SOT: Acoustic wave propogation/reflection?

Here's a (somewhat off topic) question for Brett Buck, based on one of his posts in the "Final Solution 20FP" thread.

The Dynajet, as I understand it, can be thought of as a simple acoustic pump, running through this cycle:

1. Explosion happens, valves slammed shut, high pressure wave has nowhere else to go, travels down the tailpipe.

2. High pressure wave exits pipe, triggering low pressure wave returning up the pipe.

3. Low pressure wave expands backward into combustion chamber, further lowering pressure (presumably) and sucking open reed valves.

4. Ambient pressure flows in through the opening, sucking atomized fuel with it into hot combustion chamber until it's ignited by residual heat--at which point we go to step 1 again.

Pretty much the perfect mechanism for turning white gas into noise if you ask me.

Couple of questions:

Can you elaborate on this--assuming I have it right? If I don't, help me out.

What is the phenomenon that causes the negative pressure wave to reverb up the pipe? I'm given to understand that an open pipe end will reflect a negative wave and a closed pipe end will reflect a positive wave. The latter is easy to believe, but what makes the former work?

TIA.

CyberJet

Cat Whisperer,

I believe you have missed some very basic facts. The long tube on a typical CL Jet Speed model is not a "pipe," it is instead a muffler, one which is heavy and not needed for proper performance levels to be achieved.

Why the Jet guys continue to use these things is a total mystery to me, epsecially as they don't seem all that effective as a muffler.

Continue to chuckle whenever seeing you sign off as "The Cat Whisperer." Very, very funny, including the fact that part of the joke is on me.

Funny you should mention pulse jets. There is currently a debate raging around these parts about allowing RC pulse jets to operate in remote areas. This gentleman has alot to do with it: http://aardvark.co.nz/pjet/ and his website may be helpful to you.

On the reflection issue, its easy to imagine the closed pipe reflection as if the wave is simply bouncing back the same way it hit. The reason the open pipe reflection is the opposite phase is a simple case of maintaining equilibrium.

If the fundamental wave continues travelling out the pipe with the same phase, the reflected wave (reflected from the "node" formed by the cessation of a physical boundary) travelling in the opposite direction must be of opposite phase to maintain a force and energy eqilibrium.

If it weren't the opposite phase, there would be an unbalanced force set up and the jet would spin off into space (as my teacher would probably say ). More likely the fundamental and reflected waves would cancel each other out at the node.

Put simply, the reflection has a change of direction, so it also has a change of phase.

A.

Oh, man!!

In light of a rather heated discussion we have been having, you guys have just got to see what the true crazies are doing.

A low-cost cruise missile? Built in a garage!!??

I only used to think a couple of my neighbors were nuts.

Thanks for the tip!

>Oh, man!!

>
>I only used to think a couple of my neighbors were nuts.
>
>Thanks for the tip!
>
>Dan
Dan,
I am sure that most of your ( and my) neighbors and co-workers
have come to this conclusion long ago. Grown men playing with toys!
Pat MacKenzie
P.S. has anyone heard the Tom Waits song "What's he building in there?"
http://www.lyricsdomain.com/20/tom_waits/whats_he_buildingq.html

>Cat Whisperer,
>
>Continue to chuckle whenever seeing you sign off as "The Cat
>Whisperer." Very, very funny, including the fact that part
>of the joke is on me.

Funnier still, if you choose to think of it that way, that I'm looking for Slightly Off Topic info when I spent so much time trying to keep the discussion generally revolving around C/L Stunt.

If you can't beat them...never mind.

Cy,

It's not a question of "waves", it's just pressure.

When the fuel ignites it expands, the pressure inside the dynajet will go up, pressing the valve shut.

When the gases exit the pipe pressure will drop,along wuth the temperature.

The reed problay has some spring action, and will open.

simple as that.

Thanks, Ion.

But no, it's not. (that simple)

For one thing, the valves tend to spring shut, not open.

I'm hoping the acoustic wave propogation guru will reply to the original post.

Cy,.

a valve can spring either shut or open...

OK, Cy,

I'm gonna do it your way, googlin' it.

Have fun!

http://www.pulse-jets.com/valveless/.

I also would like to hear Brett's explaination. But here is a basic tidbit. Acoustic, light, EM(radio) waves all behave the same--just the wavelengths are different. Look into transmission line theory for the explainations of the closed end/open end waveguide (pipe).

Basic TL theory describes the end condition as a step change in line or guide impedance. at any such step, reflected energy is created.

shorted transmission line "stubs" are analogous to the closed end waveguides or sealed pipes. Open waveguides are open pipes.

I recall that your understanding of the reversed phase reflection is correct. At some critical frequency, based upon the physical legnth of the guide and the frequency/wavelengtyh of the wave, you get standing waves in the guide as well. THe trick is to get a standing wave portion of neg pressure where we want it.

You see the end results of all this theory more than you think. antenna "horns" on dish systems are open ended impedance matching devices. When you tune the CB or Ham antenna for minimum SWR, (Tanding wave ratio) you are matching line impedances for optimum transfer across the step and reducing reflected energy. And audio speaker enclosure design is all about closed transmission lines and standing waves internal to the cabinet.

Like I said, hopefully Brett will step in here and correct this in someway we can all understand.

It really is all FM.

FWIW

Check out this pulse jet

http://www.interestingprojects.com/cruisemissile/images/lhmorepower.mpg

>What is the phenomenon that causes the negative pressure wave to reverb up the pipe? I'm
>given to understand that an open pipe end will reflect a negative wave and a closed pipe\
>end will reflect a positive wave. The latter is easy to believe, but what makes the former
>work?

There have been some interesting technical explanations of the negative wave reflection from an open end, but I'm guessing that you are looking for some kind of physical explanation or model. I don't know whether this will do it, but I think of it this way:

A positive pulse is a higher pressure, higher density disturbance traveling at sonic speed through the exhaust gases. It is a pressure wave caused by particles striking particles, which in turn strike others, in a succession progressively moving the disturbance downstream at a sonic speed dependent also on temperature, among other things. When this pulse or disturbance reaches the open end of a pipe, its particles are freed to expand into the lower pressure atmosphere, creating a sudden drop in pressure at the pulse location. This in turn causes following particles to be accelerated (pushed via transfer of momentum from collisions from behind) into the suddenly lower pressure area, thus lowering the density/pressure in their own wakes and creating a new disturbance that moves upstream, as successive particles are pushed from behind into newly created low pressure areas (See Edit below). So this disturbance, a negative wave made up of rarefactions rather than compressions, moves upstream at sonic speed. Actual speeds also depend on the speed of the exhaust gases at any location along the pipe.

I hope that makes sense...?

Edit: While away, it occurred to me that I may have the "cart before the horse" here. How about this? Pressure occurs through the impacts from vibration and superimposed random motions of molecules. So the molecules immediately behind those in the released pulse at the open end encounter statistically less resistance to forward motions and spring further ahead than usual, leaving a rarefaction to be filled, in the same way, by those that follow. This disturbance travels upstream at sonic velocity via the same mechanism. Hmmmm...

Yeah, Prof, that helps a lot. I got to thinking about it in my simplistic terms and one model/explaination I came up with is that the 'slug' of high pressure going out the back is free to expand when unconfined by the tailpipe. In the process of exiting, it sucks out the air behind it. The pressure wave (slug) bursts out, creating a pocket of low pressure which then travels back up the pipe. This arrives at the back of the valves pulling them open.

I understood the theory--in theory--but was left wondering 'what is really going on'.

Now I have a better mental model which 'feels' more intuitively correct than just the dry explainations I've read thus far.

Would still love to hear Brett discuss the phenomenon, by private email if he does not want to get into this on the board. His credentials seem extremely appropriate.

Thanks, Serge

While sanding - happens too often! - I realized a problem with my answer above. I just came back to alter it slightly, but perhaps fundamentally. I'll do that now. See edit above.

>
>>What is the phenomenon that causes the negative pressure wave to reverb up the pipe? I'm
>>given to understand that an open pipe end will reflect a negative wave and a closed pipe\
>>end will reflect a positive wave. The latter is easy to believe, but what makes the former
>>work?
>
>There have been some interesting technical explanations of
>the negative wave reflection from an open end, but I'm
>guessing that you are looking for some kind of physical
>explanation or model. I don't know whether this will do it,
>but I think of it this way:
>
>A positive pulse is a higher pressure, higher density
>disturbance traveling at sonic speed through the exhaust
>gases. It is a pressure wave caused by particles striking
>particles, which in turn strike others, in a succession
>progressively moving the disturbance downstream at a sonic
>speed dependent also on temperature, among other things.
>When this pulse or disturbance reaches the open end of a
>pipe, its particles are freed to expand into the lower
>pressure atmosphere, creating a sudden drop in pressure at
>the pulse location. This in turn causes following particles
>to be accelerated (pushed via transfer of momentum from
>collisions from behind) into the suddenly lower pressure
>area, thus lowering the density/pressure in their own wakes
>and creating a new disturbance that moves upstream, as
>successive particles are pushed from behind into newly
>created low pressure areas. So this disturbance, a negative
>wave made up of rarefactions rather than compressions, moves
>upstream at sonic speed. Actual speeds also depend on the
>speed of the exhaust gases at any location along the pipe.
>
>I hope that makes sense...?
>

Certainly does. The one item about it that you may have omitted is that the *time* it takes for these waves to go from one end to the other is crucial. The time is the speed of sound divided by the length. So right as the wave bursts out of the exhaust, this low pressure area propogates back up the pipe and the "suck" part of the cycle takes this time to actuate the valves and suck in fuel/air through the venturi.

I want to emphasize the fact we are talking about pressure waves in the exhaust, not the exhaust itself. A particular molecule doesn't go from the head end to the exhaust end in a 220th of a second, any more than any particular exhaust particle on a stunt plane goes from the exhaust port to the reflecting surface and back in 1/180 seconds. The guy at the front pushes on his buddy, who, pushes on his buddy, and so forth, and the pushes propogate. The particles also flow out, too, and of course the average pressure in a Dynajet is much higher than atmospheric and the thrust *is* made by particles flowing out the exhaust pushed out by this average pressure. But waves rattle around inside this stream.

It's like half a tuned pipe. In closed systems, we are shooting for the high pressure wave that starts when the exhaust port opens to travel to the end of the pipe and bounce back about the time the port closes. The Dynajet wants it to travel to the end and then disappear into the air.

> Certainly does. The one item about it that you may have
>omitted is that the *time* it takes for these waves to go
>from one end to the other is crucial. The time is the speed
>of sound divided by the length. So right as the wave bursts
>out of the exhaust, this low pressure area propogates back
>up the pipe and the "suck" part of the cycle takes this time
>to actuate the valves and suck in fuel/air through the
>venturi.
>
> I want to emphasize the fact we are talking about
>pressure waves in the exhaust, not the exhaust itself. A
>particular molecule doesn't go from the head end to the
>exhaust end in a 220th of a second, any more than any
>particular exhaust particle on a stunt plane goes from the
>exhaust port to the reflecting surface and back in 1/180
>seconds. The guy at the front pushes on his buddy, who,
>pushes on his buddy, and so forth, and the pushes propogate.
>The particles also flow out, too, and of course the average
>pressure in a Dynajet is much higher than atmospheric and
>the thrust *is* made by particles flowing out the exhaust
>pushed out by this average pressure. But waves rattle around
>inside this stream.
>
> It's like half a tuned pipe. In closed systems, we are
>shooting for the high pressure wave that starts when the
>exhaust port opens to travel to the end of the pipe and
>bounce back about the time the port closes. The Dynajet
>wants it to travel to the end and then disappear into the
>air.
>
> Brett

Well, That makes sense, and in pressure/time terms I might even
begin to understand how a Valveless pulsejet works...such as the
Lockwood-Hiller etc...
Tony

>> Certainly does. The one item about it that you may have
>>omitted is that the *time* it takes for these waves to go
>>from one end to the other is crucial. The time is the speed
>>of sound divided by the length. So right as the wave bursts
>>out of the exhaust, this low pressure area propogates back
>>up the pipe and the "suck" part of the cycle takes this time
>>to actuate the valves and suck in fuel/air through the
>>venturi.
>>
>> I want to emphasize the fact we are talking about
>>pressure waves in the exhaust, not the exhaust itself. A
>>particular molecule doesn't go from the head end to the
>>exhaust end in a 220th of a second, any more than any
>>particular exhaust particle on a stunt plane goes from the
>>exhaust port to the reflecting surface and back in 1/180
>>seconds. The guy at the front pushes on his buddy, who,
>>pushes on his buddy, and so forth, and the pushes propogate.
>>The particles also flow out, too, and of course the average
>>pressure in a Dynajet is much higher than atmospheric and
>>the thrust *is* made by particles flowing out the exhaust
>>pushed out by this average pressure. But waves rattle around
>>inside this stream.
>>
>> It's like half a tuned pipe. In closed systems, we are
>>shooting for the high pressure wave that starts when the
>>exhaust port opens to travel to the end of the pipe and
>>bounce back about the time the port closes. The Dynajet
>>wants it to travel to the end and then disappear into the
>>air.
>>
>> Brett
>
>Well, That makes sense, and in pressure/time terms I might
>even
>begin to understand how a Valveless pulsejet works...such as
>the
>Lockwood-Hiller etc...

Same thing but in both directions!

This site has as good a description as I have seen:

http://www.aardvark.co.nz/pjet/valveless.htm

Note the picture - and note that it looks red hot even in full light!

For a very similar (and exceptionally simple) version, with an additional pahse change, look up "pop-pop" boats. There's hardly any way to make a simpler jet engine - it's an external-combustion pulse-jet!

> I want to emphasize the fact we are talking about
>pressure waves in the exhaust, not the exhaust itself. A
>particular molecule doesn't go from the head end to the
>exhaust end in a 220th of a second, any more than any
>particular exhaust particle on a stunt plane goes from the
>exhaust port to the reflecting surface and back in 1/180
>seconds. The guy at the front pushes on his buddy, who,
>pushes on his buddy, and so forth, and the pushes propogate.
>The particles also flow out, too, and of course the average
>pressure in a Dynajet is much higher than atmospheric and
>the thrust *is* made by particles flowing out the exhaust
>pushed out by this average pressure. But waves rattle around
>inside this stream.

Good to think of it that way. There's a general flow of exhaust gas coming out of the system as a flow of fuel is turned into a series of (what? I don't know that I want to say explosions) burns. Within that flow there are pressure waves rippling up and down.

EDIT:

Just out of curiousity, has anyone got some numbers on the speed of the exhaust gas and the pressure waves? I understand a dynajet runs at 240 pulses per second. If a pulse is triggered every time a returning wave opens the valves, that must be the frequency of the pressure waves. How does this velocity (of the waves) relate to the velocity of the gases heading down the tube?

>
> It's like half a tuned pipe. In closed systems, we are
>shooting for the high pressure wave that starts when the
>exhaust port opens to travel to the end of the pipe and
>bounce back about the time the port closes. The Dynajet
>wants it to travel to the end and then disappear into the
>air.

I was with you until the very last sentence.

You mean that the returning wave disappears as the valves open and the lower pressure is relieved by the incoming charge?

Thanks for the reply.

>Just out of curiousity, has anyone got some numbers on the
>speed of the exhaust gas and the pressure waves? I
>understand a dynajet runs at 240 pulses per second. If a
>pulse is triggered every time a returning wave opens the
>valves, that must be the frequency of the pressure waves.
>How does this velocity (of the waves) relate to the velocity
>of the gases heading down the tube?

Well, the pipe is about 24", and it goes 1/4 of a wavelength in one pulse. So I would say 220 pulses/sec*2 feet*4, or about 1800 fps? It's really hot, so the speed of sound is higher than at STP, of course, and higher than that in a stunt tuned pipe (usually seems to be about 1350-1400 fps), so its probably in the ballpark.

The tailpipe outlet is about 1.25", and thus has an area of about 1.22 sq in. It puts out about 4.5 lb of thrust, so that means the average pressure at the outlet is something like 3.6 PSI. I think that if you care to figure the exhaust density out using the temperature that corresponds to 1800 fps (maybe 8-900 degrees F), you have enough to figure the exhaust velocity. It's should be well under 1800 fps, to check your work.

You could double-check your work by figuring the fuel flow rate, and then assuming stoichiometric mixture, the air flow rate. From this you could figure the Isp and thus the exhaust velocity.

I don't have one right here to measure, so this is just from memory.

>>
>> It's like half a tuned pipe. In closed systems, we are
>>shooting for the high pressure wave that starts when the
>>exhaust port opens to travel to the end of the pipe and
>>bounce back about the time the port closes. The Dynajet
>>wants it to travel to the end and then disappear into the
>>air.
>
>I was with you until the very last sentence.
>
>You mean that the returning wave disappears as the valves
>open and the lower pressure is relieved by the incoming
>charge

No, the fact that the high pressure wave escapes into the external world creates a low pressure in the pipe (and incidentally, creates the thrust pulses). You don't want the high pressure to reflect and return, you want low pressure at the head end to suck in fuel/air. The wave going out the pipe pulls the low pressure you need.

Get a Slinky, and hold it by one end hanging down. Then blip you hand down sharply, and see what happens. This is analogous to a DynaJet.

On a piston motor tuned pipe, you want the wave to start when the exhaust port cracks open, travel to the back, then reflect back with the pressure wave arriving about then the port closes. One wavelength = total distance out and back during the period of exhaust duration. 11000 rpm = 183 Hz, or .005454 seconds. For an exhaust duration of 145 degrees, 145/360=.4, so you want the wave to take about .005454 * .4 seconds, or .00219 seconds. 1400 fps * .00219 seconds = ~3.1 feet, divide by two to get the pipe length, or 18.45" Randy's chart says 17 7/8", but you probably want the pipe to be tuned for less than the operating RPM, so it's on the downslope. Plus these are all approximations, particularly the speed of sound in the exhaust.

This situation is analogous to a Slinky held at BOTH ends.

This is the ideal situation, whith maximum effect - i.e. 1 full wave (albeit with a lot of damping and sub-optimally effective due to a "smeared" reflection off the baffles and tail cone- but much less touchy than it could be). The 20FP muffler allows 4 full bounces in one port opening, which is much less effective than if it were full length, since it loses something on each bounce.

> The wave going out the pipe pulls the low pressure you need.

And that brings us back to the original question.

What is the explaination of the phenomenon that reflects the low pressure wave.

It's not the intertia of the mass of combusted materials, since they are the medium through which the pressure waves are traveling. Which kind of negates my concept of a 'slug' of positive pressure exiting the tailpipe and leaving behind a partial vacuum that travels back up the pipe.

I'm beginning to get that what gets set up is a standing wave pattern.

All of this decided by the length of the tail pipe (and the shape of the neck of the combustion chamber) to generate a negative pressure that pulls the valves open against their tendency to spring closed.

Except if the wave was a standing wave, the valves would not flutter open and closed and the engine would not run.

So a set of waves have to be traveling up and down the pipe.

The more I think I know, the less I think I understand.

>> The wave going out the pipe pulls the low pressure you need.
>
>And that brings us back to the original question.
>
>What is the explaination of the phenomenon that reflects the
>low pressure wave.
>pipe.
>
>The more I think I know, the less I think I understand.
>
>Cy.

Cy....I don't know what, I think I know when though...the negative
wave definitely starts(timewise), when the exhaust passes the end of
the tailpipe and starts to expand...must be generating the negative
pulse then. I "tuned" a Production roadrace 250 Suzuki this way.
Tony

So its the release of pressure out the open end that extracts some air molecules with it and instigates a negative pressure wave that begins to propogate in the other direction?

As the positive pressure wave expands outward--and out of the confines of the exhaust tube--it sucks out a little volume, which then becomes a negative pressure wave going in the opposite direction.

Cy and all...

No one, so far has mentioned ram air pressure against the upstream face of the reed array...

This should have some effect adding to pressure drop aft of the valve. As for flow out the back end, the expanded (burnt) gases do flow through the pipe and out the rear. The rate this happens is tied to the amount of fuel burned and the mass of air it was burned in, in the chamber aft of the reed bank.

The cycling frequencies of the valves works with the pressure waves superimposed on the column of burn residues as they travel back -- sort like AM radio in reverse. (There, the imposed frequencies are within our audible range -- MUCH lower than the carrier, or nominal TX/RX frequencies.)

And, Serge or Brett (or both) said it: the phase of a pressure wave reverses -- as if reflected off a physical obstruction -- at a significant change of section, like going out the end of the tailpipe. It seems odd, but it does that.

Enjoyable thread!

>Cy and all...
>
>No one, so far has mentioned ram air pressure against the
>upstream face of the reed array...

Not by accident.

The really cool thing about a dynajet is that it runs (once you get it started) while sitting still. Certainly ram air in the intake helps out once the thing is airborne.

But whatever phenomenon is at work operates independent of that and the thing will sit and run strapped to a workbench--until the tailpipe melts.

It does take a burst of air in the intake to get the initial charge in the combustion chamber, but after it lights off the thing will just run on the basis of whatever is going on inside the tube, gasping in little spurts of air in the front and blasting bigger ones out the back.

Which I find fascinating the more I think about it.