I picked up a Wolf type Sioux 25 (litres) unit some years ago and have been very happy with it although as Jason has said, it does tend to make you jump when the motor kicks in. I think the current Sioux models are 50 litres in capacity. They are very well made and come with two outputs, one of which is regulated, the other being at reservoir pressure.

The tank size will determine how long the engine can run before the reservoir pressure falls to the point where the compressor cuts in again.

A small hole will pass a LOT of air in avery short space of time.

Air is VERY low viscosity.

Supplying air at 5 psi to a 1/16″ hole will cause a 1 .5 hp compressor with a 50 litre tank to tun on a 50 /50 duty cycle I E it takes little time to drop the tank from 120 psi to 80 psi.

I use a welder’s argon flow regulator on my large compressor. I also have a rechargeable tyre inflator that can be used for short runs if all you want to do is exercise the engine or show it working to someone for a couple of minutes.

Martin C

The compressor Jason linked to appears to be a diaphram type. Still reciprocating but a lot quieter for lower pressures. Centrifugal compressors are also an option. They tend to make higher frequency noise that may be easier to suppress.
A toolmakers clamp on a flexible pipe will work as a low pressure flow regulator. It may damage the pipe so don’t use the pipe on high pressure afterwards.

You can easilly work out the required volume of air by multiplying the swept volume of the engine  by the speed. Keeping to basic units, volume in litres and speed in RPI will give a result in litres per minute.

Robert.

For silent compressors you could try converting a redundant fridge compressor.  Some firms use these as so called silent compressors. I’ve not tried it but it can be done. Noel.

Strikes me an air pump to drive small engines would make a good project, though it might be difficult.

The ideal pump would be small, quiet, and electrically driven, with a minimum of adjustments.   It should be able to maintain an output pressure of up to 10psi.   (I guess most model engines need less than 10psi.)

So far, so good, but a vital parameter is missing: it is the required air-flow, that is the volume of air needed to turn the engine within a given time.  It is measured in litres per second, or cubic feet per minute.   Although the engine requires a certain pressure to start, a certain steady volume of air is needed to keep it running.   The volume of air needed depends on the swept size of the cylinder and how many piston strokes per second, plus a hefty allowance for leaks.   At the same pressure,  a big multi-cylinder engine will need a much higher airflow (litres per second), than my little Pottymill!

Calculating the air requirement is similar to sizing an electric motor, where air pressure is equivalent to Volts, air volume is equivalent to Amps, and pressure * volume / time is equivalent to watts.   The relationship is closer than might be imagined, for example if it’s found that an electric motor takes ‘n’ watts to drive an unpowered steam engine in reverse, effectively measuring the engine’s mechanical efficiency/friction, then an air compressor motor to drive the engine must be significantly more powerful than that.

I’d bet money that apart perhaps from Jason, no-one on the forum has calculated how much air an engine needs to run, or – even better – measured it.   Hence most advice about pumps for this purpose is guesswork, or ‘this worked for me’, rather than thought out.  I’m guilty too!

Understanding that a pump for this purpose has to deliver moderate pressure at highish volume helps understand why many alternatives don’t work well:

• A car foot pump produces slow high-pressure pulses of air in low volumes.
• An electric type inflater produces fast high-pressure pulses of air in tiny volumes.
• Fans move high volumes of air at low pressure.
• Refrigerator compressors are high pressure, low volume, and expect to work into a constant load.   Many of them use the refrigerant  as a lubricant, and don’t last long pumping dry air.
• Aquarium pumps get fairly close to the required pressure * volume relationship, but they’re generally too small to run model engines.

The ideal pump doesn’t seem to be available off the shelf, the closest I’m aware of being the larger diaphragm pumps used to power artists air-brushes and paint sprays. Maybe a rubber boat inflater. Not cheap.

What does work is a bog-standard workshop air-compressor, also useful for other things.   An unsuitable high-pressure low volume per stroke pump runs automatically to keep a reservoir at up to about 20bar and some compressors just deliver that;  the volume released is only constrained by the pipework, so whoosh!   Domestic compressors are more likely to feed the air out through a pressure regulator, which can be preset to ensure the hose pressure doesn’t exceed, say 0.5Bar.  Like a gearbox, the reservoir and regulator transform the mismatch between the pulsed high-pressure air input and the what’s needed at the output.  There is a potential problem! Regulators reduce airflow as well as pressure,  so depending on their design it may not produce an exact match to the engine.   In practice, they seem to work well enough.   The problem is size and noise.  Even ‘quiet’ workshop compressors are too loud in my house.

I feel an experiment coming on.   What’s the easiest way of finding how much air (volume) is consumed by a real model engine?   If I can find the time, I’ll run my PottyMill inside a sealed deflated plastic bag, and see how long it takes for the exhaust and leaks to fill a bag of known size.  Not difficult to measure the rpm at the same time.    Knowing the input pressure and RPM, and the output volume over time, would give me a strong clue as to the ideal type and power of pump needed.

Then all that’s needed is to design the pump!

Dave

SOD, you missed a couple of types.
Centrifugal. Quite a good match but generally need to run all the time the model is running
Side channel. Similar to centrifugal but quieter and fairly rare
Screw or “rootes”. Normally used for high volume medium pressure and generally to large for our uses.

Robert.

On 26 April 2024 at 16:47 JasonB Said:

SOD, if you want some numbers to crunch for various engines.

 

…

Your Potty Mill 0.38cu in displacement per rev say 60rpm then 23cu in /min plus your leak allowance,  I don’t seem to suffer from that😏

Lucky you!  Now I’m getting older I have to allow for leaks.   Nurse is never about when I need her urgently!

🙁

Seriously though, surely all engines must leak a bit?   Imperfect piston seals, poor bore finish, tapered cylinders, and especially the valves.   I’d expect D-valves to leak even when they’re forced hard without lifting against a dead flat surface, difficult to do in a model.

Again, as an experiment, be interesting to see how much air an engine built by me leaks compared with the same model built by Jason.  I advise no-one to bet on my efforts!  I build engines to learn technique, and just tweaked them until they ran reliably – the fit and finish of the sliding parts are far from perfect.   I did notice that it’s easier to get a model steam engine to run if the fit is left loose rather than tightly engineered to a high tolerance.   Close fits are more likely to jamb…

Dave

It’s totally impossible at our GP. I’ve asked several times and been fobbed off. “have you got any symptoms”, well as far as I know I don’t, but my cousin didn’t, and he turned out to have prostate cancer which had spread. They also go into long explanations that PSA tests are unreliable. Our GP surgery regularly comes bottom of the league table in our town, but no-one does anything about it. The only way to make contact is with a ouija board. I think it’s time to switch.

On 26 April 2024 at 21:26 noel shelley Said:

Unlike Duncans cousin I was lucky ! The docs did a test…

Me too.  Aged about 55 I developed a bunch of unpleasant symptoms that could have been due to any of several possibilities, hence hard to diagnose.   My GP ordered a barrage of blood tests, and only the PSA test identified a problem.

I didn’t believe it, but a biopsy showed the alert was all too real, and my prostrate was quicky removed.   (An option only available to fitter men: I wouldn’t risk it now.)    After the op, my blood was retested once per year, and after 5 or 6 years, went positive again.   The return was treated successfully by radiation, although successive blood tests are not quite zero, suggesting there might be trouble again later.

I consider myself extremely lucky, because the disease started and was caught early.  I am the type who toughs minor ailments out, so I might have waited far too long before seeing the doctor.

My original complaint turned out to be to an inner-ear problem, very uncomfortable, but in no way life threatening.  In contrast, what was going on in my prostate was a killer, and was causing no symptoms whatsoever.

Those who believe the NHS is over-funded, and that improved efficiency is the answer, might consider that running a large number of PSA tests is expensive, and is a waste of money when most of them come back clean.  Therefore, it’s arguably more efficient to not PSA test until some other symptom suggests prostate cancer is afoot.   Great idea, and many surgeries adopt the policy to save money, except late diagnosis is highly dangerous to the individual, and treating advanced disease is always more expensive, so the total cost  testing may be higher : the GP saves money, leaving a hospital cancer unit to pick up a bigger bill a few years later.   Any fool can call for efficiency savings, but once the facts are known, efficiencies are usually seriously hard to deliver.

As a result, patients might have to push to get a PSA test done, but my experience suggests having one is a sensible precaution, especially if there’s any hint of waterworks trouble.

Dave