Miniature Combustion Chamber Design Problems

[LarryC]

I am working on a miniature pulsejet design, and have several problems due, in one way or another, to the small size of the combustion chamber (approx .75 in. diam x 2.0 in. long). The exit nozzle (approx .75 in. diam down to .375 in. diam) is rather short (about .5 in.) but as smooth as I can make it. Material is high-strength steel (sim to "chrome moly&quot about .07 in. thick. The engine does not pulse yet, due to inadequate tailpipe length, so all runs so far have required compressed air input (intended for starting only, eventually). This "starting" airstream pulls in fuel by carburetion and also draws outside air in through the intake venturi (i.e. an "ejector" effect). I can get good smooth combustion, and (judging by visible flame temperature), I am achieving lean burning, so I assume I'm running air/fuel ratios of 30 lb/lb or better.

I am not concerned about low thermal efficiency -- I understand that very small combustors are inherently "lossy" due to the large radiating surface for the relatively small mass flows involved. My main problem is that good combustion seems to be mostly happening AFTER the gases get out of the chamber, as evidenced by the red heat of the tailpipe downstream from the chamber exit nozzle (on larger pulsejets, the throat of the nozzle is always the observably hottest zone). Note that, like most pulsejet designs, this is a straight tailpipe equal in area to the nozzle throat, NOT a deLaval nozzle.

Other pertinent data: Assumed maximum achievable exit velocity is approx 1000 ft/sec; assumed combustion chamber velocity approx 200 ft/sec (but this may actually be higher right now due to the "starting" air injection -- smooth running has only been achieved with a crude "flame holder" positioned downstream from the intake venturi to brake the airflow); total engine mass (excluding mounting lugs) 2.5 ounces. The most successful fuel seems to be ethyl ether. Starting ignition is provided by miniature spark plug (Champion V-2); spark is well inside the chamber wall and about 1.25 inch upstream of the nozzle throat (this has no bearing on the problem; after a few seconds, the device runs without maintaining spark ignition).

Any suggestions would be appreciated. One thing I have considered is lengthening the combustion chamber, but of course, that would increase the flying weight.

- Larry Cottrill

 

It becomes increasingly difficult to build efficient, reliable pulsejet engines as you start to reduce their size.

As well as the problems you've acknowledged related to the heat-losses through radiation and convection from the large surface/cross-sectional area ratio, there's also the problem of dealing with gas-flows at very low Reynolds numbers.

Your experiments to date appear to be dealing solely with steady-state combustion, not pulsating combustion. The behavior of the two are quite different and attempting to optimize steady-state combustion will likely only confuse the issue.

First up, there is little to be learned by running the engine without a suitably sized tailpipe with a length/diameter ratio of at least 15:1 (larger engines can operate with an LD as low as 8:1 but small engines need a higher figure).

During the operating cycle of a pulsejet, hot gases in the tailpipe are drawn back towards the combustion chamber by the Kadenacy effect and this tends to provide a small amount of compression (about 20% above atmospheric) while ensuring that the fresh air/fuel charge is retained in the very front of the engine.

When you test the engine by blowing compressed air through it you are naturally blowing the air-fuel mixture out the back so these results will be nothing like those you will see during pulsating combustion.

As for the need to use an aspirated fuel delivery system -- why not consider the use of a pressurized gas such as propane/LPG?

This will provide a much more reliable operation and allow the fuel to be injected directly into the combustion chamber. By using this method you can not only simplify the engine design (throwing away the atomizer) but also vary the power level over a range of (typically) 60% simply by varying the fuel-flow. Within that range, the engine will vary its intake of air to ensure that a stoichiometric ratio is maintained.

I don't know if you've already seen my website at

http://aardvark.co.nz/pjet

but if not then do drop by. I've been seriously researching pulsejets for nearly two years now and have quite a bit of success in improving the SFC and power-outputs of these seemingly simple (but deceptively complex in operation) devices.

 

[LarryC]

Bruce, this is Larry Cottrill, the same guy who's showed up a few times on Kenneth Moller's Valveless Pulsejet forum, where I've talked to you, Bruno and others a few times (I usually seem to be pontificating about things I know nothing about, since my reading of the theory is almost nonexistent and my total experimental time is short). What you're seeing here is a post from fairly early on in my "tiny engine" development, specifically re my Pulsodyne(TM) engine. I've learned a little bit more since then, but at the moment, I'm trying to get my Synchrodyne(TM) design to run (as designed right on the forum in front of everybody!).

At any rate, I did do a bit more with the little Pulsodyne engine. What I found is that the basic problem with "combustion blowout" could be solved by inserting a fairly draggy "flameholder" (or diffuser, if you like) at the front end of the chamber. This created a good pulsing configuration (with the looong tailpipe, of course) that sounds _remarkably_ like my old DynaJet except much quieter (it has the same kind of 'sharp' explosive sound, not at all 'dull' or 'weak' or whatever). I think what happens on tiny carbureted engine attempts is that it's hard to get enough air to drive the right amount (and mixture) of fuel in without overblowing the engine, and this gets fully resolved by providing some serious drag at the front end. Besides achieving pulsing operation, another effect was that combustion was more as I had anticipated -- the heating of the pipe is mostly 'up front', bright red heat extending only about halfway along the pipe and then dropping off considerably (although flame is still just barely visible at the tail end if you're running in total darkness). So, I think my "scalability failure" statement from earlier still basically holds (perhaps, this relates fairly directly to what you said about Reynolds number). I need the chamber longer, which will let me shorten the pipe, possibly a lot. Admittedly, the pulsing was only under the forced draft of the 'starting air', but I think that's explained by my under-sized intake (only about 25% of the tailpipe area, vs your recommended 40% minimum), which I can easily revise upward.

As to using injection, I just don't see it as a valid option in this case. What I'm after is flyable engines in such small sizes that jet work really becomes practical for people who need to build really small, because of cost, available flying space, or whatever. I know there are some little Butane cylinders available, but I don't think that's a very high-energy fuel and even a tiny jet would use it up way too quickly to be practical. You'd need some pressure regulation (a tiny lightweight PRV?) which would add weight. Injected liquid fuel seems out of the question because of pump weight and size, requirement for power (extra batteries?), etc.

I am still convinced that carbureted fuel input is feasible, although I realize it's a lot harder to get than when you have a positive-displacement device like a two-stroke piston mill. Mostly, in the pulsejet, the suction is just a whole lot lower. If the Synchrodyne plan works well (see

http://www.cottrillcyclodyne.com),

it may be the best way to go, rather than a more traditionally aspirated design.

LarryC

 

[GrahamWilliams]

Dear Larry.
What a suprise to see you here.
I too have some questions about pulsating combustion.
Should I use this Forum?
Best Regards
Graham.

 

[LarryC]

Graham -

Sorry to take so long to respond.

Yes, this forum is very good, though for pulsating combustion questions like mine, I think it turns out that the 'Mechanical Engineering other topics' forum might be a more direct path to someone actually involved in this kind of work. On the other hand, if you're starting to deal with 'detail' flow issues like local sonic shock formation or some such, then right here would seem to be the place to start.

Larry

 

[mrwolfe]

Hi Larry.

Pressurised fuel delivery isn't necessarily out of the question. You don't need a miniature electrically powered PRV! In my youth, (LOTS of years ago!), I flew combat control line model aircraft. Pressure feed was a good thing to have to avoid fuel delivery problems caused by varying aircraft attitude. One system took a pressure feed from the crankcase or from the exhaust into a "clunk" tank, or used a pump driven from the crankshaft.

The system which I think is probably useful to you, though, involved the use of an expandable fuel bladder. The fuel tank was made from a baby's dummy ("pacifier" in the USA). Throw away all the hard plastic bits, and keep the rubber bulb. A metal or hard plastic tube was superglued into the neck of the bulb, then the outside of the neck was bound with wire to create a pressure proof seal. The fuel line was then attached to the tube. To use the tank, the fuel line was disconnected from the engine and fuel was injected (typically using a large syringe) into the bladder tank. when the tank was full, the line was clamped off and reconnected to the engine. Fuel pressure is fairly constant until the tank is nearly empty, and fuel flow was regulated using the needle valve.

 

[LarryC]

mrwolfe -

Thanks for your reply to this very old post!

Mr Bruce Tharpe, a model jet designer and flyer here in the US, has described a very similar system [though larger] WITH a lightweight miniature PRV! He makes a pressure tank out of large latex surgical tubing [won't hold up under gasoline, but will work for alcohols and some other fuels, and is easily re-built when needed]. If you have possible uses for a tiny fuel-tolerant PRV, you might find his article on the subject to be of interest -- see the second article in jetZILLA online magazine, Vol 01 Num 01:

http://www.jetzilla.com/Vol01Num01/jetZilla.html

Though I haven't flown any of my engine designs, Bruce Tharpe has many hours of experience flying pulsejets, and claims that this is the best system he's discovered for perfectly regulated, uninterrupted fuel delivery. As in the system you describe, it completely eliminates failure due to small vapor bubbles in the fuel line, if properly filled. The combination of pressurization and regulation gives perfectly smooth pulsejet performance, even during violent maneuvering of the aircraft.

Thanks again,
Larry Cottrill