50 mph ramjet???

[sockmonkey]

This is similar to an idea I had which started out as a a ramjet coiled into a spiral and spun which naturally led to a more conventional shape with vanes in the intake and exhaust which led to a design similar to what you propose. Great minds think alike etc etc.

[paul fellows]

is this relivant viewtopic.php?f=18&t=5785#p69804
posted on the hybrid section

[sockmonkey]

I was thinking something like this.

It would need to spin rediculously fast though to maintain itself, so maybe magnetic induction bearings? The grey tube in the middle is the fuel feed.

[makulit]

I have a deep interest in slow ramjets because I would like to build something like a solar updraft tower, except that I'd like what I build to be a lot more compact. Basically ramjets make the air come out the back end faster than it went in, right? and the faster it goes, the greater the thrust you can generate. So if you put the jets in series, with something like a thrust augmenter for the intake to the 2nd and third engines, the first two engines might accelerate the air so you could really do something with the third engine.

[sockmonkey]

I see what you're getting at, but how do you get the first one to work well enough at low speed to generate enough thrust for the second one in the first place? It you solved that problem, then the whole problem is solved and you wouldn't need the other two anyhow.

[makulit]

Well I'm counting on the jet to be able to double the velocity in each engine. In my first thought experiment on this I imagine rolling down the highway at 50, 3 stage ramjet on the roof. So after the first one, the velocity is 100mph. The air is also 600°F or so. Next engine takes that air and some fresh so that the intake air is maybe 450°F moving 85 mph or so. The next engine has to heat this air to 1000°F or so, and the exhaust will be traveling at 160 mph. Lets say it then enters the third engine at 150 mph, 900°F. In the last engine, heat the air to 1500°F, and get a good exhaust velocity and maybe 1/3 lbs of thrust per square inch, which while not great, is also not bad. Seems like it would use a lot of fuel.

In the solar stack, I can get at least 10 mph from the chimney effect. If I somehow have concentrators arranged to heat grids inside the stack, I could use the same idea to get a 30 mph updraft, which would be outstanding for a wind turbine. I'm pretty sure I can even do better if I introduce water / water vapor to the system because water vapor is better than hot air at rising.

hmm. I wonder if water injection would help the first case, the roof rack ramjet.

[sockmonkey]

Yeah, I see how it's supposed to work. I'm just not sure you can get enough of a boost from each stage that it won't have to be impractically long. I think others had the idea of a pulse jet as the first stage in an arrangement like that.

[makulit]

Yes, I'm worried about the practical length too. To counter that I'm thinking of several or maybe many smaller jets in the first and second stages. To simplify construction they would be 2D ramjets - like little boxes with open ends and center bodies. And really, I would have a tough time building anything but 2D ramjets If I'm going to build a lot of them that should be similar. I should probably make a model so this is more clear and to see if I'm thinking about it the right way.

[makulit]

Here's what I was thinking 20-30 lbs thrust at 50 MPH

[metiz]

Makuilt, will there be fuel injected at each stage? if so, you're going to have an oxigen and temperature problem.

[Pscott]

Just a couple of notes on your design.. Looking at the mass continuity accross the engine and some rough mathematics, it seems you only need an exit velocity of 29 m/s (65mph) in order produce 30lbs of thrust, this is based on gross thrust and neglecting engine drag. In fact assuming ideal conditions and a conventional subsonic ramjet design, the actual exit velocity of an engine with your dimensions from thermodynamic equations is probably in the region of 150 m/s. But will be far less in real life due to the poor combustion efficiencies from the low pressure recovery generated at low airspeeds.
The centrebodies will actually reduce performance at these very low Mach numbers due to the increase in drag and a reduced intake area, which will in turn reduce air mass flow and so reduce thrust. The hot exhaust from the stage in front will also drastically reduce the pressure recovery of the intake operating in it. My latest projet will see me testing the performance of a micro ramjet in the high speed exhaust of a small gas turbine engine. The high ambient freestream temperatures and the the effects of the combustion products on the gas properties both effect the engines performance. In fact running the simulation for ISA freestream consitions gives an increase in thrust of more than 400% compared to the engine in the jet exhaust where the freestream temperature is 400`C.

[makulit]

For thrust, I'm using Decker's chart from practical ramjet design, extrapolating for the lower speeds. If the thrust is actually better, awesome. If I can cut out a stage, another awesome. Regarding the centerbodies and drag: The bodies are streamlined, the airspeeds are slow, and unless I missed something the intake area has to be between 1/3 and 1/4 the combustion / heating cross section area anyway, so theres not much getting around either a necked opening or a centerbody. The centerbody will let me move the actual object around a lot without major rework, so I can fine tune the locations. (planning to bolt them in) Thinking more about the rough math you mention, I'd like to see what equations you're using. (Did you use a mass/energy flux analysis?) Also when you say 65 mph gives a 30 lb thrust, what do you mean? a 65 mph increase in velocity? What area is this thrust applied to? (or what volume is assumed?) Thanks.
Oh, and I looked at the linked page - very nice explanation of what you're doing. I especially like the background image. I would love to see that ramjet used on the P-61 up close and personal and know how much thrust it produced at what speeds. I understand what you mean about the problem of the intake being in the hot exhaust, and my plan was to let the cooler bypass air be grabbed by the hotter faster exhaust air, slowing it a bit, cooling it a bit, but significantly increasing thrust (at least based on what I read about thrust augmenters). Really I want to get building and start getting numbers. Measuring velocity and temperature will be interesting. I haven't planned that out yet.
Oh, it might not be obvious, but the open area is pretty much the same at each inlet stage.

[Pscott]

The 65mph quoted would be your total exhaust veleocity, therefore only a small increase over freestream speed.. Personally I'd remove the centrebodies and reduce the engine diameter, this will reduce drag and still give you the same cross sectional intake area.
The way I approach these things is to start simple by assuming the engine is operating under an 'ideal cycle', this means that no heat is lost, (aqdiabatic) and any compression and expansion is isentropic and fully reversable. Then it just a simple case of appling perfect gas laws at each stage, and using a few assumptions, (like the combustion temp), you can get an idea of the engines performance.
When the engine is operating under steady state conditions, (constant throttle posistion, steady airspeed ect), and with good estimates of each components efficencies then its not that inaccurate. I wrote two programs to simulate the steady state and transient performace of small gas turbines, and the steady state model was within about 5% of the experimental results for the temperatures, thrust, fuel flow, mass flow ect.. The transient model was less accurate, when compared to the engines actual performance during say a minimum to maximum throttling. This is mostly due to the heat transfer occuring in the engine as it moves from one steady state condition to another, which I'd neglected in the model at the time.
Anyway I cant write anymore because I'm literally running out of the door before I go on holiday .. Good luck though!!!

[makulit]

Thanks for your comments. I will probably keep the center bodies during development because it simplifies construction (remember I'm building engines whose dimensions are constant in one dimension) because I can tweek the positions easily. I don't know if everyone already knows this but if you wanted to make fine adjustments on a thing thats bolted or screwed together there's a neat way to do it that's like reading a vernier. So we will have a series of holes right? but the holes are 1 cm apart, hardly fine adjustment on a small model right? well if I have 10 holes on one side, and 11 on the other where the 11 first and last are the same distance as the 10 first and last, I can make 1 mm (rather than 1 cm) adjustments. You've triggered my "Have to find out" mode and I will have to look at the difference in drag between the two designs (with and without center bodies) I should be able to compare them based on the drag coefficient and the area. This weekend I hope to be able to build something for some testing. Hot Jets!
http://www.grc.nasa.gov/WWW/K-12/airplane/foil3.html
This fun little applet lets one compare drag and various shapes. When I compare flat plates on an angle (an inlet without centerbodies) to a symmetrical airfoil, I get that the drag on the flat plates (10° Angle) would be about 10 x the drag on the centerbody. If that's anywhere close, it should show up in the tests. I'll take pictures. I'm planning to set the models up like ballistic pendulums and use a leaf blower for airflow.

[makulit]

Update: I revised my assumptions regarding the flat plat at 10° for the no centerbody design and used airfoil sections instead and was able to get the no centerbody drag to be about the same as the drag on the centerbody. Meanwhile I have stumbled upon a treasure trove of NACA documentation and I intend to download it. I'll mirror the data and post links to the original and mirror later.