Page 1 of 1

Real VPJ stats

Posted: Fri May 22, 2009 4:58 am
by serverlan
Hi all,
I'm not a builder, but dable in the theory.
As such, Uflow1d has been invaluable,
I am getting to the stage where I think I can ask it meaningful quastions and get vaguely meaningful results.

What I would like, is the dimesions of a real pulsejet someone has buld, and its resonant frequency.
Something "linear" like a lockwood would be best.
Could someone provide these please?

The reason:
Bruno (where is he) and I used to debate kadency vs. acoustic resonance.
I can now see the two separate effects in uflow1d, and they seem to have considerably different timescales.
Thus, If I put a real vpj into the simulation, and know its resonant frequency, I should be ale to tell what is really happening.

I have done this with some guesswork,
e.g. a 1 meter long lockwood resonates at about 200 hz, etc.
However, the model seems to indicate that this is way off for kadency, assuming a "full" combustion chamber each time.
The implication is that in reality we get limited recharge of the chamber, and.or limited mixing.
I think that I can work back from real data to provide a figure of % effectiveness of the engine,
where a full recharge of the CC would be 100%.
My guess is that current engines don't breathe well, and the same power (or just fuel consumption, more likely)
could be generated by physically smaller engines if we could improve this proposed cyclic CC recharege efficiency.

Its all a longshot, and highly specularive for now, but a real set of statistics (or several for different sizes/designs)
would be a good starting place.

Pretty please?


Re: Real VPJ stats

Posted: Fri May 22, 2009 5:06 am
by Mike Everman
I haven't updated it in more than a year, but this should get you started. There's lots of plans floating around, so if I'm first to respond, do I get to pick the plan? ha ha

Re: Real VPJ stats

Posted: Sat Jun 13, 2009 12:20 pm
by serverlan
Just to wrap this up,

the engines are definitely running at kadency frequencies, not resonance frequencies
The simulations show what is probably a 10:1 difference betweent he two.
The resonance occurs at much higher frequencies.

The timing of the "real" jets seems to be determined by the ratio of combustion chamber size to tailpipe size.
i.e how long does it take a given tailpipe to vent a given chamber.
I'd expect that the predictability of the frequency comes about because most builders are using similar plans and calculators that give good workable engines, and that all give similar relationships between CC and tailpipe.

I tried some % efficiency estimates.
Its a bit circular - start with an combustion, claculate the breathing efficiency, plug that back in for the next cycle and calculate, etc.
Unfortunately the results don't converge. The figures were bout 70%, 35%, 28%... steadily decreasing.
I'm not sure where this leaves me. Obviously the simulations aren't good enough.
I haven't tried fiddling with initial flow conditions yet. that may make the difference.


Re: Real VPJ stats

Posted: Sat Jun 13, 2009 2:58 pm
by Eric
Try using the breathing efficiency to calculate a fuel air charge volume, and expulsion, rather than trying to use the efficiency rating directly maybe. If the numbers are far off from what you know a real engine does then you know its probably something wrong with the efficiency assumption.

Personally I wouldn't start with breathing efficiency, since its kind of a contrived construct. The engines take in and expel however much air they do, the efficiency of any given design may have to be 100% or more depending on your calculations to get it to 'sustain', and probably wont directly compare with other designs. Also the amount of superheated gas that remains in the engine after a given cycle doesn't necessarily have any direct correlation to static efficiency, and certainly goes out the window when moving.

Re: Real VPJ stats

Posted: Sun Jun 14, 2009 1:52 am
by serverlan
Hi Eric,

I actually start with a notional full fuel/air charge in uflow,
run the simulation for one cycle, and dump its data to calculate the next fuel/air charge volume.
This gives me a percentage for the charge for the next cycle.
I put that in, runce the sim, export the data and calculate the corresponding intake charge for the next, and so on.
Unfortunately, they just keep reducing. i was expecting it to reach a limit, which would be the real operating input charge for the engine.
This wwould also give me the real exhaust, CC temp/pressure, etc.

My intake efficiency is a percentage of full CC capacity, rather than the percentage of intake over output.

I'll keep it in the back of my head. There's bound to be something I can change in the model to get closer to a good model.
e.g. intake/output velocities.

Thanks for the feedback