Just Another Valveless Front End

[Bruno Ogorelec]

Mike (...) Your last design used a shape something like the letter "C". That puts the entry point into the combustion chamber on the side and fairly far back.

To reiterate the point I have often made -- there is no such thing as "far back" in a a pulsejet. There is no reason for the intake to be in front, the chamber in the middle and the exhaust at the rear.

We are not talking of ramjets, in which air and other gases proceed (generally speaking) from front to back. Flows in pulsejets are DIFFERENT. (Yes, I am shouting.)

According to seceral sources I have seen, the best aspiration you can find is achieved in the Logan pulsejet, which has a side intake at about the middle of the chamber. The NRL pulsejet (one of the more thoroughly tested layouts I know) also has a side intake at about the middle of the chamber, inclined topwards the bottom at about 45 degrees but configured so that the fresh charge hits the opposite chamber wall.

Most of the participants in this forum appear to consider the designers of those two engines fools abd cling blindly to the 'Dynajet' layout -- for no reason that I can fathom.

Bruno

[Bruno Ogorelec]

1.)
I agree, Bruce, I don't like the pre-heating of the incoming charge either

Um... what heating of the charge? How much do you think charge can get heated from a few inches of duct wall? that duct wall, by the way, is far from the hottest part of the engine. In fact, it is among the coolest. I wouldn't worry.

What I think is wrong is going for the 8-pipe layout instead of 4-pipe. We know that a 4-pipe thing works well, we have the workable measurements etc. etc. Why throw away such priceless pointers and have everything about the engine unknown? It's just one big extra reason for the engine not to work.

Bruno

[Viv]

I prefure the 8 tube layout even though Bruno correctly point out that it is introducing a new varible.

Graham wisely councils the reduction of varibles advice that we as a team have learnt the hard way.

But the 8 tube layout fits the design better to my mind, any design should have an elagant simplicity the first design has this, although I was not so happy with the second design i did like the overlap of the tubes in the inlet tract and this feature should be kept.

Build the combustion chamber and exhaust as per the Kentfield design and dimensions.

The four tubes on the Kentfield engine are mounted on a simple circular plate, if you put your front end on to this plate you can test and modify the engine with ease.

It should also be possible to make an 8 tube inlet to test and optimise the dimensions.

You may want to add the augmentor (ejector:-) at a later stage in the development.

Viv

[Mike Everman]

And I was wondering if yesterday's light activity was because I finally exposed my colossal and insurmountable ignorance! I woke up at 5am (I think because I heard Bruno shouting half a world away) after an evening of messing with the original layout, see below. I'll take the comments in turn:

the best aspiration you can find is achieved in the Logan pulsejet, which has a side intake at about the middle of the chamber.

I had seen a reference to this in mining the old forum, and that is one reason I moved to tubes. (BTW: any more info on the NRL?)

... what heating of the charge? How much do you think charge can get heated from a few inches of duct wall?

Finer point of the design, to be sure, but the original configuration has two 90deg turns with one wall being the CC; how much can it heat the charge? Don't know, but this is different from straight pipes and I imagine quite a bit of shear derived turbulence through the turns, allowing more heat transfer than normal. As cool or as cooled as they can get will have benefits, I think, like an intercooler on piston engines. Again, a plus for tubular intakes with some space and cool flow between them and the CC.

The four tubes on the Kentfield engine are mounted on a simple circular plate, if you put your front end on to this plate you can test and modify the engine with ease.

Where is that old Kentfield I had in the attic? LOL Perhaps if we could get Bill Hinote to chime in, I could make a head for his motor. He lives only a two hour drive from here, I believe...

I prefure the 8 tube layout even though Bruno correctly point out that it is introducing a new varible.

Sure is less parts to go with 4, and still interleaved, still the same potential benefits, maybe less augmentor turbulence? The more the better for intake cooling, but as Bruno says, this is probably polishing a turd.

Finally, now that some proportions are taking shape, my current efforts in the design are focusing on what I think I can make easily.
1. My welder is only good for carbon steel, so tha's what I'll experiment with
2. I have a small press, and think I can make either the sheet intakes or form the tube types in halves from sheet, from dies I can also make. Everything else is shear, bend and weld.

The finer points of experimentation are going to be aspiration, fuel flow, ignition among others. I'd be happy to send part sets to anyone that wants to build concurrently with me.
The material cost is nil, it's all labor, so who wants to get a package in the mail? We can do something really cool here.

I can't say the first hardware will be completely designed by committee, but I've grown to value greatly the input from you all, and I also know myself, I will only experiment so far after building the basic setup, even or especially if it works! I am a manufacturer with pretty good resources and I think that we, as a team, could eventually offer an extremely cool kit, and dedicate the proceeds to (our collective) R&D!

Mike

[Viv]

Take a basic Kentfield engine.

1 8 tube layout, 4x4 arrangement straight, circular plate 8 straight tubes arranged two rows of four.

Test and optimise length of tubes.

2 as above but bend the top third of the tubes to opposed format as per your drawing layout.

Adjust to work

3 try adding inlet nacell and augmenter.

4 try it in a pressed steel format

Viv

[Mark]

[quote="Viv"]I prefure the 8 tube layout even though Bruno correctly point out that it is introducing a new varible.

Graham wisely councils the reduction of varibles advice that we as a team have learnt the hard way.

It would be interesting to have a contest to create the most surreal pulsejet. Have a basket starfish worth of tubing and throw a slug of air into a morass of branching pulmanary snorkel elbows or whatever comes to mind. There's no reason why a pulsejet couldn't be made into some likeable kenetic art form. Imagine a tube-train-track and table of switches to make your fire-form travel through one set of ducts, or switching a segment of pipe track and having the combustion choo choo into a tunnel of different tubing and just play with the combustion like you would a toy train with lots of track. The fun would be to keep the entity alive and not have it fall off the track on a tight curve going too fast or die going "up hill" too slow to make it over the hump.
You could have a watertower-shaped pulsejet, or a Reynst drum with the variable floor feature to walk up the speed of combustion or slow it down in the Dali diorama as well. There could be all sorts of governing shapes to tweak the flow and route the gases as elaborate as a circuit board, with lots of air diodes for good measure. You could quickly store some created pressure in a large sphere like a capacitor stores electricity and reroute it to some other air gating track. Maybe the whole thing could become a pinball machine, coordinating simultaneous events with levers as you improve your skills with fuel, air and synchronized sparks keeping the flame alive.
Yesterday, I got out my "old" 2 1/2 gallon steel tank with a equal length of 1 1/4 inch plumbing pipe which makes the whole thing ~36 inches long. The tank is 8 inches in diameter. The tank looks like a fat little air compressor tank round at both ends with 6 various place and sizes I can screw pipe or pipe fittings into. I don't remember if I put some acetone in with the methanol in this experiment but I stood the thing on end under my peach tree that is about 14 feet tall. It's starting to lose its leaves for winter and it was so funny to watch the tree rythmically buffetted by the jet of hot/warm air ejected out of the engine running jam jar style. It ran the longest I could rememeber sounding like a Harley motorcycle purring at perhaps 20 mph.
If you stand next to the chuffing and cup your hand about 3.5 feet above the exhaust where it's not too hot, it will hold your arm up in the air almost. It's not deafening but kind of hard on the ears, but for just breathing and exhaling from the one orifice, it throws out a surprising amount of air. I have a clothes drier air amplifier of sorts, it came out of an old drier and it is composed of a tapered cone converging with a larger cone with three spars to connect them and a gap for air to be drawn with the outward flow. I should try place this over/near the exhaust of my 3 foot oddity to see if it somehow improves combustion.
Mark

[Mike Everman]

Mark, you wax more poetic with each post. "Air diodes" reminds me of some study I did 20 yrs ago on fluidic circuits. It was wild stuff, maybe old news to you cats, but one can make any possible logic, amplifier or oscillator circuit with jets, walls, branches and control ports. the air force was doing extensive work making full control systems with this, because it is solid state and most importantly, nuclear hardened.
Mike

[Mike Everman]

2 as above but bend the top third of the tubes to opposed format as per your drawing layout.

4 or 8, I really want to preserve the inlets crossing the CC, so it's two bends. I still feel that flow through two bends in a pipe is going to have very different dynamics, and if we are going to tune the lengths, it should have the bends first.
If I start with the two bends, and make the ejectors telescoping, our experimental starting point has the ejectors pointing radially out since we have no augmentor/rectifier yet, so we can temporarily put a big tube around as a blast shield.

Anyone have any analysis on flow through bent pipes as relates to straight? Google came up dry out to the 6th page or so.

Mike

[Bruno Ogorelec]

Anyone have any analysis on flow through bent pipes as relates to straight? Google came up dry out to the 6th page or so.

Mike, I e-mailed you some stuff that I have. Not much, but it should help. As for bends being problematic, please look at this Russian 'trombone'. Remember that in Russia, they were taking this seriously. This is not two-teenagers-in-a-garage stuff. If they say in a textbook on pulsejets that it worked, well, it worked. (The reason to bend it into a pretzel was to make it shorter, of course. Pulsejets are mostly indecently long.)

Your intakes are nothing like that, remember, and in the intake phase, they will be pumping cool gas, meaning lower speed, less resistance and less turbulence.

Finally, if you really worry about bends, use the 4-pipe layout and make your pipes flat (oblong section) in the horizontal plane, not vertical. This is usually good practice in bent high-speed fluid conduits.

Bruno

[Mike Everman]

Thanks much, Bruno.
I want more than anything to twist the inputs, but can so far think of no way to then interleave them up above if I did. It's what led me to sketch up the mini one piece design concept, which essentially has one pipe.
Mike

[Bruno Ogorelec]

I don't quite get it? Why can't you interleave them? They look easy enough to stagger. Am I missing something?

Bruno

[Mike Everman]

Bruno-
The depth of the concept engine in and out of the page, is 4". I made the tubes rectangular and vertically oriented to mimic the Kentfield inlet area required, and still have 2-3 times that area between them for the augmentor inlets. If I twist the tubes horizontally, they would have to go over and under creating an impossible situation for the augmentor inlets. I'm currently playing with bent round tubes in different configurations for ease of manufacture.
Mike

[Stuart]

Anyone have any analysis on flow through bent pipes as relates to straight? Google came up dry out to the 6th page or so.

Mike, I e-mailed you some stuff that I have. Not much, but it should help. As for bends being problematic, please look at this Russian 'trombone'. Remember that in Russia, they were taking this seriously. This is not two-teenagers-in-a-garage stuff. If they say in a textbook on pulsejets that it worked, well, it worked. (The reason to bend it into a pretzel was to make it shorter, of course. Pulsejets are mostly indecently long.)

Your intakes are nothing like that, remember, and in the intake phase, they will be pumping cool gas, meaning lower speed, less resistance and less turbulence.

Finally, if you really worry about bends, use the 4-pipe layout and make your pipes flat (oblong section) in the horizontal plane, not vertical. This is usually good practice in bent high-speed fluid conduits.

Bruno

Bruno, you have more neat pics than anyone. Regarding my comments to you on velocity versus pressure head, I obviously don't think you have forgotten pressure head. What I am recalling from my fluid dynamic classes is the "relativistic" observer viewpoints. That is, is the observer on the ground, or on the pulsejet. Depending on which observer you are, a flow can seem stagnated while to the other observer it is still in motion. Ultimately, what I am trying to figure out is how much additional thrust is gained by a pulsejet in motion, if any. I've found some papers on some of this, but they are conflicting.

By the way, what I am working on is technically a ramjet, though all this stuff ends up being the same because of the observer effect.

[Bruno Ogorelec]

Stuart, you are right, of course.

When thinking about ducting, I usually adopt the attitude of Keith Duckworth. He once said he designed ports in his racing engines by imagining them very, very big, with himself inside. He would then think, "Now, how do I slide down there without tearing my britches?"

How's that for placing an observer?

Thanks for the compliment about pictures. I try to have the best collection. I've put a lot of effort into putting it together. I think that we will reach a fine level of understanding only once we pool the sum total of all the world's knowledge about pulsejets and analyze what we've got. I have no hope of getting there (my math skills are too primitive) but somebody will. Maybe my efforts will help that someone along.

Bruno

[Mike Everman]

As for bends being problematic, please look at this Russian 'trombone'.

I didn't mean that the bends will be detrimental, I meant that in order to design the starting point length of the intakes, I would need to know how bends effect the "apparent" length of the tube. all indications are that bent tubes have the drag of a substantially longer straight tube of the same diameter. so you see my problem; I certainly can start with tubes with a median length per Kentfield, knowing that they are conservatively long and I will be cutting bits of progressively to get it to go resonate, but I may need to increase the diameters so that I can make them long enough to interleave them properly. Whatever "properly" means at this point. thanks for the bent tube data, it doesn't seem to want to load the figures even from the same folder, but I'll get it, and I love the trombone.
Mike