Structure: Redesign for CG

Happy New Year, and Happy 1st Anniversary of the HAPP project! It's been a whole year since I decided to undertake this project. It might also seem to have be a year since my last blog update...

There is a reason for my lack of blogging. Well, two reasons actually. The first reason is that I got a new day job in mid-July, and I've been pretty focused on the new gig. Gotta pay the bills for the HAPP project!

The second and more crucial reason is that I hit a roadblock as I started to assembly all the flight hardware, especially the structure. Turns out I had balance issues. Specifically, I couldn't figure out how to package the guts of the HAPP such that the center of gravity was sufficiently low. The CG must be lower than the aerodynamic center of pressure if the HAPP is to remain aerodynamically stable during the free-fall portion of its descent from balloon burst down to the altitude for parachute deployment. If the CG is too high, the HAPP gets tippy.

Think of a shuttlecock - most of the weight is at one end, the nose, and the center of pressure (where the wind catches it) is at the other end, the feathers. This make it stable as it flies. We're trying to achieve something similar with the HAPP, but we want it to be stable with the lower shield facing downward - you may recall that the HAPP is essentially a scale replica of an Apollo Command Module.

The "final" design I've been using was based on some internal packaging that evolved from multiple prototypes, and it relied on some rough hand calculations for the center of gravity. I thought I could roughly calculate the CG accurately enough that some final adjustments after building the main structure would be sufficient to fine-tune the CG. Not so! After weeks of staring at a lab table full of parts, I just couldn't make it work.

To find the solution, I had two options. Option 1 was to keep buying lots of parts and try, try again until I found the answer. This is OK when dealing with plywood and cardboard prototypes. This is not OK when dealing with flight hardware that's primarily carbon fiber - the stuff is expensive!

Option 2 was to model every single part in 3D software and play with the design until I figured it out. This option is cheaper, but requires time to model all the bits and pieces. Every screw. Every panel. Every piece of electronics. Every pneumatic component. All of it.

Having already spent too much money on this project, I went with Option 2. I used AutoCAD Fusion 360, which is an excellent free program. It took some weeks, but I finished the modeling, and I found my solution to the CG issue. It entailed a serious re-arrangement of the HAPP's guts. As a bonus, I was able to cut almost 1 kilogram of weight out of the design. That means higher potential maximum altitudes on flight day.

With the design completed, I have started ordering parts and fabricating the final structure (no really, it's final this time!). In the coming weeks I'll post some photos and videos of the build process. In the meantime, here's a rendering of the section between the lowest deck (the tank deck) and the second deck (the propulsion deck, which supports the pneumatic valves). I omitted the pneumatic hoses and some wiring for clarity. Two of the three canisters holding the main parachutes are just visible at the top of the image.

With the CG roadblock cleared, the project is back on track, and with a hard lesson learned. Sometimes there is no substitute for thorough, quantitative design work before you build!

Valves and low-pressure regulator hanging down from the Propulsion Deck. Tanks visible at bottom.

Bonus pic: 3D-printed thruster pack and quick-connects for air hoses, mounted on the end of the jet arm