Speed control of air engines via pressure or flow

[PStechPaul]

I have found that it is somewhat difficult to adjust the speed of my wobbler air engine using a ball valve mounted close to the intake, where I also have a pressure gauge. Also, the gauge fluctuates wildly as the sliding valve opens to admit air and then closes as air is expelled from the exhaust.
 
My thought is that there is very little volume in the plumbing between the valve and the inlet of the engine. Obviously, perhaps, the pressure will build up while the intake valve is closed, and then quickly drop as the intake opens and the volume expands as the piston is forced out of the cylinder. Thus the force on the crank will be large at first, possibly even causing some "shock" as the compressed air is suddenly applied, and then reduce until it reaches a point where the air moving through the valve can start increasing the pressure in the fixed volume at the end of the stroke.
 
I think I will try adding a Tee fitting into a larger diameter and length of pipe with a cap on it, to act as an air pressure reservoir, and it should show as a much steadier reading on the pressure gauge, as well as perhaps a smoother running engine. Then I might also consider replacing the ball valve with something like a needle valve or "pet-cock" with finer control.
 
Anyone have experience with this or suggestions on what to try? Thanks!

[Jim Nic]

I have thought a little about this problem of steady slow running on air with my own engines, especially a wobbler, where I have found it difficult to achieve a good result.  I suspect it is because I run my engines with no load so when reducing the flow of air there is little resistance to the piston turning to the open exhaust position to stop the air going out of the exhaust in rapid order.  If you load up the engine to slow it the pressure in the cylinder will remain higher and steadier for longer and the speed can then be controlled by the flow rate.  A similar effect might be seen by restricting the area of the exhaust passage.
I fully expect a more learned response to come along shortly and blow my theories out of the water but that's my £0.02.
Jim

[philjoe5]

Paul,
A heavy flywheel, low head pressure on the compressor, a needle valve and small diameter tubing (1/8" ID) will allow you to go slow.

<a href="https://www.youtube.com/watch?v=Na7uaZkItS0" target="_blank">http://www.youtube.com/watch?v=Na7uaZkItS0</a>

If you're going to CF I can show you my setup

Cheers,
Phil

[BillTodd]

I have been pondering ideas for air control for a while.

I wondered if carefully timed pulses would be better than either flow or pressure control. Pressure or flow control will work if the load on the motor is proportional to rpm (such as a propeller or other air movement load). Most small model engines (I guess) will have a variable friction load  - they'll be much tighter in one part of their cycle than others . Their relatively light flywheels will not help to smooth the load enough . Then there's the problem of heating, or cooling as the air expands , causing slow changes in the friction.

So...

My idea was to sense the motors rpm (using the air pressure in the pipe to the motor, if possible) then apply variable duration pulses, via a electric valve, timed to coincide with the motor's power strokes - the rpm would then be controlled to a large extent by the electronics, but it should be able to control the motor with a conventional throttle.

Bill

[90LX_Notch]

An automotive leak down detector works well.  http://www.summitracing.com/parts/otc-5609  They have a small regulator built into them.  They are pricey but one can build their own.  Fortunately, I have one from my drag racing days.

This is a video I made six years ago to response over at that other site.  It's Elmer's tiny running down to 2psi.

<a href="https://www.youtube.com/watch?v=hSPRFIamayQ" target="_blank">http://www.youtube.com/watch?v=hSPRFIamayQ</a>

The "pop" at the end is when the tubing popped off of the nipple.

-Bob

[philjoe5]

Bob,
I seem to recall that discussion 6 years ago because I've always been interested in seeing how slowly we can run these air powered engines.  IIRC someone suggested putting a valve on the exhaust side of the engine as well as on the intake and then balancing the valves to get the engine to turn over as slowly as possible.  I was able to get a few of my engines in the 30 rpm range without that trick, so I never tried it out.  Maybe someone else has?

Cheers,
Phil

[b.lindsey]

From a historical perspective, meaning the early industrial revolution, I would think that virtually all steam engines were controlled with flow valves. Even on locomotives, the boiler was brought up to running pressure and maintained there and the engine controlled via a throttle valve. Obviously air is a different medium but my experience has been that with larger bore engines (say around 1" bore) I will see a good bit of deflection in the air pressure gage during the cycle, especially at lower pressures. The PMR#1 I just finished was the first of that size that also has a flow valve between the air regulator and the engine, and after running it at anywhere from 6-12 psi over much of the weekend, when using the flow valve rather than pressure to regulate speed, I notice little if any deflection in the pressure gage needle. Not sure if that helps or confuses the issue more but that is what I have observed.

Bill

[geoff5269]

I have experienced this problem when testing my engines, now I always connect my small compressor via a boiler I have made which is approx. 300cc and it has an extra control and pressure gauge, and the pulses are smoothed out. I have also fitted two regulators to my small compressor and now the pressure stays even as the air runs down. My latest Benson will run at 100rpm on just 1psi  , of coarse when you run at this low pressure you have to be confident that the gauges are accurate but that's another story,
Geoff

[PStechPaul]

I was able to get my wobbler running at about 20 PSI and perhaps 100 RPM, as can be seen in this part of my TigTac video at about 3:30:
 
https://www.youtube.com/watch?v=GUFMYTKVAv0&feature=player_detailpage#t=210
 
It does have a steel flywheel that is much heavier than the smaller aluminum part specified in the plans, and it also has the generator and incandescent lamp load which may help smooth out the strokes. I also have found that the axle seems to be binding or galling which causes it to stall at lower speeds. I have taken it apart, cleaned out the old dirty oil, polished the shaft with abrasive cloth, touched up the bore, and added fresh oil, which helps to some degree. Once it gets graded, I may add bronze oilite bushings or even ball bearings, which should help a lot. I wonder if the fact that my manifold is made of SS rather than 1018 has caused a problem? Or perhaps the fit is too tight?

[b.lindsey]

Paul, I am wondering if a part of the fluctuations in your pressure gage are simply feedback from the vibration of the engine itself. Oscillators are inherently unbalanced engines and even when locked in a vise those vibrations have to go somewhere.

Bill

[Don1966]

Another cause of fluctuations is air leaks. Since the two faces have to slide against each other and may not be properly seated. To get fluctuations out of your lines feeding your engine you can make out of pipe and fittings an accumulator. This serves as a cushion to the pulsing air pressuree because it stores volume. Depending on the size of your engine would determine how big to make one. Geoff was doing this with his boiler in line with his engine.

Don

[philjoe5]

It occurs to me that since this is a single acting engine, these pressure fluctuations are going to be normal.  During the power stroke air rushes in to fill the cylinder causing a momentary pressure drop.  during the exhaust stroke the air intake is blocked causing a momentary pressure rise.  As suggested an air reservoir will dampen the oscillations.

Cheers,
Phil

[tinglett]

What are you guys using for low pressure air regulators?  Perhaps I need to look for a better source, but my current air regulators don't claim to be accurate < 40 psi.  I don't have an additional means to measure, though I could attach a couple to see how they read in comparison to one another.

I was wondering if there is something better than traditional air compressor regulators.  Maybe a regulator designed for an air brush or LP gas or something like that?  I see mentions of getting in the < 10 psi range and wonder what you are using.  I see the automotive leak detector but that's probably not something to go buy unless you need it for its intended purpose.

At some point I'd like to make some sort of manifold with at least 3 or 4 valves.  Any pointers to nice examples would be welcome.  The other comments here are interesting, too.  Perhaps some kind of little "tank" made from water hammer arrestor hardware might do the job for the wobblers provided it's feeding very small air hose?

Todd

[b.lindsey]

Todd,

I have several Mini Air Regulators I purchased from Grainger some years ago. The mini only refers to their size which compares to most hobby type compressor regulators. Most of these should be good for 5-125 psi. I think the more important thing is to get a gage that covers a smaller range that the typical compressor gage. The one I use goes from 0-15 psi thus giving a more accurate view of the actual pressure than you can get from the small 0-120psi gages found on most small compressors. This one is a panel mount type with a face diameter of 2 1/2" as I recall...so both larger in size and a much smaller range of covered psi...ideal for small air engines.

Bill

[sshire]

Todd
Here's mine. Amazingly precise.

http://www.modelenginemaker.com/index.php?topic=3571.msg91177#msg91177