SpaceX Starship SN-8 test launch and fiery crash landing

On December 9, SpaceX conducted a test of its Starship rocket, and it was spectacular.

Starship SN-8 silver metallic rocket standing vertically against starry blue night sky
Starship SN-8

The rocket was called SN-8 (which just stands for Serial Number 8), following the naming pattern for each new iteration of the rocket. Elon Musk originally unveiled the idea for the Starship rocket last fall, and the prototypes SN-5 and SN-6 flew about 500 feet before falling back down. This test of a more complete looking Starship went up 12 kilometers, the vehicle’s first high altitude test.

Many people in the rocketry community watched this live. Below is a great condensed/ time lapse video showing both the launch and the landing, in case you missed it:

Basically, SN-8 had a successful launch and flew vertically for 5 minutes, then began falling back to earth. After cutting its engines, it fell horizontally – the “belly flop” maneuver – which maximizes surface area to help slow its descent.

As a side note, there are a lot of principles in rocketry that are the same whether you’re building and flying a very small model rocket or a colossal commercial rocket, and one of them is drag and aerodynamics. Rockets are sleek and meant to minimize drag and air resistance when they’re moving vertically (or in whatever direction they are pointed), but they are really inefficient and have enormous drag if moving at an angle or horizontally. You’d be surprised how slowly even a large, heavy rocket falls back to the ground without any parachute when it’s falling sideways, and often in multiple (connected) separated pieces.

Starship SN-8 metallic rocket horizontal, falling against light blue sky
image credit: NASA

Anyway, back to SN-8: the belly flop was successful in slowing it down somewhat, and then its engines turned back on to turn the rocket again for a vertical landing. Unfortunately, it was still descending a bit too quickly when it hit the launch pad (perfectly on target) and it exploded in a fireball. But overall, this was an unbelievable achievement.

SpaceX is continuing to innovate and make things that were just recently science fiction into a reality.

My own progress in rocketry may be impressive, but it’s not quite at that level yet. I have some catching up to do!

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..

Going viral

I’ve periodically uploaded videos of some of my rocket launches during the past year (with more to come soon, of course). Generally, my YouTube videos don’t get a ton of views. Most of them have maybe 50 or 60 views; some of the more interesting ones have about 600-700. But one video seems to have really taken off – no pun intended.

What’s fascinating to me about this is: why? This is just a twelve second video clip of a rocket launch. It’s the Darkstar Extreme rocket that I built earlier this year, and this particular flight is on a K-535 motor, a common and standard workhorse motor. This video is not very different from several others that I’ve uploaded recently. Yet suddenly and without warning, the views started to dramatically increase: as of when I’m writing this, it’s topped 94,000 views.

As a nice side effect, it’s caused my YouTube channel to gain a bunch of new subscribers. Some sizeable fraction of people who casually see this clip want to subscribe – my total number of subscribers has risen from about 30 to over 170 in the past week or so. This is awesome, from my perspective.

I’m just not sure what accounts for this sudden interest. YouTube provides some analytics and it looks like most traffic (82 percent) is coming from YouTube Shorts, which is something new YouTube rolled out: a vertical video format that’s basically meant to compete with TikTok.

YouTube Shorts logo
YouTube Shorts logo

Another 13 percent of viewers are finding this through their suggested videos. Very few people are finding the video by using specific search terms (e.g. rocket launch).

But it’s still mysterious: why this particular video when I have several similar ones? Why now?

If anyone has any suggested explanations, I’d definitely be interested, since I’m still relatively new to this and figuring out how it all works!

Rocket flight data

Just to follow up on my last post, I wanted to provide some additional information and the actual flight data, and briefly explain what this all means, especially for all those folks reading this who are not familiar with anything related to rockets or flight computers. And for anyone who has significant experience flying rockets, you may find the below information interesting as well, without any explanation!

As a starting point: a flight computer is basically a very small circuit board that you put inside your rocket, and it has a bunch of neat built-in gadgets to measure exactly how high the rocket went, and how fast, and what interesting events happened when. I’ll explain more below.

This is the relevant flight data for the flight I mentioned in my last post, which went over one mile high:

flight data
flight data

So what does all of this mean?

First of all, it means that the rocket flew to a maximum height of about 7,579 ft – you can see this in “maximum height.” This measurement is made by a barometer taking air pressure readings in the flight computer, starting at ground level on the launch pad, and then many times while it’s in flight. There’s also a GPS chip on this flight computer and you can see it also independently measures the height using GPS, but I’m just going to assume the lower value is more likely correct.

The flight computer also records the maximum speed, which in this case was 904 feet per second (fps), which is equivalent to Mach 0.8, or a little bit slower than the speed of sound.

The total flight time was 145 seconds (just over two minutes), and there’s a further breakdown of how long the rocket spent going up and then coming back down.

The graph is even more intuitive:

graph of flight data
flight data graph

This reflects the same data described above. The black line is the easiest to understand: it represents the rocket’s actual height over time. As is generally the case (unless you experience a catastrophic failure), the rocket zooms off the launch pad extremely rapidly and hits a maximum height early (here, just over 7,500 ft, as you can see from the black units to the left side), and then after parachutes deploy, it descends more slowly.

The red line is speed (extremely high at first and then plummets quickly), and the orange or gold line is acceleration. Both of these units are off to the right side of the graph.

It’s definitely fun to build and fly a rocket, but with modern flight computers and the ability to record all kinds of really precise data, you can really geek out on this stuff. How high can I fly? How fast can my rocket go? Is it descending at the right speed, or do I need a bigger (or smaller) parachute next time? This can really help refine your building and flying skills through a trial and error process, because you have access to reliable data. And needless to say, this can also help you find your rocket if you lose it because it lands really far away out of sight. In that situation, you’ll find the GPS coordinates onboard to be incredibly useful!

NAR Rocket Science Achievement Award

The National Association of Rocketry (“NAR”) has established a “Rocket Science Achievement Award” program, which currently has three categories of awards:

  1. Mile Marker
  2. Faster Than Sound, and
  3. Data Downlink.
rocket launching with cloud of smoke underneath
Darkstar Extreme on Dark Matter “sparky” motor

The awards are pretty straightforward: to achieve Mile Marker, you need to fly a rocket to at least 1 mile (5,280 ft), and you can get additional awards for 2 or 3 miles, or as many as you’d like, in one-mile increments. To achieve Faster Than Sound, you just have to fly a rocket at a speed that is Mach 1.0 or higher. And the Data Downlink award involves real-time telemetry for data beyond just basic altitude and acceleration.

For any of these awards, you have to have documentation of the flight data, including a copy of the data file from a commercial flight computer. If you submit this documentation and it’s accepted, you’ll be awarded a high quality printed certificate and your name will be added to the NAR website, which is pretty cool.

I recently achieved the Mile Marker award when I flew my Darkstar Extreme rocket to 7,579 ft AGL. I plan on even higher flights in the future, of course, and I’d like to try to achieve an award in each of the three categories that NAR established. The data downlink one should be the most interesting and will require a bit of creativity.

In case you’re interested, the award page is here!