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.

Date:  14th March 2010
Location: Doonside, NSW, Australia
Conditions:  Warm 24C, partly cloudy 10+km/h wind.
Team Members at Event: PK, GK, AK, PaulK, John K and Jordan K.

We had a fun weekend at Doonside this week launching a couple of different rockets and testing a few things. The first one was flight qualifying the sustainer for the Acceleron V booster and taking Craig's flight computer for a ride to gather flight data.

Axion IVb

The Axion IVb rocket is almost identical to its previous version with the exception that the Robinson Coupling in the lowest bottle is replaced with a tornado coupling and an insert that allows jet foaming to happen. This was designed to help prevent the base of the bottle blowing up at the coupling like has happened a number of times. For this we made a small spliced pair of bottles to replace the regular 1.25L bottle at the bottom.

This is because the normal spliced pair has a capacity of around 2.1L and we are going to be putting only 1.3 or 1.4L into the sustainer. This would have resulted in pretty much all the water settling in the lowest bottle preventing jet foaming from working properly. The small capacity bottle pair makes sure some water remains in the upper bottes. The good side-effect of using the tornado couplings is that you get much better alignment of the bottles than you do with Robinson couplings. With Robinson couplings the alignment is dependent on the geometry of the base of the bottle and how well you can drill the hole in the center. We would often get instances where the coupling would be very slightly angled to one side when tightened.

We also wanted to test the new 2L bottle splicing technique on the 1.25L bottles. We did not have any re-enforcing on these bottles, as they can handle the 120-130psi pressures without problems.

LiPo Batteries

We have mostly been using 9V batteries for our flight computers because they are easy to get and are cheap. They have a couple of disadvantages though. They are relatively heavy (~45grams) and are not rechargeable so need to be replaced, which precludes them being mounted more permanently. There are re-chargeable 9V batteries, but they cannot deliver the necessary current.

So a couple of weeks ago I ordered a whole bunch of 70mAh LiPo batteries with a 20C discharge rate. This should deliver plenty of power for a whole day worth of flights. Since servos ideally need at least 4.8V we will need to use two of them 3.7V x 2 = 7.4V. They each weigh around 2g! Since we'll be converting this to 5V we can't use a regular 7805 voltage regulator because it can't maintain line regulation below about 7.5V input. So I've ordered a number of LDO regulators (L4940V5) and I'll use those instead. All up the replacement battery should weigh only 5-6grams. It will be nice and compact and will be re-chargeable. A 40 gram saving on a smaller rocket can have a measurable effect on altitude.

I've also been investigating the best way to charge them and what chargers to use, but more on that next time.

Hard o-rings

During pressure testing of the Acceleron V booster we had a leak at one of the tornado couplings. It turned out that one of the o-rings was pushed out from the gap between the bottle and coupling. What was worrying is that it happened at around 30psi. We had seen this at least once before when launching the flour rocket recently. We put that down to the fact that the bottle neck may have shrunk in the sun due to the black coupling. This would have created a bigger gap for the o-ring to pop out. This was a different case though as everything was new and no heating was involved. It may be that that particular bottle or coupling had slightly different dimensions allowing it to happen. We swapped out the tornado coupling and turned one of the bottles around and the problem went away when we tested it to around 50psi.

During the week dad chased down and ordered a harder version of the same BS-119 o-rings which should help prevent this from happening in the future. We used the harder o-rings on the sustainer test flights this weekend without issues. We will not be replacing the o-rings in the assembled booster, but if we find any more Tornado coupling leaks during testing or on flight day, then we would replace them at that stage.

Craig's flight computer test flights

Craig ('strud' from the Forum for Australian Rocketry) has been working on a nifty little flight computer for his pyro rockets. He flew it on a couple of his rockets and we had a chance to do some test flights for him on our water rockets as well. It wasn't controlling deployment but was along for the ride to gather flight data. Here are more details on his computer: http://ausrocketry.com/forum/viewtopic.php?f=32&t=1922 and also here: http://ausrocketry.com/forum/viewtopic.php?f=16&t=1998&start=30

Launch Day Report

• We arrived about 8am at the launch site, and set up our new collapsible gazebo to provide some shade. This goes up a lot faster than the poles and tarp we used to use.
• The first flight off the pad was Paul's two stage pyro rocket since we still had a couple of motors left over from last time. We wanted to make sure that the last 2-stage flight wasn't just a fluke. The rocket went up well and the second stage lit when it was supposed to, so all in all it was a successful second 2-stage flight. I believe this our longest duration rocket flight to date (water or pyro) at 1:05 min.
• The next flight was the first flight of the new Axion IVb sustainer. It used the tornado coupling insert to generate foam. We launched it at 120psi as that is going to be the initial launch pressure for Acceleron V. We decided to drop the launch pressure to 120psi from 130psi for the first flight to give us a greater safety margin. The rocket took off slowly and the jet foaming appeared to work well on the flight.
  
  We flew the rocket with our altimeter and the original MD80 camera. On review of the altimeter data it showed a single 30 foot spike and the rest of the plot was level. We are not sure why this was the case, but may had been a corrupted calibration value? The two subsequent flights also showed a similar problem. The next day I hooked up the altimeter to the computer and did a proper reset and it appears it has come to life again. I hope this does not happen on the actual flight. *fingers crossed*
• The next flight was identical, but this time we mounted Craig's flight computer to the mid section of the rocket. In case the parachute failed I wanted to have at least a couple of bottles worth of crumple zone between the ground and computer. The rocket had a similar flight, but unfortunately his computer did not detect launch and did not record anything. With foam and lower pressure, the acceleration is very low so I am not surprised it did not detect it.
• For the next two flights we swapped to the Axion rocket and used a 15mm nozzle and fitted the launch tube to the launcher. We also removed the jet foaming insert in order to get the launch tube to go in. We did not use any foam since we wanted to get a high-G take off to hopefully trigger the flight computer.
• On both flights the flight computer detected launch and recorded good data. These flights peaked at roughly 20G. In the onboard video you could hear Craig's computer beep after the rocket reached apogee, which is what was expected. This was triggered by the onboard timer.
  
  We also benefited from these test flights since this was the first time we got some real accelerometer data. (See below)
• The last flight also had a little character mounted in its payload bay. Unrelated to rocketry this little character "Junior" has been travelling around Australia for about a year, visiting all sorts of people and they always take photos of him in different locations. It was Juniors turn to fly in a rocket this time.

Flight Data

Following is the captured accelerometer data from both flights. We noticed a large spike in the acceleration just after the launch tube phase. At the time I put this down to an error in the data.

Actual captured data from the two flights from Craig's flight computer

When we got home we ran simulations with Clifford's simulator to see if we could compare the results to what we saw. There was a problem though because the launch tube phase looked quite different. Clifford's simulator showed a large acceleration during the launch tube phase, higher than the start of the water phase averaging around 170m/s/s. The accelerometer data showed closer to 50m/s/s. The timing also did not look quite correct with the air pulse starting around 0.39s.

Clifford's simulator results

So I decided to run Paul Grosses' simulator to see if there was any difference. Paul's simulator showed the launch tube phase much closer to the accelerometer data. It showed the launch tube acceleration phase at 45m/s/s, however, it showed a large spike at the start of the air phase. I know both of the simulators use an instantaneous impulse function developed by Bruce Berggren to model the air-pulse phase and so the observed decaying air-phase is not shown. The timing though on Paul's sim looked correct and much closer to what was captured in real life.

Paul Grosse's simulator results

So it was onto Dean Wheeler's simulation to compare all three. Dean's simulator seemed to be the most accurate in modelling the observed data. What was a big surprise was that the sim even showed the big spike at the end of the launch tube phase. His simulator also showed the correct timing and the decaying air pulse.

Flight 2

Flight 3

Dean Wheeler's simulation results

It appears there may be a problem with Clifford's simulator when a launch tube is used. I compared the results for altitude prediction without the launch tube and the results were very close in the different simulators, however, when a launch tube was used Clifford's simulator appeared to give more optimistic results for altitude. We'll do more experiments next time and compare the real data to the three simulators again, taking altimeter data into account.

Flight Details