Each flight log entry usually represents a launch or test day, and describes the events that took place.
Click on an image to view a larger image, and click the browser's BACK button to return back to the page.

(Photos: Andrew Eltobaji)

Self portrait shortly before landing.

(Photos: Chris Barnes)

Date: 12th February 2012, 7:30am - 12:30pm
Location: Doonside, NSW, Australia
Conditions:  Calm 0 - 5km/h wind, warm ~25C, blue skies
Team Members at Event: John K, Paul K, PK and GK

After the delays over Christmas and rainy weather we finally got a break on a beautiful morning with ideal launch conditions. On the way out to Doonside we drove through heavy fog, but by the time we got on location the fog had lifted. We arrived about an hour earlier than normal to help out as NSWRA was having a working bee to get the tall grass mowed down around the launch area.

Rocket preparation went well. We chose the smaller of the two parachutes we had prepared (820mm diameter). The rocket was filled with 1.3L of water and set up on the launcher. We had the extra long hose extension this time and the direct scuba tank connection with the large pressure gauge. For the higher pressures we fill the rocket straight from the tank and shut it off when it gets to the desired pressure. We took just over 2 minutes to pressurise the rocket to keep compression heating down.

The rocket also used the break wire option to trigger the backup timer just in case the high acceleration damaged the built-in 2G acceleration switch.

Here is Shadow's full build log.

Launch 1

The first launch was at 330psi (~23 bar) since the pressure chamber had only been hydro tested to 335psi. The rocket released easily and made a loud bang during the air pulse. (You just don't hear it well on the video) The rocket went straight up and you could see it coast for a long time. Right at apogee the parachute popped out and the rocket slowly drifted down and landed in the tall grass maybe 70m away from the pad.

The maximum altitude reached was 947 feet (288 m) with a total flight time of 54.2 seconds. This was our new personal best altitude.

In the onboard video you can hear the uMAD continue to beep it's armed state even after the parachute deployed. This tells us that the parachute was ejected by the backup timer rather than the uMAD. In the video you can see that when the parachute popped, the rocket was still mostly pointed upward and since the parachute is attached at the top of the rocket, the uMAD never had the chance to fire.

From the altimeter data we can roughly calculate that the peak velocity on the way up was: ~260 feet/s or 79m/s or 285km/h or 177mph

We also see from the altimeter plot that the descent rate for this parachute was: 20 feet/s or 6.1m/s

Video from the on-board camera showed that the rocket rolled approximately 270 degrees from launch to apogee. This means that the fin alignment jig worked quite well. With the sunny conditions the video turned out great.

With all the humidity in the air the slow mo video showed that the rocket produced a brief contrail about 20 meters above ground which was several meters long. This was after the end of the air pulse when very cold air was coming out of the rocket. (see picture on left) We saw this on both flights, but have never seen it on our regular rockets.

Launch 2

Seeing that we were close to the 1000' mark we decided to go for a higher pressure of 360psi (~25 bar) for the second flight. We were going to have to pressure test the rocket on the pad to see if it would hold. We were reasonably confident since tests on short lengths of the tube held at least 500psi.

Rocket prep and setup were identical to the first launch.

When Paul tried to launch the rocket he had a hard time getting it to release. I'm not sure why that was because the pressure was only an extra 10% and shouldn't have been an issue. But after a couple of tugs the rocket launched in a similar manner to the first launch. Straight up .... and straight down again. The rocket pitched over at apogee but no parachute came out. The rocket came down very fast and you could hear it whistle shortly before it hit. It landed a couple of meters from the launch pad!

Because the soaked soil was very soft it buried itself about 1 foot into the ground. The entire nosecone section and payload frame were destroyed. We had to use the old shovel recovery technique for this one.

The altimeter and camera were at the lowest point in the payload and so had the maximum protection. Sadly though, the camera didn't survive and the SD card was bent and broken. We tried to read the data off it, but it just heated up when we plugged it into the adaptor. To our surprise the batteries survived and were still powering the altimeter and the altimeter was still recording! The altimeter had a damaged LCD display from a failure a couple of years back, and this crash, just did more damage to it. The rest of the altimeter is mostly in tact. The computer was able to connect and download the data without issues though. We will most likely retire it now, or use it on other non critical flights.

The data showed 1017 feet! (310 m) . Woo hoo :) This is now our new personal best altitude though if this had been a certification flight, it wouldn't have passed due to the damage, so we are going to have to try it again. The data plot also had a very large negative spike on impact. This would have been caused by the pressure increase in the payload bay as it collapsed around the altimeter.

The peak velocity was 275 feet/s or 83.8 m/s or 302 km/h or 187mph

Here is a highlights video from the two launches:

What went wrong?

We still don't know exactly but have some ideas. We did not re-test the deploy mechanism after the first flight, as it had worked fine, and there was no reason to believe that something may have been damaged. It is possible that something went wrong when the rocket landed, but when we loaded the parachute for the second flight, the servo had moved to the locked position, the piston was also okay to lock down.

After the crash we noticed that the ground wire was cut between the batteries and the deployment mechanism. At first we thought this may have somehow happened when we re-assembled the rocket, but the uMAD was beeping on the pad which means it had power so the line wasn't cut at that point. The altimeter is connected to the batteries through a different set of wires and those were fine.

After pulling apart the ball of tangled aluminium, balsa and fiberglass that once resembled a deployment mechanism, we discovered the Servo Timer II was undamaged except a couple of bent pins. Even the RC servo still worked! Though the servo motor horn was broken it was in its deploy position meaning that it was sent a signal to move and it did. The linkage between the servo horn and the lever arm was also intact. If the servo horn was in the deploy position then the release arm must have also been pulled back. We were able to reconnect the batteries and both the STII and servo functioned as they were supposed to. 

As a result we can only conclude that there had to be a mechanical failure of the piston due to the high G forces. (around 55Gs )

We had tested the piston dozens of times on the ground and we never had a stuck piston issue. Our best guess at what might have happened is that during acceleration a portion of the parachute, or shock cord may have wedged itself in the gap between the piston and the payload bay wall and prevented it from sliding out. The piston appears to have never moved as the springs were still compressed as they are when in the stowed position.  

It is also possible that somehow the nosecone was wedged so tightly into the body tube during acceleration that the springs weren't strong enough to push the parachute and nosecone out. The springs exert about 50N of force to eject the parachute.

The camera and SD card were damaged and will need to be replaced. We have a spare camera so we will use that. The altimeter is usable, but we will most likely replace it. The uMAD was destroyed. The nozzle bulkhead flew half way down the rocket (up to the coupler) on impact. This bulkhead is just free floating anyway so we will just push it back into the tail and it should be good to go. Because the bulkhead flew forward during impact it would have reduced the shock somewhat as well. The forward bulkhead also survived without damage.

As John was digging out the nosecone debris, he hands me two bits of wire and says "Dad, here you can still use these" :)

The top end of the pressure chamber will be cut square and extended again with a coupler. We could potentially increase the length of the rocket somewhat during the repair and give the rocket a little more capacity. This would help offset the added weight of the repair coupler. We will also try to make the payload bay and ejection mechanism a little shorter and lighter.

Though this will be a bit more of an expensive fix, the offering to the rocket gods was well worth the milestone. We are looking to make the next version of the deployment system fully redundant with two isolated circuits and separate batteries.

What went right

Despite the crash there were a lot of good things that came out of the 2 flights.

• The rocket did not bend (at least what we could see) during the launch and was able to withstand the acceleration forces. This was one of our major concerns as that was a major problem with our FTC rockets.
• The launcher worked well to hold and seal the rocket.
• The rocket came off the launch tube cleanly.
• The deployment mechanism worked .... at least once :)
• The parachute attachment point worked well.
• The rocket didn't spin very much.
• The nosecone did not separate on burnout - This was another one of our main worries because it had to be fairly loose so that the springs could push it out. The parachute being fairly loose in the payload section meant that it would fly upwards after burnout also pushing against the nosecone.
• Pressure chamber held up well under the pressures.
• The rocket was able to land without damaging the fins or nozzle (on both occasions! :) )
• The parachute was a good size for this rocket, and did not tear or rip out the shroud lines.

Simulating the flights

During development we used Dean Wheeler's simulator for predicting the rocket's performance. With the best estimates for the various rocket parameters we had a predicted altitude of: 980 feet (299m) and a max speed of 79m/s for the first launch at 330psi.

The results from the first flight put the simulator about 30 feet over. The predicted peak acceleration was close to 50G at burnout.

The simulation for the second flight @360psi predicted: 1099' (335m) and a max speed of 85 m/s ( 306km/h or 190 mph). Here the simulator was 82 feet over the measured value.

Leaving all other parameters the same, we tweaked the drag coefficient on the simulator so that the predicted altitude matched the first flight's measured altitude.

Re-running the simulator with the new drag coefficient for the second launch we get a predicted altitude of 1036 feet (316m) which is close enough. So from now on we can use the new drag coefficient for the rocket and hopefully get more accurate predictions for the next round of flights.

The simulator also predicted a crash down velocity of about ~250km/h ( 155mph )!

Flight Details