Kazoo engine

[resosys]

Dear Mike.
'Queens Belfast' did some work on a small valveless with a cycle time of only 4.10^-3s. They had to use 4 Propane injection points mounted at 45 degrees to the main flow axis.
I'll send you a copy of the details if you like.

Graham,

I'd be interested in details. I've got a little baby lockwood cut and bent. I'm waiting for a friend to weld it up for me since my TIG skills are not quite there yet. The combustion chamber is around 1.25" with a 19.5" tailcone. It's cute, if nothing else.

Since it's just mild steel, I'll MIG it to death if he doesn't come through soon.

Chris

[Mike Everman]

Ditto on the details, Graham, I'd love to see that. And don't worry, I won't make any decisions based on little engines.
I've made some test bends on a 1", I know, I know. I may do nothing more than spray some Aqua Net in it and touch it off...
I MUST try the 2", I must, I must! Then I'll do a 75 or 100mm, I promise.

[Viv]

Hey thats not bad at all, im impressed! is it just those collars that keep the creases so clean?

Viv

[Mike Everman]

the collars worked great, but the jury is still out. this mat'l is .049" wall, 1" od. You can see on the top picture how it buckles in the middle, bound to be worse when mat'l is thinner. I'll squish some 2" tomorrow, and Graham, I'll get some 3" collars ordered tomorrow too.

[Mike Everman]

It just occurred to me that this bend (the laminar flow U-bend, that is) will be smooth in one direction, and quite turbulent in the other… Imagine a rolling vortex like a tumbleweed down a pipe.
Faaaascinating.

Oh, yeah, I need to correct myself for posterity:

1. I think it's interesting that the surface to volume ratio is in fact constant from end to end, which I find a possible reduction in variables normally difficult to iterate experimentally. When I iterate the basic dimensions, at least S/V will not be changing too.

Of course, I meant that the fortnights per furlong were constant, my mistake. Oh, and the surface area.

[resosys]

Last night I did some layout work on a straight through triangle engine design. I printed three copies onto paper, cut them out, and taped them together to make sure my math and eyes weren't too far off. It looks pretty good, if not a bit odd. I'm really looking forward to welding one up this weekend.

I cut down an image of my baby lockwood combustion chamber and intake tube as it sits right now (see attached).

If anyone has any thoughts on special requirements a triangle engine may need in terms of ratio changes, fuel inlet placement, etc... please post them. It would be nice to go into this thing with some ideas to try during the test phase.

Thanks,

Chris

[Mike Everman]

Aw, Chris, isn't she cute! See attached for another size challenged engine. Most of the guys here will say we're MAD, MAD I TELL YOU! I'm just having fun bending stuff. I learned I've got to make a fixture to do this right, no matter what size I do.

[Mike Kirney]

That looks amazing. Mike, you're amazing. If Larry lived on the Starship Enterprise, he would build something like that. If it works you should make 88 different lengths and build a pulsejet piano. You can save time and make 64 jets for a short scale (like a Rhodes or a synth). What frequency do you expect it to resonate at? Playing some Deep Purple on an internal combustion pipe organ would be a lot of fun I think.

[Bruno Ogorelec]

Looking at the photo, I'd say instinctively, based on years of gazing on various engines, that it is too short. If it's that small, it should be proportionally longer.

[Mike Everman]

thanks Mike. I swear to Budda I'm going to make the intergalactic train whistle! Maybe a Pan flute. Either way, by definition it will be quite harmonic. And that stray dog problem? He he.
You know, if you blow on this thing it whistles! Only from the intake end, though.
Bruno, it is the proper length for a Kentfield style Lockwood, with a .902" CC. I'll check it again.
I got a piezo igniter and some methanol to play with tomorrow...

[Bruno Ogorelec]

Bruno, it is the proper length for a Kentfield style Lockwood, with a .902" CC.

Mike, my I ring a warning bell? You sound as if you had a dimension-predicting program for pulsejets. You don't. You have a good scaling software, but it is only good for pictures. It does not take into account fluid mechanics and thermodynamics. Software that does that is hugely complex. It's been done, but it runs on a few supercomputers at a few North American universities and still takes hours to process the information, so that the researchers have to apply for computer time months in advance.

Milisavljevic has a dimension-predicting method, he says, but I don't know how complex it is. Maybe he has developed something simpler than the one currently in use on the big Cray at the University of Calgary, for instance. You can ask him.

I am saying your engine is too small to run in self-sustaining fashion to start with. The pumping losses will overcome its ability to aspirate. If it is ever coaxed into working, it will be with a longer pipe.

[Mike Everman]

Yes Bruno, and no, I have no expectations that this little tube will run! I will however push some propane through it an see if I can get a POP out of it. Hey, it's kinda the right shape, it's stainless steel, but man, it only has about .4 sq. in. of exhaust area! If there is anything I've learned mining this forum is that pure geometric scaling is pure folly. You're preaching to the choir, my friend. I'm just getting some tube smashing in before I do a realistic size. I'll likely skip the 2" and go to 4.
I'd like eventually to create such a reliable scaling tool or methodology, of course, because small engines are so cool. Maybe useless, buy still very cool!
I dream of an array of high SFC engine tubes in a duct. I'm sure that small means low efficiency, and at the same time not so sure.
Small means high frequency, right? And the consensus is that things just don't happen well too fast, apparently. At the small scale, perhaps you want it anything but self sustaining, and force a lower frequency so the burn will be better. Dare I say "timed ignition"? But timed ignition can be organic and uncomplicated if you let it. If the engine is spring mounted, then it's rebound after the pulse could bump a piezo stack to touch off the next charge, then no matter what the size, you can tune the spring mount to make the motor's mass oscillate at whatever frequency blows your skirt up. In this way you could limit a large number of very small combustors to practical frequencies, without a complex system. What would be good? 120 Hz?

[Bruno Ogorelec]

Well, I guess I'm an unreformed reactionary. I still say, 'there ain't no substitute for cubic inches'. I still prefer a Harley to the Japanese screamers. OK, maybe not Harley; I like fine technology too much to go for Harley's affected anachronisms, but a BMW flat twin, oh, yes.

Two questions.

(1) Off topic. Do your robots have anything to do with biotechnology. biochemistry, pharmaceuticals etc? My daughter is writing a paper on robots in those industries and is looking for some interesting background info. She's majoring in biotechnology. One day, she will be growing pulsejet plants hydroponically in our basement. Kirney can tell you all about it.

(2) Closer to the pulsating topics. I am mad enough to go back to valves. And not even the lightweight reeds but old-fashioned heavy-duty steel poppet valves you normally find in piston engines. I think I have a good reason to go that way with a pulsejet. Would you consider a design of such a pulsejet and evaluate it for practicality of manufacture?

[Tom]

nano technonlogy. Last time i heard, they were working on nano robotics. Little robots that would go round in your viens and capillaries, clearing blockages, etc...

Tom

[milisavljevic]

DELETED.