New and Improved Lady Anne

[Eric]

This video is from a chinese, it was all a matter of having the fuel injector at a certian spot, with the straight tube injector It took rapidly pulsed air to quickly kick it into pulsejet mode, or simply changing the injector position, though its a little different in design

I was getting some really wild action from that engine that I didnt get on video, like quickly jumping from the high to low howling at about 10 htz with really wild audiable tones.

Eric

[larry cottrill]

Well, I tried uploading a sound file a couple of days ago as a Christmas gift to you all, but found to my astonishment that the forum doesn't like uploading MP3 files. So, yesterday I tried making a WAV file out of it, but that file was HUGE! Apparently, MP3 has way superior compression - no wonder it's such a popular format for music and such. Anyway, using my fabulous MP3Cutter tool, I cut out a tiny slice of the MP3 (made directly from the camcorder soundtrack), used MP3Recorder to make a smaller WAV file, and I hope I can upload that. It isn't much, though - just a couple of seconds worth.

I also ran the short MP3 through the Windows built-in sound player and took a screen snapshot, so you can see what the waveform looks like. Man, that is a harmonic-rich sucker! Pretty far from a sine wave, that's for sure. Of course, a lot of that probably comes from the positive pressure reflection at the choke cone of the main tube. Remember in looking at / listening to these that this is air-driven, not a sustaining run, so they need to be taken with a grain of salt.

L Cottrill

[larry cottrill]

OK, it says 'Extension WAV is not allowed'. What IS allowed? I guess it doesn't matter - MP3 and WAV are my only choices, i.e. the only ones I can do.

Well, I tried ...

L Cottrill

[larry cottrill]

OK - here it is - the MP3 file promised earlier, encapsulated in a ZIP file.

First, you'll hear (just barely) the Model T Ford spark coil, then the compressed air in the starting tube, then some bangs as I try to 'home in' on a throttle valve setting that works, and finally the resonant scream. The interruptions are when I try to see if it will sustain. If you hear a loud humming noise come up, that's just my air compressor coming back up to full head, in the background.

Sorry I wasn't sharp enough to get it out there on Christmas day. Anyway, enjoy!

L Cottrill

[mk]

Man, that is a harmonic-rich sucker!

Unfortunately, it is not. "HF ringing" -- to paraphrase you -- was well surpressed. (!

[Irvine.J]

Ok I ran it through VA and found a couple of things may be helpful, or maybe mean nothing.

1. Your frequency at the most audiable (most gas) I think was 366
2. It also held a 355 for a while
3. Brief periouds at 344 and 322
And only once made 312 before you could save it and bring it back up briefly.
Though it usually died anywhere under 322. This might help for something?

[Irvine.J]

Larry, is this kinda sorta in the ballpark here...
We know that the speed of sound is 340.29mps (At sea level)
And that your engine seemed most prominent around 366hz. (Or 366 CPS)
Therefor λ = Vw / f = 0.929
then half the half of the wavelength is 0.464 meters.
Your engine on the schematic is .546 meters, perhaps its too long?

I really don't know and I'm just going on some info off one of the many PDF's available, (The length of the PJ should be half the resonant frequency.) Though as we discussed its only very preliminary testing you've done so far. Still it might be food for thought?

[larry cottrill]

James -

Interesting observations. However, in terms of calculation, you're not taking into account that the wave speed is MUCH faster in the hot gas in the pipe than it is in normal air. In fact, the speed of sound varies depending on where you are in the pipe! This is why your guesses at internal temps are so important to UFLOW1D and NUDiS, both of which look at local temps to decide how to propagate the pressure within the tube.

Basically, you are trying to match the "cold" open-pipe frequency of the intake duct to some odd harmonic of the hot closed-pipe frequency of the main duct. (An oversimplification, of course.)

L Cottrill

[Irvine.J]

I'm Still learning :D... *places bucket on head and sits in the corner for a while.*

[larry cottrill]

James D, IrvineJ et al -

All right, forget what I said about the straight nozzle - have a look at my Guinevere testing thread and what I learned last night about fueling at that deep location. Basically, it is MUCH better to have some definite spread in your fuel spray pattern at that location. I will make another "lemniscate" fuel pipe and fit it to this engine for further testing. I didn't have such a tube available last night, though. The "lemniscate" end is really easy to produce, and you might try it for a little more testing of your Rev02 build, unless you feel you're already doing something just as good.

A couple of things I did learn last night in trying to run the Lady Anne Rev03: My starting air pipe is very poorly arranged for safety, having its inlet near the rear of the engine - my hand gets almost fried as soon as the pipe gets really hot, and rich extinction flameouts are really exciting, too! Also, its rearward location effectively prevents lengthening the engine with any kind of sliding sleeve. Duh. Anyway, some changes are in order for both the air and fuel tubes before more testing will be meaningful (not to mention safer for the operator!).

I was able to get a long enough night run to see a good heat pattern something like what you (James D) showed in photos of your earlier version engine. However, with the present fuel setup, it is hard to get good running because throttleability is so poor and the fuel valve setting so ultra-sensitive to the slightest change.

L Cottrill

[James D]

Larry, that really sounds much, much better than I was expecting, it can't be far off running.
The way it switches between the two frequencies is very interesting; do you think that means the intake length is not quite right?

I'll have to give mine a go with your lemniscate nozzle type, and see what happens.

Just out of interest, I plugged the intake right back to the combustion chamber and screwed a small microphone into the spark plug hole. I measured a (cold) acoustic length of 595mm for the main duct, this figure is probably not absolutely accurate but I’m confident that it is somewhere between 590 and 600
Anyway, it’s a lot more than 546mm plus end correction, but I'm not sure why, or how useful it is to know that?

[larry cottrill]

The way it switches between the two frequencies is very interesting; do you think that means the intake length is not quite right?

I think it's more like a change in combustion location creates a sudden change in flow somewhere, altering local conditions. I can't imagine how to explore such a phenomenon, really.

I'll have to give mine a go with your lemniscate nozzle type, and see what happens.

Yes - please try it !!!

Just out of interest, I plugged the intake right back to the combustion chamber and screwed a small microphone into the spark plug hole. I measured a (cold) acoustic length of 595mm for the main duct, this figure is probably not absolutely accurate but I’m confident that it is somewhere between 590 and 600 ;) Anyway, it’s a lot more than 546mm plus end correction, but I'm not sure why, or how useful it is to know that?

That was a really cool thing to try! I have no idea why it would be so far off - could there be a temperature correction you didn't make (or didn't make correctly)? An error in temperature can make quite a difference in frequency vs length calculations, because the density of the air is so crucial to the speed of sound. Hard to say how useful - again, the accuracy of the dimensions doesn't mean much if you're just guessing internal temps, which we always are for a running engine.

L Cottrill

[leo]

Just out of interest, I plugged the intake right back to the combustion chamber and screwed a small microphone into the spark plug hole. I measured a (cold) acoustic length of 595mm for the main duct, this figure is probably not absolutely accurate but I’m confident that it is somewhere between 590 and 600
Anyway, it’s a lot more than 546mm plus end correction, but I'm not sure why, or how useful it is to know that?

Cool that you measure that, how do you get resonance, did you blow over the exhaust with something.
I suppose that the frequency is lower than you expected, because a pulsejet is not a strait pipe, and you always have it also work a bit as a Helmholtz resonator too, this effect will always make it resonate slower never faster.
Maybe this cold testing is a good start for later corrections on acoustic temperature, to get a real average internal temperature.

[Eric]

Sounds like it is in rijke tube mode. If it is it wouldnt be producing much thrust at all. If you tip the engine so it is pointing straight up, you might be able to get it to sustain in this mode if it is, since the convection will help pull in air and may give it just enough to sustain itself, though that probably wont help get it into pulsejet mode.

Eric

[James D]

Larry,
I measured the temperature quite accurately, any error probably came from the measurement of frequency but I did it several times and took an average.

Leo,
If you play white noise nearby, the pulsejet (or any piece of tube with a microphone inside) will show it's character, it doesn't really resonate as such but will 'favour' certain harmonics and show peaks on a spectrum analyser.(btw, I think it was Bill Hinote who suggested that, some time ago)

If you want your pulsejet to really come alive and resonate (kind of) you need to feed the sound from the microphone in the combustion chamber back to a large speaker/guitar amplifier near the puslejet. You should then be able to get strong feedback at the frequency of the pulsejet, I've tried this and its quite interesting.