Proposed Design - The Short Lady Revealed

[larry cottrill]

Here is the design of the "Short Lady" that I ran UFLOW1D curves for on a thread I started a couple of weeks ago. This is a logical development of the 'focused wave' design I originally posted along with the proposal for the Elektra I. This is much better than the original design -- UFLOW1D showed that lengthening the conical chamber greatly improves the reduction of standing waves immediately after the combustion event.

The idea is that everything about the chamber [except for the intrusion of the intake pipe] focuses the pressure wave rearward into the tailpipe for maximum effective conversion of energy into tailpipe mass "piston" momentum. The slightly domed front plate and purely conical chamber wall create the focusing effect without significant standing wave reinforcement. See the older thread for the UFLOW1D pressure curves:
http://www.pulse-jets.com/phpbb2/viewtopic.php?t=939

See what you think. I have included the Pipe and Data pages of the supposedly "best" UFLOW1D run, where I added the small conical tail flare shown in the drawings. The 'Output Points' on the Data page represent the L/8, 2L/7, 2L/5 and 2L/3 stations along the engine length. The flat-sided version should yield internal cross-section areas very close to the conical model, though the fit will not be perfect, of course.

Any of this is open to criticism, of course, but if I've made reasonable assumptions about temperatures, the location of the explosion event and so on, it looks like a heck of an engine to me ...

L Cottrill

[steve]

Looks good larry, you want me to build it, or are you going to try it first?

If you would like (and if some of our "sort of" resident experts think it will work) I can have it built in a week or so. I am down in the shop at school for at least an hour per day (today I was there for 2.5!) and am running out of things to build.

It looks very promising and I would be thrilled to be able to try out a new design!

[Mike Everman]

Ha ha, "running out of things to build" ! That's rich, Steve. You should do the bigger one, IMHO. ;-)

[larry cottrill]

Looks good larry, you want me to build it, or are you going to try it first?

If you would like (and if some of our "sort of" resident experts think it will work) I can have it built in a week or so. I am down in the shop at school for at least an hour per day (today I was there for 2.5!) and am running out of things to build.

It looks very promising and I would be thrilled to be able to try out a new design!

Steve -

I have no problem with you giving it a shot, as long as you keep us posted with everything that happens. I would suggest you make the intake unpinched for propane for initial testing, and then try adapting it for liquid fuel later -- but, use your own judgment.

Would you try the conical chamber or the flat one? I have no idea which would actually work better in practice. It will be hard to find equipment to roll a cone that small.

Steve, this is one that you need thrust measurement for [assuming, of course, that it runs at all!] -- so there's another craft project you can put together in your copious free time ;-)

Best of luck!

L Cottrill

[larry cottrill]

Steve et al -

There are a couple of details that I think need to be given attention for this machine to work well.

First, the transition from the chamber cone to the tailpipe should be as nearly perfectly smooth as you can make it. So -- don't weld the front plate dome onto the cone first. Instead, weld the cone to the tailpipe; that way, you can reach in there with a big round or half-round file and smooth out any roughness that happens to be there from your welding, and you should definitely plan on spending a little time to do so. Also, try to make sure that you have a perfect 'butt weld' fit between the narrow end of the cone and the pipe. If the end of the cone is slightly elliptical or has a small 'flat spot' where it comes together, do whatever it takes to make the front end of the pipe match that for a perfect fit and get it perfectly aligned with the out-of-round zone before you weld them up. Use tack welds at about 8 points around the circumference, then finish weld, then smooth up the inside.

Our second concern is that this engine should be extremely good at fully purging the combustion mass, possibly leaving no residuals to ignite the next air/fuel charge! So, you should provide three or four small 'flameholders' at the velocity node, i.e. the inside wall of the front dome. A simple way to make one would be to use a small flat washer with an underlying blob of weld on the dome wall to prop it up at one edge. Then, weld the edge at that point to the blob and the opposite point on the edge to the dome wall. It should look like a sort of 'raised manhole cover' that some combustion gas can hide behind between cycles. Exact orientation should be non-critical; location should be out near the edge of the dome, at three or four equally spaced points. I think such flameholders might be crucial for getting sustained operation from this design. We need to make sure that the explosion phase of each cycle starts right up against the domed front wall, and these will help with that, too.

L Cottrill

[Mike Everman]

Our second concern is that this engine should be extremely good at fully purging the combustion mass, possibly leaving no residuals to ignite the next air/fuel charge!

Respectfully, I think not. To my eye, you'll still have plenty of free radicals in the CC boundary layer. I think you're over analyzing it. Have a cup of coffee, look out the window, watch the pretty girls go by. Life is good. ;->

[Bruno Ogorelec]

I love it, Larry. I have good reason to believe you are onto a good thing with this. I am not at all surprised that Uflow shows it to be good, too.

But (there's always a 'but' somewhere), why the flat wall one? What conceivable purpose would it serve? It will only be structurally inferior, while performance will certainly not be better.

You say that making such a small cone might be difficult to do. In my opinion (for what it’s worth), this is a typical case in which one should go for a cylinder (large diameter tube) converted to a truncated cone by means of making triangular slice cutouts to one side. Bruce Simpson proposes it on his website and I see no reason to avoid this relatively simple trick.

A relatively easy alternative is making a multi-sided (say, six-sided) truncated pyramid. It can be folded from flat sheet or built up from individual trapeze-shaped sides. No need to mention that mounting the intake tube on a flat side will probably be somewhat easier than doing it on the web of a cone.

In fact, I would make the intake tube square-sectioned, too. Forgo the intrusion of the intake pipe into the chamber and have it begin at the side wall in the same way it does in the Chinese. Another huge step towards easy fabrication. Bends are infinitely easier to do with square sections than with circular and oval ones. That’s why SNECMA built their Ecrevisses with square section bends for a long time.

I am absolutely certain that the tube section will have little (if any) influence on performance on this level. Moreover, a square section tube with a circular bend will probably perform better than a round tube with an angular ‘bend’ like the one featured on both the Chinese and your Short Lady.

Next, don’t forget the ‘bustle’ tailpipe. It increases the thrust greatly. Look at the Escopette for inspiration. You might wish to start with a cone and end with a circular-section tube. You are better equipped to work out the nodes at which the transitions should be located than I am.

All of the above said, I again encourage you and anyone else to build and develop this design. I am sure it will be a good one. I know of very good results achieved with a similar approach. As encouragement, let me show you a neat MEW jet, model 307, that seems to have been inspired by similar thinking, but should have had the reflector added. If they only thought of Bruce’s ‘double dome’ valve-shield-cum-fuel-vaporizer….

[larry cottrill]

Respectfully, I think not. To my eye, you'll still have plenty of free radicals in the CC boundary layer. I think you're over analyzing it.

Mike -

Well, maybe you're right -- there's nothing much to disturb that boundary layer throughout the cycle. And if all those radicals are in charge, which they must be, we can't control the locus of the explosion precisely, anyway.

An interesting and refreshing boundary layer experiment is this: Pour a tablespoon or two of Hershey's syrup into a glass tumbler, then fill with cold milk. Using a long spoon, note the effort required to eliminate the chocolate entirely from the inside surface of the glass. When finished, drink the mixture.

Alas, my intuitive grabs have seldom been dignified by anyone employing the term 'analyzing' ...

So, Mike, in broad terms, what do you think of this beast?

L Cottrill

[mk]

Building the cone (If s.o. didn't read the subject):

I don't think it is a problem to build the cone, being similar in size to the exhaust cones of my homebuild tuned pipe for my 80ccm Yamaha.

The easiest way getting the roll-off is using Eric's pulsejet calculator, the augmenter and cone creator sheet, of course. Use ODs in this case and add a 5mm wide "stripe" to the roll-off if you'd like to get an overlap for welding.

Then "pre-roll" the cone by holding it in the vise. Move from the edges to wards the symmetrical line of the roll-off sheet metal.

Get the final shape by hammering the "pre-bent" roll-off around a model pipe, its OD should be a bit smaller than the final small cone ID.

Now tack-weld the cone by holding it in a vise, screwed pliers or whatever you prefer.

Give the cone openings their more or less completely round shape and their exact diameters by using a bigger and a smaller model pipe or an anvil.

Weld up the cone completely as the last step.

[larry cottrill]

I love it, Larry. I have good reason to believe you are onto a good thing with this. I am not at all surprised that Uflow shows it to be good, too.

Thank you ... I thought you'd like it. I think this is going to be a good one.

But (there's always a 'but' somewhere), why the flat wall one? What conceivable purpose would it serve? It will only be structurally inferior, while performance will certainly not be better.

Just an alternative that I thought might be a bit more buildable. Not necessarily better, though it would have a more 'channeled' flow to the front of the chamber, like the Elektras.

You say that making such a small cone might be difficult to do. In my opinion (for what it’s worth), this is a typical case in which one should go for a cylinder (large diameter tube) converted to a truncated cone by means of making triangular slice cutouts to one side. Bruce Simpson proposes it on his website and I see no reason to avoid this relatively simple trick.

Yes, I'm sure that would work pretty well, as long as you were sure you could keep the inside pretty smooth, especially as you close in on the nozzle zone where you connect up with the pipe.

A relatively easy alternative is making a multi-sided (say, six-sided) truncated pyramid. It can be folded from flat sheet or built up from individual trapeze-shaped sides. No need to mention that mounting the intake tube on a flat side will probably be somewhat easier than doing it on the web of a cone.

Yes, that would be workable too, of course. To me, welding a tube to the outside of another curved surface is no big deal, but that's because I have no qualms about leaving gaps that need to be fillet welded in. Take a look at the construction photo of joining the tailpipe onto the Elektra II to see what I mean.

In fact, I would make the intake tube square-sectioned, too. Forgo the intrusion of the intake pipe into the chamber and have it begin at the side wall in the same way it does in the Chinese. Another huge step towards easy fabrication. Bends are infinitely easier to do with square sections than with circular and oval ones. That’s why SNECMA built their Ecrevisses with square section bends for a long time.

I am absolutely certain that the tube section will have little (if any) influence on performance on this level. Moreover, a square section tube with a circular bend will probably perform better than a round tube with an angular ‘bend’ like the one featured on both the Chinese and your Short Lady.

Possibly, but I think the important bit is the wave getting out easily, not a minimum impedance path for airflow in. Square tubes have their own problems; modern HVAC engineers don't like to use them as much as round and oval sections, now that those are more easily available due to modern fabrication technology.

Next, don’t forget the ‘bustle’ tailpipe. It increases the thrust greatly. Look at the Escopette for inspiration. You might wish to start with a cone and end with a circular-section tube. You are better equipped to work out the nodes at which the transitions should be located than I am.

Hadn't thought about that, though I had considered tail-end augmentation; I know you mentioned recently that augmentation at the intake might even be more effective. The long diffuser exhaust you're talking about is certainly something to consider as well.

All of the above said, I again encourage you and anyone else to build and develop this design. I am sure it will be a good one. I know of very good results achieved with a similar approach. As encouragement, let me show you a neat MEW jet, model 307, that seems to have been inspired by similar thinking, but should have had the reflector added. If they only thought of Bruce’s ‘double dome’ valve-shield-cum-fuel-vaporizer….

I remember that one being advertised when I was a kid building U-control models. Though the 307 was fairly lightweight, it was bulky and low on thrust. Probably just inadequately developed. I think the name stamped through the side of the chamber from the back side is a particularly effective artistic touch ...

L Cottrill

[Mike Everman]

So, Mike, in broad terms, what do you think of this beast?

In general, I don't know what to think of it. I think that there are insurmountable difficulties believing UFLOW analysis on this and the Chinese. I've been trying on the Chinese to either linearize it, or divide the problem. The latter I'll be trying, but can the result be believed? Who can say? If it gives you a warm fuzzy, I'm all for it. At the very least it will get you to something like the proper length, if the temps are close.

I'm not on board with the stuff you're doing with the flows, snorkels and intake pinches, but the proof is in the putting. I think you'll certainly get it to sustain, but I'd rather see the intake terminate as it enters the CC proper, and I can't substantiate why, so don't ask me to explain it, It's just a "feel thing".

I like to see that you're concentrating on the acoustics a bit more than the flows. I think "flowcentric" and "flow diode" thinking is a trap I fell in for a while that is fun but ultimately self-defeating, IMHO.

Nice work, I can't wait to see it work!

[steve]

I rolled up and welded the tailpipe today, tomarrow I will try to finish the CC cone.

larry, dont worry about the cone, It's nothing compared to the one I had to do for the exaust for MK's K-pt01 That stupid thing took me an hour to roll!

[larry cottrill]

I rolled up and welded the tailpipe today, tomarrow I will try to finish the CC cone.

larry, dont worry about the cone, It's nothing compared to the one I had to do for the exaust for MK's K-pt01 That stupid thing took me an hour to roll!

Steve -

Way to go. Don't forget to take pains with the smoothness of the nozzle zone. Sounds like you're well on your way -- man, I can't wait to see it!

Just one more little thing you can do for us if you would: Since I'm not building one [yet] try to keep a simple record of accurate measurements as you actually get the assemblies put together. Especially note the overall length [inside, if you can get it somehow], the location of the beginning and end of the intake tube and the intake tube centerline length! Incidentally, the length shown for that tube is supposed to be the acoustic length, which means the real piece should be .6 x ID shorter in total physical length overall. Try to hit this one just as close as you can cut it -- it might help to weld it up at the elbow before making the angle cut on the inside end, so you can get it right where you want it. It's a hard thing to measure, especially with an angled cut like that at one end. Be sure and get the total finished weight as well, and measure the point that seems to be the CG with the plug in place. (You do want this to end up as a flight engine, don't you? ;-)

Good luck, and thanks again!

L Cottrill

[steve]

Incidentally, the length shown for that tube is supposed to be the acoustic length, which means the real piece should be .6 x ID shorter in total physical length overall

...wait, what????

so what you are saying is 133 - (.6 x 19) = 121.6mm ?

the intake should be 121.6mm and not 133mm long?

Be sure and get the total finished weight as well, and measure the point that seems to be the CG with the plug in place. (You do want this to end up as a flight engine, don't you? ;-)

oops, well, too late for that! I used heavy sheet metal and crude sloppy welds so this won't exactly be the lightest version possible. If It does work though, I will probably build a seccond high quality version to take thrust and weight measurements on.

[Anthony]

Steve, you're really impressing me. You go head-first into things, and seem to find your way so easily. Good luck man, although I'm sure you won't need luck and that everything will run fine hehe.

With someone as Larry on the drawing board anyway, it's gonna end up running for sure if it doesn't in first place!