To the right you see a frame grab of the launch and the altitude and velocity plots from the PerfectFlite MAWD altimeter. It hit 7292 feet (2222m) and 650 mph (291 m/s). It went out of site on launch so we didn’t witness the parachute coming out, even with the binoculars. But once it got low enough I could see that everything had deployed correctly. It landed about a half mile away in a field owned by a very unpleasant man. I won’t get into that story…

The data matches the simulator from RockSim very well. It went about 10% lower than predicted, but that makes sense. I use a few O-rings and tape and things that mar the perfect surface that RockSim calculates with. The velocity plot matches surprisingly well. The altimeter doesn’t measure velocity so you have to compute it numerically using the time and altitude. That is why the data looks noisy.

The descent velocity was about 35 mph, which is a tad faster than they normally recommend, but this rocket is all G10 fiberglass and pretty hard to damage. Everything was in great shape when I got it back.

Here’s the (short) video of the launch:

It was a nice launch season but too short! Now it’s time to build some stuff for next year…

1. Take your time loading the motor and stop and ask if you have any doubts whatsoever. There is exactly one correct way to load the motor. The first flight of Canis Minor failed because the motor CATOed because I loaded it wrong.
2. Make sure things that shouldn’t come apart can’t come apart. I lost a (broken) camera because the nose cone came loose when it shouldn’t have.
3. Drill holes in any bulkheads in the booster foreword of the motor. I destroyed Canis Minor because I didn’t do this and it lawn darted.
4. Make sure your deployment strategy takes the rocket design into account. I destroyed Canis Minor because I didn’t do this and it lawn darted.
5. Use wadding and deployment bags. I melted the streamer on the Supersonic Aura and broke a fin on landing. I also burned a parachute.
6. Turn off the electronics if you are working on your rocket on the launch pad. I inadvertently set off the ejection charge when I remove the payload bay to add wadding.

I think that’s it so far. These are mistakes I will not make twice!

Incident Details: On Saturday, August 15th, 2009 the rocket Canis Minor was launched on 2 occasions from the North Branch launch site of Tripoli Minnesota. The first flight was powered by an Aerotech J350W motor and achieved an altitude of 6554 feet above ground level. Deployment was successful and the vehicle was recovered undamaged. All systems performed nominally.

The second flight was powered with much less thrust using a single grain Cesaroni “G” motor. The purpose of the launch was to try to set the Tripoli MN “G” motor altitude record. An altimeter was onboard with the sole purpose of recording the altitude. Electronic deployment was not used. Instead the flight utilized motor ejection to deploy the parachute. A Rocket Hunter radio transmitter was also flown as payload to assist in recovery.

The vehicle launched successfully and just after apogee a puff of smoke was seen by most observers on the ground. The parachute did not deploy and the rocket was aerodynamically stable and ballistic when it impacted the ground. The impact was not observed because it occurred in a corn field and the rocket was not recovered. The Rocket Hunter signal was not present and the rocket is presumed to be entirely destroyed.

Incident Cause: While all systems appeared to have functioned normally the flight plan failed to take into account a bulkhead separating the motor assembly from the parachute bay. The design of the rocket explicitly ruled out using motor ejection for recovery. The motor ejection gasses never entered the parachute bay and thus the parachute bay was never over-pressurized and failed to deploy.

The responsibility for the failure lies in the launch design and preparation team (LDPT), which failed to take the rocket design into account when preparing the vehicle for launch. The LDPT consisted solely of Michael Koppelman and he is entirely responsible for the incident. While he is not particularly happy about totally destroying his rocket, he thinks this shit is fun and he thoroughly enjoyed himself.

Photograph of the last flight of Canis Minor courtesy of Rohn Blake.

It went 11,511 feet above ground level and reached about 1200mph. It landed just less than a mile away. It traveled about 20-30 mph horizontally, which is about right because the wind was blowing pretty good.

The plot shows the altitude from the altimeter. The main chute came out at about 600 feet. It landed at 30 ft/s (20mph). Perfect flight! Too bad I didn’t see very much of it!

I built a Performance Rocketry Little Dog Dual Deploy. I have a little video which describes the rocket on youtube. The first launch was a disaster which resulted in a cato. (cato is rocket dork slang for a CATOstrophic (sic) failure of the rocket motor.)

The 2nd launch was (mostly) a success! I was going to launch with a J motor but because of clouds I used an H motor instead, so a quarter of the total impulse. According to the sim it should have gone about 2000 feet, but it ended up going about 1500 feet.

The rocket wobbled a bit early in the burn. The streamer was supposed to pop at apogee and then the main chute at 700 feet but they both came out shortly after apogee. I think the payload pay just fell out of the booster when the nose cone was blown off. But as failures go, that was a pretty good one, as the rocket descended nice and slow and landed in view.

I was happy with the launch because I loaded 2 motors and they both worked fine. My altimeter worked, too, and fired the deployment charges. My failures were the wobble, which might have been solely because of the slow speed, and the fact that the main deployed too early.

It’s fun.

Like, you know, me. I’m an amateur scientist who enjoys launching high-power model rockets. The ATF, in their infinite wisdom, treats the motors that my rockets buddies and I use as explosives. This makes them hard to ship, hard to buy and hard to store. For example, I need to get an explosives permit from the ATF in order to by certain kinds of igniters, a device less capable of harm by far, than my gas grill.

The uninitiated should know that model rocket motors are ridiculously safe. Even if you put them on the ground and set them on fire, they just burn like a flare. Almost no one is ever harmed in the sport of high power rocketry, even people who make their own motors from raw propellent.

In terms of terrorism, these motors are just not that useful. You can buy, without a license, far more dangerous materials, including fertilizer and diesel fuel. You can buy 50 pounds of black powder at a sporting good store without a license. You can buy kerosene, which is literally rocket fuel, without a license. But I have to get a license to play with model rockets.

It’s insane.

Maybe if Obama reads my blog he can IM me from his Blackberry and I’ll convince him to make some calls. We’ll get this sorted out.

But I keep asking myself — what is so interesting about rockets? I can skip to the end of the book on this little hobby and see that I will never have the capital necessary to actually play around with the cutting edge of “amateur” rocketry (unlike Jeff Bezos, Elon Musk and John Carmack). I’ll build and launch rockets and build and launch payloads, like video cameras or whatever, and then I’ll have done it all and I won’t need to do it anymore.

But the goal isn’t necessarily to do something that has never been done. Sometimes great goals are ones that do things that many people have done. It’s the whole thing about the journey and not the destination. It was the night before my last launch and I grabbed my wife by the shoulders and shook her and said THIS IS SO FUN. I was having a blast. It’s really fun stuff and I’ve thought a bit about why.

1. The Need for Speed. Rockets go fast. In fact, I was thinking about it and I don’t know of any other thing, period, that humans deal with, besides light, that goes faster than rockets. I’m working on a rocket right now that I calculate should go about 1.2 times the speed of sound. That’s over 900 miles per hour. What else do you know that goes that fast? This is a 2 foot rocket and it can go faster than every other thing I know of excluding multi-billion dollar air and spacecraft. That’s cool.

2. Math and Physics. For me, those are good words. Any activity I do which can involve math and physics is a good thing. With rockets you calculate various aerodynamic values to determine stability. Then, given the physical parameters of the rocket and motor, you can very accurately model flight behavior and predict altitude, velocity and acceleration. With cool software like RockSim you can design rockets and virtually fly them to try to optimize for altitude, payload or speed. So far the best I have is a design for a rocket that will go 20,000 ft. It’s basically a motor with fins at that point.

3. Electronics. I’ve not taken the step of total nerdery where I build electronics from scratch. Quite the opposite, I’m pretty much a beginner tinkerer and haven’t done much more than basic soldering. With rockets you can use electronics to deploy parachutes or to fire motors while in flight. You can fly payloads like video cameras, accelerometers, iphones and GPS. I imagine a world where high power rocketry meets the high-tech servos and airfoils of remote control (RC) aircraft. Rockets are getting more sophisticated and more capable as electronic components get smaller and cheaper.

4. Pyromania. Rocket motors are cool. The big ones are like the size of a fire extinguisher. The ones I’ve used are about the size of a nice Polish sausage. But you also fabricate ejection charges, to blow off nose cones and such. This is a black powder controlled explosion inside the rocket. So Myles and I will be going in the backyard to do some ground testing. This means blowing nose cones off of rockets with electronics. That’s fun!

5. Zen and the Art of Rocket Building. With rockets, beauty is functional. Rockets should be very smooth, very symmetrical, very aerodynamic , very strong and very easy to see. Building rockets is a little like building models but the fact that the thing will fly very fast and very high and be subject to very strong forces makes it more of an engineering project. Rockets are torn apart all the time, once one gets into high power rocketry. Building rockets is a slow and detail-oriented process that is quite satisfying, both on an aesthetic level and on an intellectual one.

6. The Final Frontier. The reason that Jeff Bezos, Elon Musk and John Carmack are into rockets is because it really is the final frontier. Orbit is still the province of governments, as we speak. Can amateurs get fractions of the payloads into orbit for fractions of the cost? What would it take to get my iPhone into orbit? Can we design the necessary guidance to put a rocket into a specific orbit? Can amateurs get to the moon?

So, um, that’s why. I like to amuse myself and I find rockets very amusing. I’m going to launch some rockets and see what sort of fun stuff I can make ‘em do. Maybe I’ll be the first person to get an iPhone into orbit. Or maybe I’ll just have fun trying.

Throws Things At Sky

massdistraction caught me being a law-abiding pyromaniac.

I got out and launched some more rockets. Although it doesn’t exactly compare apples to apples, you can see on the image to the right that things either went very wrong with my initial iPhone launch or the GPS data I’m getting is screwed up.

The black line is a flight with the red iPhone rocket but instead of an iPhone I flew a standard altimeter (a PerfectFlite MAWD). The red line is the prediction from RockSim for that flight. You can see it matches almost exactly.

The green line is the plot of the GPS data from the iPhone and the yellow line is the prediction for that flight. You can see it doesn’t match at all.

One would be tempted to think that the time stamps on the GPS data is just stretched out or something, because it would look a lot more normal if you compressed it so it had a faster climb at the beginning and a descent rate like the other flight. But I just checked again and on the video it is a 74 second flight. That is 20 seconds longer than it should have been.

Thus, the only theory that makes sense to me is that the motor burned slow and long. From the video I tried to determine the velocity and acceleration of the launch and it matches the GPS data within uncertainties. So I think the GPS data is good and the problem was with the motor.

As a rocketry dork aside, I was using the Copperhead igniters when I launched the iPhone and I used the FirstFire igniters the 2nd time. I think the Copperheads did not work correctly or I used them incorrectly, because one other flight seemed to have a weird start with them.

I tried to fly both the iPhone and the PerfectFlite altimeter at the same time but I couldn’t get them both to fit in the payload bay. That is the next test I need to to.

Finally, here is video of the 3 launches of my iPhone. The first one is the failed launch due to a motor failure. The 2nd is the one where I got all the data that I’ve posted. The 3rd one I got almost no data but the motor performed better as the flight seemed much more normal.

Sorry Elton! Don’t sue me, but I couldn’t think of a more appropriate song.

iPhone Rocket Launch and Interview (from MobileOrchard.com) from Mobile Orchard on Vimeo.

I’ve been meaning to put my video stuff together but have had some technical difficulties. I’ll get that finished up soon!

If anyone from Apple should stumble upon this: I want to be Apple’s Official Rocketeer! Make it happen, Steve.

I don’t know if I will, but if I go all the way and get the Level 3 Certification I can buy motors with 14000 N-s of thrust!

People tend to fly bigger and bigger rockets on the bigger motors. This is obvious, in a sense, because these rockets can’t be flown on smaller motors. For me, though, once my rockets are big enough to fly interesting payloads, such as iPhones and digital video cameras, I’ll be more interested in speed and altitude than bigger rockets. I don’t have the budget or the engineering skill to make really big rockets. I do have the technical chops to do some cool stuff with payloads though.

This is fun!

This is kind of a cool plot using only the GPS data:

Here is a plot of the first few seconds of the launch from the accelerometer:

You can see that the data peaks at 3g when I was expecting 12g. The acceleration heads back up from negative territory quicker that predicted as well. The prediction is from RASAero.

Here’s the whole flight from the accelerometer:

The red line at 1 at t < 0 the phone is sitting upright in the rocket before launch. Most of the acceleration is in the Y axis at launch but some is in Z as the rocket leans over into the wind. The ejection charge, in theory, goes off at 15 seconds or so but at 22 seconds the accelerometer goes nuts and by 27 seconds the iphone is now inverted, hanging upside down from the parachute. The data dies for some unknown reason when it hits the ground at t=74s.

I need to look for rotation in the X and Z axes with Fourier or something.

It’s taken me awhile to understand the data I got from the first launch. Using the video, the GPS data, the accelerometer data and my own recollections, I’ve put together a rough timeline of events. I’ve notated it on the plot of the altitude. A couple of weird things happened.

First of all, with the engine I used, it should have gone to 1200 feet or so. It actually went 670 feet. This was an Aerotech G80-13 engine. Every calculation I ran using various software packages for this purpose had the altitude between 1200-1500 feet. So why did it only go to 55% of the predicted height? I noticed it says in the Aerotech instructions that if the igniter doesn’t penetrate all the way up the tube you can get a reduced thrust situation. That is apparently what happened.

But why did it go up for so long then? The 13 second delay wasn’t nearly enough, apogee wasn’t until almost 30 seconds into the flight. I’ll note that I don’t have error bars on my altitude data because I didn’t log it, which was stupid. Is it possible that GPS is under-reporting the altitude? The calculated velocity is quite slow from this plot — about 11 m/s or about 25 mph. Did the motor burn with less thrust but longer?

As I watched it, you could see the rocket turn over at apogee before the chute came out. So I think the ejection charge went off at 15 seconds or so but the payload bay didn’t come all the way out. Things get really turbulent at 22 seconds and when the rocket tips over I think the payload bay finally comes loose. On the video you can hear me say that I see the chute at 31 seconds. I’m not sure if it was fully deployed or if that is the anomaly just before 40 seconds.

The descent looks like a constant, terminal velocity of 3.8 m/s (8.5 mph).

I’ll show you the accelerometer data next!