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.

Well it has been a few weeks since the last update. Currently we are on a hold in launches while we are on holiday and also do further rocket developments. We will start launching again next week.

Since the last update the following events have taken place.

Events

• We bought a new 12V SLA (Sealed Lead Acid) battery for the compressor. This one has an 18Ah rating, the old one only had 7.2Ah. This should take care of running out of power during the later part of the launch day. We will also bring along the old battery as backup just in case. With higher volume rockets and higher pressures the rockets take longer to pressurise.
• We bought a new compressor. This compressor is pretty cheap (AUD$20 at Bunnings Hardware Store), but says it can go up to 250psi. We haven't tested it yet to this pressure but we will get there.
• We bought a new digital scale (AUD$29) for accurately measuring the weight of rockets in order to determine the optimum fill volume.
• Although we are spending money on these items, they are multi-purpose and the batteries are great for camping.
• A lot of consideration has been given to the problem of reliable parachute deployment. As the rockets get bigger and carry payloads such as cameras and altimeters it is important to have a reliable deployment if we want to reuse the rockets and payloads. The nosecone-off-at-apogee approach has been relatively successful, but the rate of failures for us, either too early or not at all is too high. So we decided to go with a horizontal deployment system.
  
  The horizontal deployment system shoots the parachute out of the side of the payload section, allowing the nosecone to permanently be fixed in place.
  
  While the ejection mechanism is relatively simple to construct the detection of when it should release and having enough force to release the pin holding the door on the chute is a little tricky.
  
  We wanted to stay away from air-flaps as we see these as causing unnecessary drag as well as the potential for asymmetric airflow around the rocket.
  
  Chemical reaction based release mechanisms have also been used often by others, but we wanted something a little less messy especially with kids around.
  
  From what we hear balloon in the nosecone type release mechanism also has a number of limitations.
  
  Fixed timer based systems can work quite well, but parachutes are likely to deploy at less than ideal altitude and at potentially high speed.
  
  So we have decided to design and build a lightweight flight computer (FC) and some sensors that will detect launch and apogee.
  
  Some reasons for going down this path:
  a) Repeatability of function;
  b) Configurable;
  c) The parachute will not deploy until computer is operating, can turn rocket upside down while working with it on the launch pad;
  d) Quick turn around time for the next launch;
  e) An excuse to do more electronics.
  
  Here is a typical flight profile of what the flight computer will do:
  
  It will detect a launch and start an inhibit delay. This delay will disable the apogee sensor (due to its sensitivity) for a fixed period of time, this will ensure the parachute is not accidentally deployed during the boost phase, the air-pulse phase and the immediate deceleration due to air resistance just after the air-pulse. During this time the rocket forces can be quite violent and can interfere with a sensitive sensor.
  After this delay, the FC starts listening to the apogee sensor, and any trigger from it will cause the parachute to deploy.
  After a further delay, if the computer did not get an apogee sensor reading (as we can see that that could happen) The computer will trigger the parachute anyway. This is a redundancy feature to ensure that the rocket always attempts to deploy the parachute.
  
  After 3 minutes the FC will start emitting a loud noise to assist in rocket recovery should it fall in bushes where it is hard to locate.
  
  The FC will also measure the time between launch detect and the apogee sensor firing. Upon landing, the FC will beep out the time in a similar manner the some altimeters report altitude. These values can then be used on the subsequent flights and averaged by the FC to assist in adjusting the inhibit and safety delays.
  
  The computer timing will be adjustable on the launch pad via a simple push button interface.
  
  Some functions that the FC will also be able to support in the future are:
  a) Touch down sensor to measure total flight time,
  b) Initiate other events such as cameras or drogue chutes, wings etc.
  c) It may emit IR signals on launch with a particular delay to trigger still camera's on the ground.
  
  When we started designing the FC we were going to build it from a few discrete components, perhaps a couple of 555 timers or a 556, but as we though about other functionality, the component count grew, and so it was a natural progression to use a microcontroller. We have chosen the PIC16F628A as the micro of choice. This is a very capable processor in an 18 pin package that requires minium external components, and can handle all the functionality we need and then some all for about AUD$3.
• We bought a PIC programmer from Modtronix Engineering for ~AUD$70 and set it up on the computer. They had very good service and quick delivery.
• The pull pin will be activated by a tiny motor from a toy. These motors run from 1.5 - 3V. These can also be found in mobile phone vibration motors. We decided to use this option because of the low voltage requirements when compared to direct solenoid driving using high voltages and currents or capacitive assist discharge.  The tiny motor when geared can generate quite a bit of force and a lot more travel than a solenoid. Overall the mechanism is around half the weight of a solenoid solution.
• The battery is a lightweight 6V alkaline battery.
• We are building a new 4L rocket with a Robinson coupling that will utilise the flight computer and carry a video camera payload.
• When the FC is fully developed and tested I will provide full details as well as  the .ASM files for the PIC.
  
  
  In the next update I will discuss the apogee sensor and progress on the FC.

Location: Denzil Joyce Oval. (launch site #4)
Where exactly is that? Click the above link to see a Google Earth place mark. What is Google Earth?

Conditions: Mostly sunny, light breeze, 24 degrees C.

Rockets:

Team Members: PK, GK, Paul K, John K and Jordan K.

Number of launches: 14

Today was a great day for flying with very good conditions. It had been about a month since we last launched (we were on holidays), so we wanted to try a few new things before the big rocket and flight computer are finished.

Flight Day Events

• First off the launch pad was a new rocket called D.Y. (named after the suburb where we are launching) This new rocket was built to test a number of design ideas including a new strut design for a ring fin, re-enforcing bands around the widest part of the bottle to allow for higher pressures, a weighted nosecone and lastly a new parachute obtained from surplus flares for illuminating battle fields. The rocket performed very well. It was our first single 2.25L bottle rocket.
• The weighted nosecones worked well in all instances (except one) and actually often deployed on the way up near apogee, with the parachute deploying, but the nose cone kept flying considerably higher than the rocket.
• "OO" flew 5 missions today with very good successes. This is a good rocket with consistent performance.
• We also probably achieved our own personal altitude record. On "Clifford's" second flight, the parachute failed to open, so we were able to measure the total ballistic flight time of the rocket. On video replay this was 9.92 seconds. Using Clifford Heath's simulator and the rocket's parameters the best fit flight time gives an apogee between 105 and 120 meters (350 - 390feet). The rocket was named after one of the kids favourite toys.
• 
  High Altitude Clifford rocket
  
   
• Brotanek II hadn't flown for a while and had its parachute and the foam in the nosecone compressed for about a month. There wasn't enough spring in foam for the nose cone to easily separate near apogee and the air pressure was enough to prevent the nosecone from separating. The rocket had a good flight but sustained considerable damage on impact.
• We had the new compressor going for quite a few launches, but with the higher rocket capacities and higher pressures the compressor runs much longer. I think we over cooked it towards the latter part of the day. At the start of the day we had no problem going up to 140psi although at that pressure the amount of air going into the rocket had slowed to a trickle that achieving much higher pressures would have taken a considerable amount of time.  The rate of launching rockets didn't give enough opportunity for the compressor to cool down sufficiently.
  
  At the end of the day the compressor would get up to about 90psi and then an awful stalling noise would come from it. We quickly shut it down. Luckily the pressure lines  going to the rockets have a quick release mechanism with a return valve so we were able to swap the line under pressure to the other compressor and finish filling the rocket.
  
  On the next launch attempt we filled with the old compressor and let the new one cool down. On the following launch we used the new compressor again and again it made that awful noise so we swapped the line again to the old compressor but the compressor's return valve looked like it failed and water started frothing from somewhere in the compressor. You can imagine we weren't terribly impressed and may have used the odd 4 letter word. That was 2 compressors basically stuffed in one day. Luckily they are fairly cheap to replace.
  
  We will use a SCUBA tank with a pressure regulator on the next launch day, before we can figure out a good way to fill the rockets. The tank should allow us to fill the rockets quicker and to higher pressures. A tank refill should last for a few days of launching and we have a convenient access to refill the tanks so it should not be too bad. The only reservation I have is that it is big and bulky and cumbersome to carry around.
• The new battery worked well and lasted the full two hours of launches without trouble.

Flight Record

Design and Development

• The first prototype of the flight computer (V1.1) was built and tested (not flight tested yet). This prototype does not have an apogee sensor yet but uses the launch detect sensor and a simple time based deployment. This was basically used to understand the PIC development environment and debugging process. As a result of the prototype it was discovered that while running from a 6V battery pack made from 4 x AA batteries there was enough current to run the PIC and the chute deploy motor, however, running from the tiny 6V battery that is rated at 150mAh, the noise ripple from the motor (even with 2200uF capacitor across the rails) was far too big, and was causing the PIC to reset itself. We measured the noise ripple under load on a CRO and it was about 0.5V. A quick test to see if doubling the capacitance would help, met with the same result.

  The motor draws about 75mA unloaded. We run it only for 250ms which is enough to activate the release mechanism. The PIC itself draws about 8mA with 1 LED on.

  The design was modified to run the motor from a separate power supply. This might sound like going a little overboard, but the second battery (identical to the one powering the PIC) is quite light and the circuit only includes an extra opto-coupler. Using this second battery also allowed us to remove the big capacitor which is the same size as the battery. This will allow us to run other motors or actuators from a different voltage to the PIC in the future.

• The idea behind the weighted nosecone was to add much of the rocket nose weight to the detachable nosecone. The idea was to design the nosecone's shape to have less drag and fall faster than the rocket. The weight was to help dislodge the nose at or near apogee. The theory here isn't based on ideal calculations for drag, but is based on experience as to what happens to the rocket at lower air speeds near apogee and how wind forces buffet the rocket.  The weight was provided by winding a heavy copper electrical wire around the nose cone. The total weight of the nosecone was around 40 grams.
• Another new design idea was to reduce the diameter of the base part of the bottle (where the nosecone fits) in order to provide a looser fit for a narrower  nosecone. This allows the nosecone to be made from the same diameter bottle. Using a smaller diameter nosecone reduces the drag on the rocket. From experience we have found that the nose cone has to overlap the bottle somewhat otherwise it falls off too easily.
  
  To reduce the diameter of the bottle simply put a cap on the bottle and submerge the base about 3 cm in boiling water. That part of the bottle will shrink quite quickly. As the air heats up inside the bottle it provides a little pressure that helps keep the base forming a nice shape. You may have to practice this a little but it is easy. As always be careful with hot water ... yadda yadda.
  
  We used this technique with the Clifford rocket. At some stage I want to work out at least mathematically if the reduced drag on the rocket is worth the slightly reduced volume.
• We are always looking for the cheapest and lightest way to construct the rockets. The ring fin is supported by 9 bamboo skewers held at each end by a single piece of tape. This design provides a very light structure (~40 grams) that is very robust. This design was used in two rockets and flight tested on 6 launches.

  One rocket hit hard from over 100 meters without a parachute causing no damage to the fin structure. We will probably use this design in the future for small rockets.
  Cost of fin: ~$0.10  (tape and skewers)

• We obtained 6 parachutes from old flares. These have a reasonable size hole in the parachute which makes it fall faster, but still very acceptable. These parachutes are well made with strong lines. We may sew the holes closed if we want to set a flight time record, but for now they are fine the way they are.
• The re-enforcement bands we tested are just made from recycled plastic box ties (see picture). I tried to weld a band together in the workshop by squeezing  the overlapping ends of the band between two pieces of metal held in a vice and heating them with a blow torch (The metal not the ends). This was unsuccessful as it was too hard to control the temperature, and it was either too much or not enough.
  
  Using ten staples from an ordinary stapler instead worked just fine and gave a really strong bond.