Rod and Tube Motor With a Perfectly Flat Pressure Trace

Nov 16, 2026 Last reply: 2 hours ago 3 Replies
Jump to replies (3)

The most surprising part of this build is the grain itself: instead of one cast charge, Andrew Reilley loaded the case with two separate propellant pieces. There is an outer tube that regresses outward much as a conventional BATES grain does, and a free-standing rod hung from the forward bulkhead that consumes itself from its outer surface inward. The maker nicknamed that suspended rod 'The Propsicle', and the geometry behind the nickname is genuinely clever. Putting the motor together involved some serious tooling. The case is a filament-wound carbon fiber composite, produced by a machine laying tow over a rotating mandrel. Up front there is an aluminum bulkhead plus an aluminum nozzle closure ring, a phenolic insulator washer bored on a metal lathe, and two black O-rings. The propellant was cast in a 3D-printed cylindrical mold, with the rod mold lined in mylar to help release. The rod carries a threaded rod for mounting, then the whole core slides into the outer case. OpenMotor handled the grain regression and performance modeling beforehand. The simulation called for a 2.245 second burn with 1.517 lb of propellant at a flat pressure near 690 psi. The actual firing ran 2.572 seconds on 1.544 lb, delivering 1440.1 N-s of total impulse at a specific impulse of 209.713 s. Peak pressure reached 897.281 psi, but that came from a brief spike near 880 psi at 0.05 seconds after the igniters briefly clogged the nozzle. After that, the trace held a flat 700 psi plateau for the rest of the two-second burn, closely matching the prediction. The firing took place nozzle-up in a vertical steel test stand, strapped down to a gravel pad with two ratchet straps. One open item: a note attributed an apparently high C* reading to nozzle slag. Whether that slag can be cut down in future firings seems like the natural next question for this motor, and I would happily watch Andrew Reilley chase it.



Project by Andrew Reilley. Watch it on their YouTube channel:

formatting link


The choice of a four-port divergent nozzle is the detail that grabs me. Splitting the flow through four ports instead of one big throat is a bold way to match a dual-geometry grain, and the measured average pressure of 640.212 psi landing so close to the simulated 690 psi suggests it worked. My only thought is that four ports must make machining and inspection much more finicky than a single throat would.

How did the maker keep the inner rod centered inside the tube during casting and firing? The threaded rod mounting on the forward bulkhead is mentioned, but with the rod burning from its outer surface inward, even a slight tilt at ignition seems like it could make the regression uneven. Was the mylar lining in the rod mold the whole story, or was there extra centering hardware in that 3D-printed mold?

For the next firing I would suggest scaling up the igniter. Since the igniters briefly clogged the nozzle and caused that 880 psi spike at 0.05 seconds, a slightly larger pyrotechnic charge, or one placed differently, might clear the ports faster and keep the start transient clean. That spike is the single blemish on an otherwise textbook pressure trace, and it seems like an easy thing to tune out.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required