Concertina Pneumatics Lab




It wasn't long after starting to play concertina that I started wondering about how these things work, what sorts of pressures and flows are involved, etc.  And it wasn't very long after starting to do some experiments to find out that I realised the need for some basic measuring devices, set up in some convenient arrangement.  I needed a "Concertina Pneumatics Laboratory".  Surprisingly, Ebay didn't seem to offer such a thing, so I decided to build my own.

Of course one quickly runs into the old paradox that faces all researchers - you don't know how best to approach the task until you've completed it, and I have no reason to hope for better luck.  The only precaution I could take was flexibility.  Only time will tell if I was flexible enough.

"Cheap" and "Dirty" were my other watchwords.  I'm not being paid for this; indeed it's taking away time I should probably be spending on more profitable endeavours, like making flutes or fixing carillons.  Further, as fancy construction and fine materials are not going to add anything to the quality of research, it's out the window with them.  So, if you are prepared mentally for an abomination, we can risk uncovering it to your naked gaze....

There it is, in all its primitive splendour.  I'll walk you through what it comprises....

Pressure Regulator

On the left we see a typical air pressure regulator.  It takes air from the workshop compressed air line and reduces it down to safer pressures.  We'll be using it just off the bottom of its range.  This isn't Mythbusters - we're not looking to blow up concertinas here!

Ball Valve

Not shown, because it hadn't shown up in time, is a ball valve to disconnect the whole system from the compressed air line.  It will go between the regulator and the bright-red quick-release connector protruding from the regulator.

Flow Regulator

Next we come to an air-flow regulator - essentially a needle valve.  I figure it will be handy to be able to keep a control over maximum airflow rate of which the system is capable.

Low Pressure Regulator

The knob to its right is a low pressure regulator, working in the 0-1.5psi (about 1000mm water column) range.  It's ratted from my "Magnehelic Flute Leakage detector".

Pressure Gauge

The big round dial is a Magnehelic Differential Pressure Gauge, 0 - 10" water column, also part of the original Flute Leakage detector.  As we'll be favouring metric measurements, we'll regard it as 0-250mm of water. 

Flow meters

Then we come to the first of 4 air-flow meters.  Well, not just meters, but combined meters and controllers.  The knob (at top or bottom of each) operates a needle valve, which can be used to set flow very finely.  Or backed right out so it doesn't influence the airflow in the system to which it is connected. 

The meter itself consists of a tapered bore in a block of acetate.  A bead lives in this bore and is blown upwards by the air flow through the bore.  The heavier the bead and the steeper the taper, the more airflow is needed to push the bead up the scale.  Wonderfully "first principles".

Why four?  To cover the range of air flows we can expect to encounter.  The four flow meters are:

  1. The 1 SCFH (Standard Cubic Foot per Hour) measure that came with the Magnehelic Flute Leakage detector.  We'll regard it as a 0.5LPM (Litres per Minute) meter.  I found it generally too sensitive for concertina work, other than measuring leakage through one pad at a time.

  2. A 5 LPM meter from China, ten times less sensitive than the first one.

  3. A 20 LPM meter also from China, four times less sensitive again, and

  4. Another one of the same, to give us up to a total 45.5LPM capacity if we need it.  That's a winecask full of air every 6 seconds!

Electric pump

At the extreme right, lurking behind its white on-switch, is the aquarium pump that powered the Flute Leakage Detector.  It's rather underpowered for concertina research purposes, but I want to be able to reconstitute the original arrangement for flute testing.

U-tube Manometer

Not built into the pneumatics lab box because of its size is your good old school laboratory manometer - essentially a U-tube half filled with water.  Another wonderfully "first-principles" tool.  Some blue food colouring makes the water easier to see, and an in-built rule makes it easier to note readings.  You can also see that if you even momentarily overload it, the blue water makes a bit of a mess on the base of the instrument!

The Magnehelic pressure gauge is more convenient than the manometer, but if you want to measure pressures in several parts of a system at the same time, manometers are cheap and easy to build.

Digital Manometer

Not shown in the image above is a digital manometer, again from China.  Neat little hand-held unit, it can measure up to 1400mm of water (10psi), and read out in any of 11 units of pressure.  Connecting it up with the other two pressure gauges, they all agree to within 2%, plenty good enough for our purposes.  Reassuring.

And inside?

You might think that the most intriguing part is yet to come - how is it all connected up inside?  You'd be disappointed - there is absolutely no interconnections inside.  All of the items are brought out to that row of tubing connectors along the bottom of the panel.  The aim of the lab is to give us flexibility - different setups can be constructed at will and within minutes using scraps of poly tubing, T-junctions and anything else we find necessary.  And I can quickly reconstitute the flute leakage detector when work calls.

Room to move

You'll notice some blank panel space to the right of the flow meters.  Room to add more devices if necessary.  We really don't know where this is leading!

Real-world adapters

And included in the image, in front of the lab, is an adapter to connect to the end of a concertina.  It's a reminder that we are going to need all sorts of adapters and jigs to be able to measure the various parameters we expressed interest in at Concertina Issues to be investigated.  But no rush and no problem.  We have a well-equipped workshop in which to make those devices, and we can add them as needs dictate.

And it works!

You've probably noticed that the blue water is not at rest, but in fact is indicating a pressure of 254mm of water (the difference in height between the two columns).  The Magnehelic pressure gauge is agreeing, calling it 10".  The air supply is the internal aquarium pump (via the white switch - down is ON in Australia), regulated by the low pressure knob.  I've also tested the compressed air line system, and can confirm that we can supply 40 or more LPM under controlled conditions to the hapless test subject.

Well, it almost works!

If you look very very closely at the flowmeter directly to the right of the pressure gauge, you might notice something distinctly odd.  The tiny bead is sitting just above the 0.2LPM mark.  But where is this cupful of air per minute going?  The output of the flow meter is going to the manometer, and the manometer water level is stable.  It sure ain't getting out that way.

I quickly eliminated all the obvious solutions, eg, a tube connection is leaking.  So we're looking for something not quite so straightforward.  Demonic infestation, for example.   Then a small light began to glow in the back of the mind.  Could this be an "Alternating Current", rather than a "Direct Current" problem?  An aquarium pump doesn't provide a constant flow of air, but a continuous series of pulses, as the vibrating reed inside operates a simple pump.  Who knows what a floating-bead type flow-meter thinks of that?  And its response is likely to be dependant on the nature of whatever is connected.  Not Good!

And this is not just an inelegance.  If we assume that the pump waveform is similar to a half wave rectified mains signal, the crest factor is 2:1.  That means that the peak pressure would be twice the average pressure.  Our manometer or the Magnehelic pressure meter might be measuring 250mm of water, but the peaks could be reaching double that.  The inertia of the air in the lines will tend to even out the peaks but even so, we might be compromising our measurement accuracy by testing say pad leakage at pressures the pads and springs were never intended to have to deal with.

And we know it's real - I was able to prove the pulsing nature of the pump simply by putting the end of the tube from the pump into my mouth.  Drrrrrrrrrrrrrrrrr it vibrated, as my cheeks slowly filled.  Surely, we need something to iron out those pulses into a (relatively) constant flow.  We need the pneumatic analogue of the electrical "reservoir capacitor" you find in every power supply circuit.  A plenum.  ("Plenum" is Latin for "full", and is the opposite of "vacuum".)

Easily made, about 350mm length of 50mm diameter poly water pipe, capped off at both ends, with the joints taped over, just for luck.  A simple tube connector on each end.  Popped into the (previously) empty box and cut into the pump line.  Back round to the front, turn it on, problem solved!  Pop the tube in your mouth, and your cheeks inflate smoothly without the vibrating effect.  Lovely.  I think we're there.


Hmmm, now we have no excuses.  We're going to have to do something!

Back to McGee-flutes Index page...

Created: 18 June 2014