Cast-Nozzle I
Rocket engines made from trash
Flight Test
How To...

I watch with envy as Stuart Leslie's beautiful aluminum micro-engine goes together using his drill press as a metal lathe.  Clever.  Very clever.  Now why didn't I think of that?  Simple.  I am not that clever.  I am a klutz.  As will be amply illustrated on this very web page, and again on several that follow.

But I want a small reloadable engine case like Stuart's, and I want it very badly.  I want it because my propellant is well adapted to making tiny grains, and I am actively testing new propellant formulations and configurations.  Smaller reloadable engines should reduce the cost and effort of testing as well as increase the safety margin.

So I was walking around Home Depot the other day and happened upon some aluminum tubing, 1/2 inch outside diameter.   I bought a 3-foot piece just because.  Tried it as a blowgun.  Played it like a trumpet.  Twirled it like a baton.  Hit myself in the head, and through the circling stars had a vision of pouring molten metal into it, thus casting a nozzle-blank.

The tube measures exactly 0.5 inches outside diameter, 0.400 inside, leaving a wall thickness of 0.050 inch, just right for a micro-engine, don't you think?

Now for the nozzle experiment.  I want my nozzle to have an exit cone, so I grind down a large nail to a sharp point, somewhere in the ballpark of 12 degrees from center.  Since this is a test of concept, I am not overly concerned about the exact angle.   I would also like for the nozzle to have a convergence cone inside, but will burn that bridge when I come to it.

The nozzle pin is clamped in a vise-grip, and a short piece of tube placed over it.  A bit of aluminum is melted in my handy little charcoal foundry, and some of it poured down the tube onto the nail-point.

The result is a crude, off-center nozzle.  And I squashed it a bit too much in the vise while trying to wrench the divergence pin out of the solidified aluminum.  But it worked pretty well, considering.  Definitely worth another try.

Another Try

This time I decide to pour from the nozzle end.  I plug the tube with a metal rod (the rest of the large nail) so that only the top inch or so is open, pour it full of molten AL and ram the cone former in while the metal is still liquid.

It worked!  Sort of.  The aluminum hardened almost the instant the former touched it, so the exit cone is not very deep.  I will use it anyway.

I cut off the greebly end with my chop saw - a carbide blade cuts aluminum pretty well, and doesn't seem to dull it too quickly.  Warning!  In addition to eye and ear protection, be sure to wear a big hat - those chips are HOT coming off the saw.  One fell down my collar - I don't dance very often nor very well, but did have some interesting moves at that moment.

A minute or so on the grinding wheel cleans up the edges rather nicely.

Now to drill the throat.  To keep from mashing the tubing flat in the vise, I try the old gunsmith's trick of making lead blocks to clamp around the tube, and dusting the surface with powdered rosin.  I did a poor job of making the lead blocks.  It worked better by wrapping the tube in an index card with a little rosin on it.

The head-end is threaded with a 7/16 inch coarse SAE tap. (Apologies, I didn't get a picture of that.)  I wanted to use fine threads but the Tractor Supply store only had NC threaded screws.

I make something of a convergence cone with a drill bit.  No, it's not 30 degrees, but it is the bit that I have.

My case isn't nearly as pretty as Stuart's, but looks like it might work.

Now that the case is complete, I heat a little rcandy and make a single hollow uninhibited grain.  This one is hand-rolled, thus a bit uneven and dirty since SOMEONE didn't wash his hands.  But it should produce high thrust of short duration, giving the new case a good test.

This Frankenbolt weighs several times what the case does.  But since it is just for static testing, that won't matter.

Test 3/30/02a

Click here for a video of this static test
(660k file, 5 seconds of video)

Grain wt:  6.3 grams
Diameter:  0.4 in.
Core diameter:  0.125 in.
Grain Length:  2.0 inches

Nozzle throat diameter:  0.125 inches
Initial Kn:  287
Final Kn:  254
Thrust duration:  0.22 seconds
Estimated thrust:  7 Ns

Thrust duration is determined by looking at the sound file.  Here the ignitor pops at 3.94 seconds, thrust-sound starts at 4.11 seconds and ends at 4.32 seconds.

(Illustration generated with Goldwave)

Interesting that the nozzle is now displaced a hair backward.  There is evidence of leakage around the perimeter.  I take this as a bad sign.  Nozzle blowout seems likely if the pressure is increased.


But I have to try it again, this time with tiny Bates grains.  Aren't they cute!  These were pressed into a short section of the same aluminum tube in which a layer of plastic-tape-covered manila-folder paper has been rolled.  The cooled grain is peeled, coated with epoxy, and rolled in heavy paper again.  The hollow core is made by inserting a wooden dowel through the warm propellant while it is still soft.  In this case, it was a bamboo skewer about 1/8 inch diameter.


Click here for a video of this static test
(435k for about 3 seconds of video)

Total grain wt:  6.1 grams
Diameter:  0.38 in.
Core diameter:  0.125 in.
Grain Length:  0.82 inches (average)

Nozzle throat diameter:  0.125 inches
Initial Kn:  127
Max Kn:  168
Final Kn:  164
Thrust duration:  0.35 second
Estimated thrust:  7 Ns

This one burned longer than the first, about 0.35 second.

Beware that the height of a measurement on the sound-graph above may not be an accurate measure of the volume of sound generated by the rocket engine.  The camera used to record this audio has an automatic level control.  Thus the apparent gradual decrease in volume during this burn may be an artifact of the camera reducing its sensitivity in the presence of a loud sound.

Kn is skewed toward the end of the burn because the grains are longer than the optimal Bates length.

The nozzle is still set back, but not any farther.  It shows about the same leakage as in the prior photo above.  I'm sure this tubing can handle much more pressure than has been applied here, but the nozzle might well blow out.  So I need to find a better way to fuse the nozzle into the case.  I will try many things...

Next:  Larger tubes!  Nozzle secured, leakage enhanced!  G, H engines!  Headers blown, flaming propellant disturbs local ants!  Reloadable "model rocket" engine!  Face of Elvis seen in smoke trail!


Jimmy Yawn
Recrystallized Rocketry