Parametric De Laval Nozzle, for Rockets!

Here's something fun to play with!

I'm working hard to understand both the nozzle design and OpenSCAD, so i only have a very simple version modeled.(attached below)

plug in your parameters and you will have a simple conical De Laval nozzle negative. By casting something around it (plaster, cement, etc) and then melting out the core, you can have your own rocket nozzle.

I hope to make this into a printable nozzle that you could attach to things. obviously not for combustion, but attaching it to an air compressor hose could be fun, or a balloon. Should stand up well to steam, too, as long as it's bellow the melting temp of the plastic (please, please be careful with hot steam).

of course, if the code is terrible please help make it better.

So far the best page i've found for just general, rule-of-thumb dimensions is here:
http://www.braeunig.us/space/propuls.htm

Anyone with experience with openscad who wants to try a parametric bell nozzle, Please do! That page has all the parameters about half way down.

This is cool! I wonder if you could combine it with this to make water rockets. Although the nozzle shape probably doesn't affect the performance of a water rocket that much...

You could also make ceramic nozzles if you have access to a clay printer like this one or this one.

It would be fun to put in embedded cooling channels. Making a model of an A4/V-2 rocket engine would be really fun too.

I've got the convergent section done for a bell nozzle (attached below). the divergent bell is really tricky. I can get an elipse by using a conic section, but getting it tangent to the throat exit radius is, I think, beyond me.

I'm even having trouble getting the conic nozzle to be tangent to the throat exit radius. Openscad makes it hard, at least for me.

From what i understand, water rockets wouldn't benifit, since it's an incompressible flow situation. I've read it would even hurt performance, probably from the surface tension slowing the flow. However, I'm wondering if it would work in a vacuum.

OK! This one actually looks really good and follows the specs. (well, the throat is from the Bell section and the rest is from the Conic section of the website listed above) How well the specs work is beyond my ability to calculate, but at least it's something to test.

The "hull" command makes the nozzle cylindrical. I don't know how much force there is at the throat, but if there's a lot, then this makes that part thicker. it could also be used for mounting holes.
Commenting out the hull command makes it much lighter, though.

Slight change to the file. the funnels were being cut incorectly.

I'm still not having any luck with the parabola for a bell nozzle. I tried a 2d plotting module, but that wasn't great.

This little chunk of openscad will create a parabolic bell. You can control the "sharpness" of the parabola and the thickness of the shell. It works by rotate_extruding a bunch of quadrilaterals around the Z axis. With some more work, the circular parts of the throat can be added, and the tangents made to line up.

Thanks! I'll see what i can do with it!

The nitty gritty of rocket nozzles is complex. it is rocket science, after all. I suppose as it gets more complex, the inputs should be more in line with the variables for the formulas rocket scientists use. That's all very complex, though, and beyond my abilities, so I think this version will be the "Hobbyist Nozzle".

EDIT:
Your code if fine, but i need to integrate it better into mine. I need to figure out the starting angle of the parabola at different sharpnesses. ~60 deg works for .36, and ~80 deg works for .16.


EDIT 2:
I attached a cleaned up, modulized version.

Edited 2 time(s). Last edit at 03/10/2014 04:05PM by Buback.

Ok I've just learned that we can describe the shape of the bell as a quadradic bezier curve. we need three points:
the1st is the point where the divergent nozzle is tangent to the throat radius;
the 3rd is tangent to the exit radius;
and the 2nd point is where those two tangents intersect.

I would like to be able to define the angles of the two tangents as variables.

Still learning how to program this in openscad, though.

Thanks to Bezier curve scad by donb on thingiverse!
This is much, much better! I need to clean up the code, so i won't post it yet, but this really looks like a rocket nozzle.

Hmm. This just occurred to me...

I'm assuming there's nothing wrong with this in regards to ITAR. All the information i pulled for the nozzle is from Wikipedia or elsewhere openly on the web.

I put the code up on my Github, if you want to check it out and contribute.


It is pretty good, but there are a couple places that need refinement. I kinda just guessed on the second point for the bezier curve. it's been way too long since i've been out of school

The p1 tangent, where the bell meets the throat, also doesn't look tangent at certain angles.

the combustion chamber is just a very rough outline.

Much progress! Check out the github for the latest code.