LUNAR’clips 2000        Volume 7, Number 3

It was Murphy’s Law. The very first time I launched an electronics payload, I was treated to the awful sight of my rocket and parachute going their separate ways.

I was sure when I recovered the wreckage, my brand new $150 Black Sky altimeter would be history. But I was lucky, and the soft ground at the Lucerne Dry Lake bed cushioned the impact. There was relatively minor damage to the airframe and the altimeter was fine. Amazing when you consider the inadvertent tumble recovery was from an altitude of 4200 feet and the model was a 53” long LOC Vulcanite on an I161 motor!

I was especially surprised at the failure considering I had flown the model earlier in the day on the exact same motor and had had a beautiful flight. The model had been very high when the separation occurred, so I couldn’t see clearly what happened. The shock cord was intact and still connected between the nosecone and body tube with no visible tearing. My conclusion was that the chute had been torn away…but why?

The altimeter held the key to what had happened. If you look at the altimeter traces below you’ll notice that at about 8.25 seconds there is a huge spike in the acceleration. This is the motor deployment charge firing and the altimeter accelerating away from the booster. Unfortunately for my parachute, the deployment charge fired while the model was still going at about 250 feet/sec (~ 170MPH). I’m going to guess this is fast enough to shred the ‘chute. But I still didn’t understand why my ‘chute popped before apogee. Time to look at the simulations.

I had one major goal with the launch of my Vulcanite: I wanted to achieve close to one mile in altitude. In anticipation I ran several simulations using Rocksim. (For those not familiar with it, Rocksim is a great simulator for Windows that gives accurate modeling of a model’s performance. You can try it at www.apogee.com.)

The simulation traces below show the expected performance of the Vulcanite with an I161W-10 motor. Ejection was timed to coincide with apogee at about 5000 feet. In fact, if you compare the acceleration, altitude and velocity traces from the simulation model with the actual measurements above, you’ll see very good agreement (e.g. the peak velocity at 2 seconds being 760 ft/sec in the simulation and 790 ft/sec in the actual flight). The only significant difference comes at the ejection charge firing time: the simulation says 12 seconds into the flight, but in reality it fired at 8.25 seconds. So just as I expected from the accelerometer trace, my real-world motor was to blame.

Thinking back to the assembly of the motor, I remember having some trouble getting the ejection charge seated properly. I guess it’s possible something I did caused a premature firing. In any case, I think I’ll be using a longer ejection next time I fly this model!

In the end I got very lucky since I was able to completely repair one of my favorite models. Some wood filler, CA, and new paint was all it took. And now I have a great deal more respect for the power of simulation since it proved to be extremely accurate. If you’re not already simulating your models, I highly recommend it. It may not prevent every failure, but it sure helps you avoid most.