That looks like a very useful experiment. I had been wondering about the repeatability of the numbers we've been getting, and this does call the repeatability into question. How much tubing is there between the regulator and the flowmeter, and the flowmeter and the manometer tap? If un-even flow is a problem, perhaps we should be looking at ensuring at least 8 diameters of tubing between elements of the chain.

I checked the expected pressure drop in longer runs of 13.3 mm tubing: at 20 L/min, we can expect 200 mm of 13.3 mm tubing to impose a pressure drop around 0.13 mm H2O, an insignificant amount. We would expect the flow to be mostly laminar, at the bottom end of where it might start getting turbulent.
Last edited by Tunborough on Thu Mar 30, 2023 7:37 pm, edited 1 time in total.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Blowing machine

Post by Terry McGee »

OK, it's a "little later", and the compressor has been left on to top up its tank. Given I could see no signs of hysteresis, and that the 10L/Min results seemed fairly stable, I thought here to concentrate on the noisier 20L/Min flow. So, I ran 20, read the pressure, dropped to zero, then came back to 20, etc. With these results: 79, 78.5, 80, 79.5, 80. Nothing much startling there, I think. It gives us some idea of repeatability.

The dancing ball and the indecision on the manometer continued, so I did cast around to see if there was some sort of filter or at least settling chamber I could plumb in as a test. I found part of an irrigation filter that had convenient fittings. It's not very big - about 6" long with a bore of about 1". And not quite straight through - I couldn't see light through it but it didn't have much resistance to the breath. I tried it first after the resistor, and got 80, 81, 81, 80.5 and 81. Fractionally higher than before but possibly not related. No change in the visible noise.

Then tried it after the flow meter. No apparent effects.

Then tried it in the feed to the manometer, hoping it might make it less indecisive. No change.

And, just to be sure I wasn't chasing gremlins, I plugged one of the many scraps of chopped up pipe littering the bench on to the end of the calibrator - an overhang of 45mm - and the pressure reading dropped to 71.5mm.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Blowing machine

Post by Terry McGee »

Tunborough wrote: Thu Mar 30, 2023 7:29 pm That looks like a very useful experiment. I had been wondering about the repeatability of the numbers we've been getting, and this does call the repeatability into question. How much tubing is there between the regulator and the flowmeter, and the flowmeter and the manometer tap? If un-even flow is a problem, perhaps we should be looking at ensuring at least 8 diameters of tubing between elements of the chain.
Bit hard to be sure, as some of the plumbing is inside the box that houses the various items. But making pretty safe guesses, we'd be in the order of:
- Regulator to Resistor, 200mm
- Resistor to Flowmeter input, 350mm
- Flowmeter output to Manometer tap, 375mm

These would be without the Filter I mentioned above being in any of the lines.

There are two T junctions in the two later lines to allow the second Flow Meter to be connected. I can give you measurements of any of these if needed.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: Blowing machine

Post by Terry McGee »

Oh, and I tried switching to the RH Flow Meter. It also danced at 20L/Min, but at a noticeably higher rate. It shivered, while the LH one blundered. That suggests to me that it's just a natural effect of a dancing bead, not reflective of any system noise.
Tunborough
Posts: 1419
Joined: Sun Dec 05, 2010 2:59 pm
antispam: No
Please enter the next number in sequence: 10
Location: Southwestern Ontario

Re: Blowing machine

Post by Tunborough »

Terry McGee wrote: Thu Mar 30, 2023 7:38 pm Bit hard to be sure, as some of the plumbing is inside the box that houses the various items. But making pretty safe guesses, we'd be in the order of:
- Regulator to Resistor, 200mm
- Resistor to Flowmeter input, 350mm
- Flowmeter output to Manometer tap, 375mm
This should be more than enough to give us "fully developed flow".
Post Reply