Going down in flames: how to self-destruct with rockets

The big day was finally here.

man standing in front of small trailer holding large red and white rocket
my pride and joy

I finished building the L3 Fusion rocket in early September and was ready to launch – once the wildfire smoke cleared in the PNW – as soon as the opportunity arose. And in late October, I had my chance.

The plan

On a frigid Saturday morning, with my wife joining the small crowd gathered at the rocket launch out near Walla Walla, WA, I went through my pre-launch checklist and got the rocket ready for flight. It was mostly ready to go – the black powder charges were prepared and loaded inside the rocket, the M-1297 reloadable motor was already built, the wiring for all the electronics was nearly complete. All I needed to do was plug each flight computer into its respective battery, turn on the GoPro camera, and seal up the rocket with a few rivets. And, of course, install the motor. Easy enough.

I’ve described this rocket before but just to quickly recap, the L3 Fusion is a 5.5″ diameter, nearly 8 ft tall high power rocket specifically designed for level 3 certification. It’s available from SBR at fusionrocket.biz and I highly recommend it. The rocket is cardboard and therefore lightweight (only 11 lbs before adding the M motor, which itself weighs another 11 lbs), but it’s reinforced and double-tubed from top to bottom, and then coated with an epoxy – basically making the rocket incredibly strong despite the light weight. On an M-1297 motor, this thing should fly to 9,000 ft or higher.

The key word, of course, is “should.”

I was a bit nervous, but mostly hopeful and excited. The temperature that morning was brisk – around 30 degrees F – and it didn’t take long for my fingers to get cold and then start to feel numb. It’s particularly difficult when you’re trying to mess with very tiny wires and electronics – think eyeglasses screwdriver (which is literally what I was using to attach wires to flight computers).

man holding large rocket horizontally above his head, walking through field
getting the rocket on the pad

But I had built this rocket entirely under the watchful eye of the man who designed it, with his recommendations. We even filmed the entire build as a tutorial for future generations, so this event might go down in history. I can’t say I built the rocket flawlessly, but I was pretty confident the flight would be successful.

As you have probably guessed by now, it was not.

rocket launching with cloud of smoke underneath
L3 Fusion liftoff!

The disaster

The countdown began: 5… 4… 3… 2… 1…

With a thunderous roar, the rocket shot off the pad and climbed into the sky with lightning speed. An M motor is a pretty powerful one, and so this was expected. What was not expected was just a few seconds into the flight, as we watched it ascend and disappear into the sky, was another loud boom. The smoke behind the rocket, which was otherwise basically a vertical line, suddenly changed as the rocket veered sharply from its trajectory.

It broke up and fell back to the ground in multiple pieces, and the certification attempt was a bust.

We mounted a search with half a dozen people scouring the hilly area where we saw the parts land, and we were able to find and recover everything except for the rocket’s three fins. The fins were completely torn off, but a lot of the rest of the rocket was largely undamaged. We even found the electronics, despite the fact that the e-bay fell separately from the rest of the rocket and it’s quite small and difficult to spot in small bushes and tall grasses on a hill.

damaged rocket parts scattered on ground
the aftermath

The aftermath

You can learn a lot from studying a rocket failure, just by seeing what happened to the airframe. You can sometimes learn even more if you recover the electronics and download the flight data (assuming they’re still working properly), and/or from an onboard camera like a GoPro.

In this case, it seemed obvious that the fins experienced fin flutter, which is a phenomenon where the forces acting on the fins are much higher than they should be under normal flight conditions, and the extreme vibrations can either change the rocket’s trajectory or even destroy the fins.

Leaving aside complicated discussions of aerodynamics, fins are really important to a rocket. The rocket itself is streamlined and has a motor at the bottom which accelerates the rocket upwards (vertically), but anytime the rocket deviates from that vertical path, the fins stabilize it. The air pushing against the broad fins with large surface area pushes the bottom of the rocket back into place. It’s an ingenious system that self-corrects without the need for a sophisticated computerized guidance system. (Very sophisticated and large rockets tend not to have fins precisely because they do have such computerized guidance systems.)

Without fins, the rocket has no stability. In this case, the moment one or more fins were damaged due to flutter, the rocket careened significantly off its straight trajectory. Since it was still traveling at very high speeds just a few seconds into the flight, the forces acting on the rocket were tremendous and it was almost instantly destroyed.

close up view of damaged rocket with missing fins
the fin can, minus fins

As you can see in the picture above, the entire bottom of the booster section of the airframe was destroyed and all three fins were torn off. Some of the rest of the airframe was damaged, despite the fact that it was double tubed and reinforced with some serious epoxy. And the drogue (smaller) parachute disappeared into oblivion.

But much of the rocket was surprisingly undamaged. The larger parachute never even unraveled and was completely fine, along with both white shock cords connecting everything together. The nose cone and electronics were in great condition as well. Unfortunately both flight computers had their batteries ripped out during this event so they lost power and stopped recording data after the first few seconds, but both are in perfect working order and only needed new batteries, an easy fix.

The conclusion

It also seems clear that the cause of the fin issue was my own flawed construction technique. Typically, with previous rockets, I’ve built the fin can (i.e. the section of the rocket consisting of the motor mount tube and the fins) outside of the larger diameter rocket airframe, and then inserted the fin can into the airframe. This allowed me to use plenty of epoxy attaching the fins to the motor mount tube at the root edge of the fin, and to build up thick epoxy fillets.

In this case, however, I inserted the motor mount tube into the airframe first, and then attached the fins “through the wall” of the airframe tube. I likely didn’t use nearly enough epoxy on the root edge of the fins when inserting them – and because of this, at least one was yanked off during flight when it experienced flutter.

The reboot

I knew what I had to do. Rebuild the entire rocket (salvaging a few parts from the original if possible, like the parachute and shock cords) and this time, build the fin can outside the airframe and use plenty of epoxy on the fins. Make sure those fins are securely attached and incredibly strong.

Which is exactly what I did, for my level 3 certification attempt #2, just three weeks later.

How did that attempt go, you ask? Well, let me go put on some coffee and I’ll tell you all about it..

Preview: L3 Fusion rocket build

As mentioned previously, I’m partnering with SBR to build the new L3 Fusion, which you can pre-order now. We decided to film the build from start to finish, in the SBR workshop studio.

the studio within a workshop, recording me building a rocket
behind the scenes

My previous high power rocket builds have been relatively slow. Don’t get me wrong – I generally don’t procrastinate, and once I get excited about a rocket project, I dive in and don’t come up for air until it’s complete.

But my techniques are far from perfectly efficient, and there’s often a substantial amount of long-curing epoxy, and then waiting for it to cure. And then… repeat. Each cure takes hours or even needs to wait overnight, and I’m doubly and triply reinforcing everything to make sure it’s sufficiently strong. Plus I’ve only built a few larger high power rockets so far and I was inevitably much slower in the beginning.

With this L3 Fusion build, however, I was able to move at a much more rapid pace. One big reason was using a fast-curing 12 minute epoxy. You just mix the resin and hardener and get to work. With a 12 minute cure, it’s amazing how quickly you can build!

me building a rocket, in workshop studio
yes, yes… this feels about right

That being said, for purposes of filming every step, this was still a marathon build session. It took a couple days of nonstop construction, even though the steps themselves are pretty simple and there was nothing that I hadn’t done before. At least, in principle. The “marathon” aspect was only because of filming and trying to make the most efficient use of our time.

We did have a few minor technical difficulties. Despite an excellent studio setup with a camera and tripod, external microphone, bright lights, and so on, there were several long stretches where we captured excellent video but the audio was completely missing. This meant we had to go back and figure out what went wrong with the microphone (troubleshooting this was much more difficult than you might think) and then re-shoot some of the steps. This turned out to have a silver lining, though, because the second time through I was substantially less inept than my initial attempts. Still inept, that is, but less so.

me building rocket with parts on workbench
enjoying the build

All in all, this was a really cool project, an awesome rocket build, and a successful video shoot. As soon as we’re done with the editing and have a final product, SBR will share the video on its YouTube channel (and I’ll share it as well, on my channel). In the meantime, if you’re looking to embark on your L3 certification, I highly recommend that you consider this L3 Fusion rocket!

L3: a story of hubris

At the beginning of 2020, before we collectively realized that COVID-19 would completely upend everything we know and love and that this year would be written off as a total disaster, I had set a couple of goals in a naive attempt to start the year off right.

picture of me looking wistfully into the sky, with large pine trees in background
overconfidence at its finest

Society might be crumbling, but looking back on the first quarter of 2020, I seem to have held up my end pretty well. My goals were to build my first high power rocket and electronics bay, and to transform my backyard shed into a workshop, primary for rockets and related projects. And I set an ambitious schedule of getting certified at not only L1 and L2 (realistically, this would already be fairly challenging) but also L3, the highest rocketry certification and something significantly more daunting, even for a seasoned pro (aka, not me). How has my progress stacked up so far, as of early April?

Well, I did build my first high power rocket, the HyperLOC 835, and I completed the e-bay. After I got my hands on some black powder, I conducted some ground testing. The rocket and e-bay can be used for both L1 and L2 certification, once society returns to some degree of normalcy and launch events are held again. I’ve studied for the written exam which is also required for L2. I have no doubt that I’ll get the L1/L2 certifications in the coming months. Piece of cake.

Plus, the workshop is nearly finished and I’d modestly deem that a spectacular success (although all of the electrical work was a joint effort with my friend Darrin, and by joint effort, I mean any success was entirely due to his expertise and generosity). The workshop bodes well for all future rocket construction, too.

But L3 certification is another story.

I started looking more closely at the NAR criteria for L3 certification. Objectively, and superficially, the criteria are similar to L1 and L2: build and successfully fly (including recovery) a rocket with a motor in a certain class (for L3, the motor must be classified as an M, N, or O). In addition, the L3 cert requires:

  • you must already be L2 certified;
  • you must have a member of the NAR L3 certification committee (“L3CC”) as your official advisor;
  • you must submit an L3 rocket design to the advisor for approval before beginning;
  • you must very thoroughly and comprehensively document every construction step along the way.

The rocket itself must also be built to certain specifications. For example:

  • each parachute event must be initiated by redundant control systems;
  • the rocket must have a safe rate of descent (20 ft/ sec is desirable);
  • you must be able to externally disarm all pyrotechnic devices on board the rocket, and so on.

However, what I didn’t realize – foolishly, in retrospect – was that it might not be a great idea to dive into the L3 rocket immediately after getting certified as an L2. Aside from the fact that the L3CC advisor might simply require additional experience before allowing you to proceed with an L3 project, more experience is clearly helpful. Rockets can be built in a large variety of sizes, and from a large number of different materials, from cardboard to fiberglass to aluminum, and more. They can be single-stage or multi-stage. They can involve different components and electronics. And there’s a lot to be said for a broad range of experience, including failures (which are plentiful, and which present excellent learning opportunities).

Let me be clear. I still fully intend to get the L3 certification as soon as possible. I’ve heard many people, whether they already have the certification or just intend to get it at some point in the future, talk about deliberately waiting years and emphasize that there’s no rush. This certification is a goal of mine, and I want to get it sooner rather than later. But at the same time, I fully appreciate the need for additional experience, something for which there is just no substitute. So I will build more and fly more, fail more and succeed more – so that when it comes time to embark on the L3 journey, I will be unquestionably ready.

I suppose it might be more accurate to say not that I was overconfident (perhaps), but rather than I was merely under-informed when initially setting the goal. I’ll go with that.