Proposed Design: Simple Two-Tube Valveless

[larry cottrill]

Comes now the Reynstodyne Whale Shark(TM), a new design that gives you a valveless engine body built from just three main pieces: two tubular sections and a washer. The trick here is that you choose a chamber tube whose diameter is as close as possible to 1.5 x the tailpipe tube diameter. The one shown in the drawing uses 1.5-inch (nominal) EMT (steel conduit) for the chamber and either 1.25-inch OD antenna mast tubing or 1-inch (nom.) EMT for the tailpipe. No cone forming required.

What you do is vise flatten the rear portion of the short chamber piece slightly - just enough to make its width identical to the OD of the tailpipe section. Then, cut a big notch out of the bottom rear portion as shown. The only welds needed are super simple: a fillet weld along each side, then a butt weld around the bottom to close the seam between the front of the tailpipe and the rear of the notch, plus welding on the front plate. The upper wall of the tailpipe forms the lower wall of the intake, which becomes a crescent-shaped section, as was proven to work wonderfully in the Fo Mi Chin II design. The flattening of the chamber tube is so slight that the top and bottom will be practically semicircular in section.

Nothing to it. This is probably the easiest construction engine I've designed yet. The mounting fins are totally optional, of course. This design should be scalable slightly smaller or much larger, as long as tubing sections of the proper ratio are available (head for your local muffler shop).

About the only thing critical is that the chamber tube OD should be very close to 1.5 x the tailpipe OD or just a bit smaller, to leave you with about the right intake area. The flares can be hammered out after all the welding is complete.

Comments, gentlemen?

L Cottrill

[Mike Everman]

Larry,
That's no doubt in my mind, your best design yet. I don't see why it won't resonate well. As all straight pipers of this size, though, it will be weak. If you're set on the path to the most out of a straight 1.25" tail to compete with a Red Head, you'll have to settle for 1/4 to 1/3 of the thrust. Only tail expansion will get you more, IMHO.
Still, this is a sweet piece of design for manufacturing!

[Mike Everman]

BTW,
The path length for the intake, acoustically corrected, of 7.3 should be 1/3 of the over-all acoustic length, or 21.9", for a duct length of roughly 21.5.
This 1/3 is important, and IMHO, the 1/8 is not, and well, if I may, just wrong.

My recent studies point to changes in diameter (to include the intake's entry to the CC) should fall on odd factors of the 21.9" length, enforcing the odd modes (pressure nodes) of the duct by making points of changing velocity. So your pinch point needs to be where? I'll have to think about that. L/5 sounds good, though some would say that the velocities are low there, so this location is less important than elsewhere.

Likewise, the L/5 for intake length is a good rule for a starting point; more germain on the intake is the aspect ratio (L/d) of it, though keeping the L/3 for your path length.

Not easy to make adjustable if you follow, but means that for every addition or subtraction from the intake length you want to make, there is an adjustment to the tail to preserve L/3 acoustic ratio. This of course makes the pinch in the wrong place, so it's good that it may not be so important! ;-)

[Eric]

I am waiting for Mark to jump in with some interesting, yet little known facts about whale sharks....

I think im coming down with something and my brain is totally not working, but it looks like it should run... I would make one that just has the tailpipe jammed in the bottom so it can be moved back and forth without requiring building new prototypes.

Eric

[Mark]

I think Larry's jet should have 5 gill slits on either side and some sort of filtered intake.
Mark
http://www.enchantedlearning.com/subjec ... hark.shtml
http://new-brunswick.net/new-brunswick/ ... whale.html

"Generally harmless"
http://www.flmnh.ufl.edu/fish/Gallery/D ... shark.html

[larry cottrill]

Still, this is a sweet piece of design for manufacturing!

Mike -

Thanks! I think this is the best recipe for a beginner's engine KIT that I've ever come up with. The kit would simply have the front plate and plug mount welded on, and the chamber pinched and notched. The starting tube would be modified so that it was pre-welded to the upper side of the chamber pinch - you could verify and adjust exact alignment of the fuel pipe within the intake shell before wiring the tailpipe on for tack welding, so you'd know you had perfect alignment. The finish welding is extremely easy, with no pinch points or other "problem" areas.

A shoo-in for a beginning welding student.

L Cottrill

[larry cottrill]

BTW,
The path length for the intake, acoustically corrected, of 7.3 should be 1/3 of the over-all acoustic length, or 21.9", for a duct length of roughly 21.5.
This 1/3 is important, and IMHO, the 1/8 is not, and well, if I may, just wrong.

I just use it because it has worked OK for me in the past. I agree, though, that it's the location of the intake flare that really counts.

My recent studies point to changes in diameter (to include the intake's entry to the CC) should fall on odd factors of the 21.9" length, enforcing the odd modes (pressure nodes) of the duct by making points of changing velocity. So your pinch point needs to be where? I'll have to think about that. L/5 sounds good, though some would say that the velocities are low there, so this location is less important than elsewhere.

Interesting. I'll try to handle that precisely on some future design, and see what it accomplishes.

Likewise, the L/5 for intake length is a good rule for a starting point; more germain on the intake is the aspect ratio (L/d) of it, though keeping the L/3 for your path length.

This one has a pretty low aspect ratio, obviously, which isn't that good. It might appear to the casual observer that the intake has no definite "bottom end", though of course it does, because of the sharp edge of the upper lip of the tailpipe tube. Still, overall, this intake might be pretty "soft" in terms of resonance.

Not easy to make adjustable if you follow, but means that for every addition or subtraction from the intake length you want to make, there is an adjustment to the tail to preserve L/3 acoustic ratio. This of course makes the pinch in the wrong place, so it's good that it may not be so important! ;-)

Have you ever concluded what the allowable length errors seem to be, in percentage terms? (I realize this may vary from one design to another.) It has always seemed to me that if really tiny adjustments make large differences, it has to be the influence of pretty high-order harmonics, because the tiny length changes would be such a small part of the wavelength of the basic operating frequency (and even of the lower harmonic frequencies). Comments?

I feel that the basic layout of this engine is not unusual, acoustically. What is really different here is going to be how the masses flow. Looking at expansion in the explosion phase, the impedance out through the intake channel will be extremely low (true of the tailpipe also, but there the 'piston mass' is much greater). I have said before that one of the strengths of the Lockwood geometry is how rapidly he can purge the chamber out through the front end. The same should be true here, although of course we don't have that very small choke at the tailpipe. I would not be surprised to see this engine running with 6 or 7 inches of visible flame out behind the intake flare!

It should be really fun to see this one go, even if a bit weak as you predict.

It is impossible to model this one accurately with 1D geometry. A crude attempt with UFLOW looks promising - however, it can't be very close to how the thing will really behave, massflow-wise. Resonance looks good, but could probably be improved with a skinnier intake. It's really hard to know what area to use, since these oddball cross sections have some pretty ineffective area down in the tight corners.

L Cottrill

[Mike Everman]

"Have you ever concluded what the allowable length errors seem to be, in percentage terms? (I realize this may vary from one design to another.) It has always seemed to me that if really tiny adjustments make large differences, it has to be the influence of pretty high-order harmonics, because the tiny length changes would be such a small part of the wavelength of the basic operating frequency (and even of the lower harmonic frequencies). Comments? "

My comments at this point are anecdotal. Length tolerances I'm sure are a pale second to diameters chosen. I think your L/8 works OK because it is so close to L/7 or L/9. M. shouldn't bother to tell me this is purest bullshit. I know already it's blue-sky!

Maybe we should have a little contest between the hard core builders here, for "best" (whatever THAT means) implementation of a straight piper at 22" long. Any takers? I'd love to take a stab, and we'd get more data if they all go on the same test stand. I'm hoping by year end to have real-time SFC for small motors, though for an RC model, "full military" thrust capability might be one category of "best", SFC be damned. Sounds fun to me, anyway, even if I do think they will all be slightly anemic, someone will come up with a new take on it, and shine (by a few ounces) for this eclectic group. We'd see annulars, multi-intakes, snorkelers; what fun would that be? ;-)

[pezman]

If you squash a 2" ID pipe into an ellipse that just touches the walls of a 01.25" OD pipe, then the two spaces on either side of the central pipe are about 0.865 sq. in. -- i.e. just a bit larger in cross section than the central 1" ID pipe (see figure)

If the central pipe is 1.25" ID and the 2# ID pipe is squashed to touch it, then the two "extra" areas around the central pipe have a total area of about 1.29 sq. in, just a tad larger than the area of the central pipe (about 1.227 sq. in).

I wonder if it is feasible to make a jet like this just by squashing a large pipe with carefully chosen dimensions around a smaller pipe with carefully chosen dimensions and skipping the welding altogether.

[pezman]

A twist on this would be to have the central pipe serve the role of the intake, and mount it at a slight angle (i.e. it starts at the center, then breaks through the wall of the outer pipe at some point). That way, you'd get an expanding tail section.

[larry cottrill]

pez -

Sounds great to me - give it a shot!

L Cottrill

[larry cottrill]

Maybe we should have a little contest between the hard core builders here, for "best" (whatever THAT means) implementation of a straight piper at 22" long. Any takers? I'd love to take a stab, and we'd get more data if they all go on the same test stand. I'm hoping by year end to have real-time SFC for small motors, though for an RC model, "full military" thrust capability might be one category of "best", SFC be damned. Sounds fun to me, anyway, even if I do think they will all be slightly anemic, someone will come up with a new take on it, and shine (by a few ounces) for this eclectic group. We'd see annulars, multi-intakes, snorkelers; what fun would that be? ;-)

Ha - that sounds great! Of course, there are a few problems to iron out: Will they be required to have the same tailpipe diameter? How close to the 22 inches do they have to be to qualify (what I mean is, you might need to allow a little tolerance for tuning)? What constitutes a "straight pipe" engine (mine has a bulged end due to the second, larger tube)? And so on.

Just a few details for the Rules Committee to work out before we get started ...

L Cottrill

[larry cottrill]

My comments at this point are anecdotal. Length tolerances I'm sure are a pale second to diameters chosen. I think your L/8 works OK because it is so close to L/7 or L/9.

Mike -

Remember where my L/8 comes from - it was not chosen arbitrarily. Bill's observation was that successful engines have intakes L/5 long that terminate at an intake L/3 from the pressure antinode (the front reflector plate in my designs), taking the acoustic end corrections into account. (Note that we are talking about wave path lengths, not one-dimensional measurement along the axis). Now, L/3 is .33L and L/5 is .20L - subtracting leaves you .13L of space at the front end, which I noticed is perilously close to L/8.

I actually suspect that the reality is that, at least with the basic layout I like to work from, the L/3 location of the intake is far more critical than anything else. I just looked at the original FWE once more with UFLOW1D - of course, it has to be modeled as a "linear FWE" like Steve and Eric built, i.e. the intake is rotated around the pressure antinode to get it out in front, to "one-dimensionalize" it. I looked at this both with the graphic output of the program and with my density plot animation, concentrating on what I could see in the intake tube. Doing this, I have to conclude that the claim that the much-made-of resonance of the intake tube in and of itself as a powerful modifier of the action of the engine is hogwash. There, I've said it. Sure, you can see little ripples moving back and forth, but they aren't big enough to mean much - they appear to just "ride on" the fundamental pressure wave without much real influence. Of course, this flies in the face of what we are supposed to believe. To me, the intake duct appears to be simply a calibrated leak having an appropriate impedance and a nicely sized enclosed mass.

Or, do we believe that the little vibrations in the intake are a kind of "straw that breaks the camel's back"? Or, the butterfly that tips the boulder over the edge? If that's true, how can we prove it?

In my earliest running engines, where the combustion chambers were very poorly formed (Elektra I and II, with flat octagonal boxes), the characteristic frequency of the engine was determined entirely by the intake pipes! These were narrow tubes, i.e. fairly high L/D ratios. Of course, only a certain tailpipe length (extraordinarily long!) would allow the engines to sustain, but messing with these lengths didn't change the frequency of the engine roaring under forced air! In other words, you just had to make the tailpipe agree with the intake/chamber assembly to get resonance. Well-formed chambers like the FWE and Fo Mi Chin, however, responded to different tailpipe lengths by oscillating at different frequencies (as we are taught to expect). Also, the well formed chambers reasonably fit Bill's L/3 criterion at final tuned geometry, while the Elektras were way off (needing the vastly overlong tailpipes to hit resonance)! This was true even of the original Fo Mi Chin with its odd arrangement of delivering intake air into the front end of the tailpipe (in that case, the L/3 criterion was only approximately satisfied, with the intake pipe length FAR short of L/5 !!! See comparison outlines, below - note that none of these silhouettes shows the location of the inside end of the pipe!).

L Cottrill

[hinote]

Maybe we should have a little contest between the hard core builders here, for "best" (whatever THAT means) implementation of a straight piper at 22" long. Any takers? I'd love to take a stab, and we'd get more data if they all go on the same test stand. I'm hoping by year end to have real-time SFC for small motors, though for an RC model, "full military" thrust capability might be one category of "best", SFC be damned. Sounds fun to me, anyway, even if I do think they will all be slightly anemic, someone will come up with a new take on it, and shine (by a few ounces) for this eclectic group. We'd see annulars, multi-intakes, snorkelers; what fun would that be? ;-)

Hey Mike:

Let's start a new, dedicated thread to the "Chinese building contest"--OK?

Please see the attached photo.

As you can see, I've already started on my version of an engine to submit.

I'll let you deduce what's going to happen to the pictured part.

Bill H.
Acoustic Propulsion Concepts

".......some day soon we'll be flying airplanes powered by pulsejets."

[Eric]

22" is a little short. That puts a properly scaled chinese or escopette styled engine at some pretty damn small cross sections.

Or are you talking about ones with no tailpipe expansions?

Bill I was just drawing up an engine just like that on inventor this afternoon that I was going to post. Sort of like a little thermojet.

Eric