Blowing machine

[david_h]

Are all requirements the same? I think Tunborough, who wants to be able to produce a whistle geometry from theory and maybe some empirical factors that might apply to all whistles, has the most stringent requirement.

If one has a whistle, to such design or otherwise, and have measured its characteristics there is a lesser requirement on the model - to predict what geometrical adjustments are needed to make a specific change to its characteristics without changing things we are happy with.

An even less stringent requirement is to be able to describe whistles - pressure, air requirements, tuning, tone, volume etc in a more quantitative way. So we have a better idea how another model of whistle will compare with what we have.

[trill]

@Terry_McGee

. . . The bit that bothers me is that the whistler covers a range of pressures, and if that were to be enough to introduce an error, it will be an error that varies with the pressure, not just a simple offset. . .

Well, our old Gen takes ~200 mmH2o do get D7. That’s 200 out of ~10,000 mm for 1atm. That’s like 2%.

You’re worried ? Really ?

@Terry_McGee

. . . but we’d have to somehow convince ourselves that it can be trusted at the slightly higher pressures than a whistle or calibrator might reflect back. . .

Well, the mfr claims “(2+.5FS)%”.

I have a query into a sales rep.

[david_h]

Theoretical question. A quote from the University of Wollongong paper linked on an adjacent thread.
" To further improve the design, the wind-way or narrow duct through which the breath passes, was increased to allow a higher volumetric airflow rate to account for any wall smoothness issues and to aid in the cleaning process."

Do we then have to blow more air - but at the same pressure so as to maintain the required velocity - or does the increased volumetric flow in the jet somehow compensate? I have been looking online in the last couple of days but didn’t find anything.

[Tunborough]

@david_h

Theoretical question. A quote from the University of Wollongong paper linked on an adjacent thread.
" To further improve the design, the wind-way or narrow duct through which the breath passes, was increased to allow a higher volumetric airflow rate to account for any wall smoothness issues and to aid in the cleaning process."

Do we then have to blow more air - but at the same pressure so as to maintain the required velocity - or does the increased volumetric flow in the jet somehow compensate? I have been looking online in the last couple of days but didn’t find anything.

Yes, you have to maintain the same velocity, so a bigger windway means bigger air flow. Air pressure would be the same or maybe a bit less since there would be a little less loss through the windway.

[david_h]

Thanks. If windway exit to labium distance is changed does that then change the velocity required for the same note?
I am wondering, in the context of ‘resistance’, if an ‘easy blowing’ whistle is one that gets through the air faster even if it has the same ‘backpressure’ or is one that requires less pressure.

[trill]

@Terry_McGee

. . . but we’d have to somehow convince ourselves that it can be trusted at the slightly higher pressures than a whistle or calibrator might reflect back. . .

This just in from a distributor:

"For example, if you have a 100 SLPM model,

at 100 slpm, your accuracy = +/-(2.0% + 0.25%x (100/100)) = +/-2.25%"

There’s no doubt in my mind it would be more accurate that our dancing balls.

No doubt at all.

[trill]

@Tunborough

Yes, you have to maintain the same velocity, so a bigger windway means bigger air flow. Air pressure would be the same or maybe a bit less since there would be a little less loss through the windway.

There’s also the question of “power”. As it watts.

Remember, power = volume flow x pressure.

The breathing muscles (intercostals) will know.

[david_h]

@trill

The breathing muscles (intercostals) will know.

I suspect that proprioception is more idiosyncratic than a flow meter. Breathing muscles are set up to be mainly controlled be CO2 levels, not a wind instrument. One output from this is going to be that if we have a whistle type that Terry has measured we can know what a range of pressures and flow rates feels like.

[Terry_McGee]

@trill

@Terry_McGee

. . . but we’d have to somehow convince ourselves that it can be trusted at the slightly higher pressures than a whistle or calibrator might reflect back. . .

This just in from a distributor:

"For example, if you have a 100 SLPM model,

at 100 slpm, your accuracy = +/-(2.0% + 0.25%x (100/100)) = +/-2.25%"

But that still doesn’t answer the question “If you put the meter in a line where the pressure is (say) 400mm H20, will it affect the reading?”

I wonder if we could download a manual for such a meter somewhere? The manual might have recalibration curves for flow against pressure?

There’s no doubt in my mind it would be more accurate that our dancing balls.

No doubt at all.

It’s certainly offering greater accuracy and resolution at STP. But it would be a bitter blow to pay all that money and find out then that it’s highly susceptible to pressure.

[Terry_McGee]

Now Tunborough, what do you make of my experiment which showed that the 4 x 30mm Calibrator showed 60mm of pressure at 20 L/Min if at the end of the food chain, but 50mm if inserted above the Flow Gauge? Do you still want me to test the other Calibrators above the Flow Gauge?

[Tunborough]

@Terry_McGee

Now Tunborough, what do you make of my experiment which showed that the 4 x 30mm Calibrator showed 60mm of pressure at 20 L/Min if at the end of the food chain, but 50mm if inserted above the Flow Gauge? Do you still want me to test the other Calibrators above the Flow Gauge?

Even more confused. In the earlier numbers, with that calibrator below the flowmeter, 20 L/min produced a pressure of 80 mm H20. Then with the calibrator above the flowmeter, it gave 56 mm. Now it gives 60 mm above, and 50 mm below. Were there differences in the plumbing between the earlier numbers and your latest trial? I am still interested in numbers for other calibrators above the flowmeter, but I better make sure I understand the plumbing.

ETA: For the record, 56-60 mm agrees well with theory; 50 and 80, not so much.

[Terry_McGee]

OK, let’s crack that conundrum first. I’ll go down and do a run at 20L/Min only, on the same Calibrator, above and below the Flow Meter. And I’ll check the plumbing carefully in case something is coming unstuck.

Hmmm, wrap your head around this!

I set up the Calibrator below the flowmeter, with its far end open to atmosphere. The Pressure takeoff is between them. 20 LPM produces 75mm.

Now I have to switch to the Calibrator before Flowmeter setup, but, being suspicious, I do it in stages…

Firstly I add a short length of 13.5mm bore tubing to the open end of the Calibrator. It extends about 30mm. Flow still 20, Pressure drops to 67! I pull it off and put it on again, same result.

I then plug in the T-joiner which will be my lower Takeoff, and the 6" of 1/4" bore tubing I will need to bring it to the Flow meter. So that tube is still open to atmosphere. Flow still 20, Pressure goes up to 83.

I put the Manometer in Diff mode, and measure between the two takeoff points. 59.

I go back to non-Diff mode, and measure from lower takeoff point to atmos, 23. Oooh, 59 plus 23 is very close to 83!

So now I swap to Calibrator first, Flowmeter to air. Go to Diff Mode between Takeoffs, Flow 20, and I’m seeing 48.

(I can’t measure to air at T/O points as they exceed Manometer.)

Which measurements would Sir prefer?

[trill]

@Terry_McGee

. . . It’s certainly offering greater accuracy and resolution at STP. But it would be a bitter blow to pay all that money and find out then that it’s highly susceptible to pressure.

I think I finally understand your concern.

I’ll look into it.

[Terry_McGee]

Good man!

[Terry_McGee]

So my ever-fertile mind (or is that “overly-fertilised” mind?) wondered if I messed around just a little bit with the Calibrator after the Flow Meter could I trick it into playing up? So, here’s the setup:

Pressure Regulator via Resistor > Flow Meter > upper Pressure Take-off > 4mm Calibrator > Lower Pressure Take-off > 6" of tubing > air. Manometer Diff Mode between the two Takeoff points. 20L gives 60mm.
Remove the 6" tube. No change.
Add 2’ tube. (When I lapse back into Imperial, it’s because I’m too lazy to measure. I can obviously measure if it becomes significant.) 20L gives 58mm.
Squished the 2’ tube to reduce flow to 18L, then increase Regulator pressure to get back to 20L. 58mm.
Removed 2’ tube, reset to 20L, 59mm.
Added 6" of 13.5mm tube plus the 2mm by 30mm Calibrator. Flow drops to 18L, Pressure 45mm.
Raise Regulator pressure to restore 20L/Min, Manometer reads 58mm
Remove tube and 2mm Calibrator, reset 20L, 4mm Calibrator back to 60mm.

We do see a little bit of change, but I’m being a bit mean to it, particularly when I add the 2 by 30 Calibrator. At that point the Pressure at the upper Takeoff (both Calibrators) is 1340, and at the lower Takeoff (2mm Calibrator only) 1270. We’ll never see levels like that on a real whistle.

It does seem extraordinary that the 2mm Calibrator has 1270 across it, while the 4mm Calibrator in series has only 70mm. Mud wasps? Or does that just remind us the power of squaring?

[david_h]

Post mainly composed on my morning walk before Terry’s last, but I think it still stands.

@Terry_McGee

Firstly I add a short length of 13.5mm bore tubing to the open end of the Calibrator. It extends about 30mm. Flow still 20, Pressure drops to 67! I pull it off and put it on again, same result.

Was that with it running? Can you do it with the system running and the flow set below meter full scale so you can see how pressure and flow reading move?

I will have a think about the numbers in the last post. However, since we know that 20l/min is not always 20l/min but that 60mm of H2O is always the same (and 60mm of H2O or very close) I think adjusting the regulator to bring the pressure across the calibrator back to the same and reading the flow meter would be easier to interpret than the other way round.

You have got another flow meter, but it is only roughly calibrated - the calibrator open to air (or the low pressure side take-off attachment) and manometer (using both take offs if attached). Adjust to keep the pressure across the calibrator the same whilst squishing the tube between the meter and the calibrator. You can watch the flow meter change with pressure knowing that the flow rate we are interested in has not changed. To find the pressures you could put the water manometer across the calibrator (where you can keep it in range) and the digital manometer on a take-off between the flow meter and the clamp. You might be able to work out a correction table - or at least plot a curve of the behaviour that might help explain any oddities in the Flow - sqrt(pressure) curve (or the Flow -Pressure curve).

[Tunborough]

@Terry_McGee

It does seem extraordinary that the 2mm Calibrator has 1270 across it, while the 4mm Calibrator in series has only 70mm. Mud wasps? Or does that just remind us the power of squaring?

Air speed coming out of the 2 mm calibrator is 4 times that coming out of the 4 mm calibrator, over 100 m/s. Four times the speed takes 16 times the pressure, so 1270 mm H2O isn’t surprising.

@Terry_McGee

Hmmm, wrap your head around this!

I set up the Calibrator below the flowmeter, with its far end open to atmosphere. The Pressure takeoff is between them. 20 LPM produces 75mm.

Now I have to switch to the Calibrator before Flowmeter setup, but, being suspicious, I do it in stages…

Firstly I add a short length of 13.5mm bore tubing to the open end of the Calibrator. It extends about 30mm. Flow still 20, Pressure drops to 67! I pull it off and put it on again, same result.

No idea why a short length of tubing would reduce the pressure. If I’m following you, the reduced pressure is across both the calibrator and short tube, with both the second manometer port and the short tube at atmospheric.

It reminds me that I should have pointed out a lesson from the orifice plate standards. When measuring pressure drop across an obstruction, the recommendation is to measure from at least 2 pipe diameters (25 mm) above the obstruction to 8 pipe diameters (100 mm) below the obstruction. I wonder if we’re getting puzzling results because we’re measuring too close to the calibrator exit.

[david_h]

Is it that flow near a high speed zone is going to be turbulent (so with local pressure variations) and take a lot of pipe diameters to settle down to laminar?

I have only read the Wikipedia page on orifice plates. I find the page on Bernoulli’s equation unhelpful compared with a textbook. However, this bit “A little downstream of the orifice the flow reaches its point of maximum convergence … … where the velocity reaches its maximum and the pressure reaches its minimum. Beyond that, the flow expands, the velocity falls and the pressure increases.” seems relevant and follows directly from Bernoulli’s equation. With a long pipe the flow rates are they same either side of the plate so velocity will be the same. If the pressure is lower downstream where has the energy gone? How does the mud wasp do it?

Is there less chance of getting confused by working straight to atmosphere for calibrators and whistles?

[Tunborough]

@david_h

Is it that flow near a high speed zone is going to be turbulent (so with local pressure variations) and take a lot of pipe diameters to settle down to laminar?

It won’t necessarily be laminar, but the flow profile will be more uniform. Just past the outlet, the speed will be very fast at the outlet, and very slow in the backwater at the inner surface of the pipe. Farther downstream, the speed will be moderate through most of the pipe, except right at the inner surface of the pipe.

@david_h

If the pressure is lower downstream where has the energy gone? How does the mud wasp do it?

Gone to frictional losses from pushing its way through the narrow windway.

@david_h

Is there less chance of getting confused by working straight to atmosphere for calibrators and whistles?

You would think so, but there’s a discrepancy between flowmeter → calibrator → open air, and calibrator → flowmeter → open air, and even flowmeter → calibrator → short tube → open air. The whistles pretty much have to vent to open air, but it would be nice to understand the source of the discrepancy first.

[david_h]

I wonder if the short tube could reduce drag at the exit.

And if it is safe to ignore the lengths tubing.