Rocket Assembled!

This weekend marks the initial assembly of the test rocket. If you take a look at the pictures, everything looks pretty clean. Well, the pictures make it look a lot more effortless than it turned out to be. The initial attempts at welding went… poorly. I had machined a simple welding fixture to keep everything in place when I gave the parts to the welder. Unfortunately, the weld went through in multiple places, securing both the rocket and the fixture in place. After chiseling and boring away for a few hours — all but destroying the original pieces — I realized that I probably needed to machine new parts and start over.

I cut off the failed weld from the combustion-nozzle assembly flange. I was able to salvage most of this piece. However, both flanges and the injector dome had to be machined all over again. As I already had both material and the CAM programs available, this was not nearly as daunting as it first seemed. Within two weeks I had remade the parts and prepared for a second welding round. I discussed what might have gone wrong with the welder, and we made a plan to improve on some of our methods. I added an inset to the flange rather than attempting to fixture it from the inside. The welder thought this and a slower weld process would have better results.

As you can see, the methodology improvements paid off! After receiving the parts back from the welder (which they were kind enough to do free of charge!), I did some final machining on the welds to clean up the finished appearance. I drilled the flange holes to the proper size to accommodate #6 bolts. Finally, I drilled and tapped two holes (1/8″ NPT) for the instrumentation on the combustion chamber.

The end product looks great given the difficulties I have had to get to this point! I’ve learned a lot of lessons putting this together, as painful as some of them were. It’s not perfect, but it’s in good shape to gear up for a test in a few months. There are a few minor final assembly activities (fitting O-ring, etc), but the bulk manufacturing for the rocket is now complete!

Now that the rocket is complete, I have turned my attention to the test stand. We have been collecting pressure transducers, thermocouples, load cells, and all of the fun instrumentation! The plans are coming together for this, and some parts are already ordered. I don’t expect the test stand will take much longer than a month to complete at the current pace. I’m tentatively shooting for an April test fire. We’ll see how that plays out.


Rocket Engine

The Plan

For this post, I would like to step back a bit from the typical software discussion, and provide a little insight into my latest hardware project (okay, there is some software!). Over the last few years, my friend and I have been designing components for a small, hobbyist-level liquid rocket engine. We had a few goals in mind when we started, and trust me when I say they were quite lofty:

  • Generate “some” thrust — maybe 50 lbf
  • Replaceable injector for design iteration
  • Don’t get burned

Okay, so they were actually pretty basic. The first thing we did to work towards achieving this goal was break the task up into several iterations. The first iteration happens to be a glorified propane-air torch that also has a “kick” to it. Perhaps the coolest feature of the rocket is the replaceable

injector, which is based on a coaxial pintle design. In preparation for the next iteration of the rocket, we will test various pintle designs to optimize performance and reduce the number of parts that will have to be remade. These can later be integrated into the injector dome when performance properties have been measured and confirmed.

The Challenge

One of our early challenges was manufacturing. After we had created the drawings for each component of the engine, we set out to collect some quotes and get some parts made. It turns out that machine shops charge exorbitant prices for one-off parts. It was about half as expensive to have the parts made in aluminum, so we had the shop start with the throat/nozzle. Of course, this change in material meant that combustion would have to be largely undersized, and the burn times would have to be drastically reduced. For this iteration of the motor, we decided, these were acceptable tradeoffs.

However, the price did get me looking for alternatives. I found a local maker space that offered access to their CNC mill and lathe for a very reasonable monthly cost. I figured that by the time I learned to machine the parts myself, I would still come out ahead. And so I have spent the last few months between my space simulator project and my day job learning how to machine and perfecting my methods. I am just now beginning to complete the final set of parts for the rocket. We hope to have these ready to be welded in the next few weeks.

Pintle Details

The pintle is really the star of the show. This device is easy to manufacture, and provides excellent mixing characteristics. The concept is simple: inject the fuel (in this case propane) radially from the center tube (known as the pintle). Flow the oxidizer (air) through a concentric opening traveling perpendicularly to the fuel. The two propellants collide and ideally atomize as they travel through the combustion chamber.

Because this is threaded to the body, we can swap the pintle out easily and collect performance data for each. The fuel-air throat areas can be easily altered by changing the pintle OD and injector hole diameters. The number of injector holes can also be optimized, although as they become smaller, they begin to pose a manufacturing challenge.


I hope to dive into some details about the rocket as it gets closer to completion. We are moving towards constructing the test stand and software to support the static fire. If the rocket can be assembled by March, I hope to have the first hot-fire test as early as April or May. As always, stay tuned and we’ll see where this goes!