| Date: 13th May 
                                    2007 Location: Workshop
 Conditions: Pleasant 
									since it was indoors.
 Rockets:
									(click the name for rocket details)
 
 
										
											| Name | Capacity | Notes |  
											| Acceleron III | 24.75 
                                            L | A new rocket expanding on 
                                            Acceleron II's capacity. It is also our first 
                                            two stage rocket booster. |  Team Members at 
									Event: 
									GK and PK DevelopmentNow that Acceleron (I and II) has 
                                    successfully flown 16 missions we are ready 
                                    to start adapting it to its intended purpose 
                                    - adding a staging mechanism so that it will 
                                    be able to release a sustainer to reach 
                                    higher altitudes. Booster UpgradeWe disassembled Acceleron II this week 
                                    and extended each of the segments by another 
                                    2L bottle. This makes the total capacity of 
                                    the booster around 25L. 
                                     As always we pressure tested each 
                                    assembled segment to make sure all the new 
                                    joints were sealing well. The first segment 
                                    held just fine, but when we were filling the 
                                    second segment, the segment let go at one of 
                                    the couplings at around 40psi. There was a 
                                    pretty loud boom in the workshop but luckily 
                                    only the nosecone got a little crushed. 
                                     We've seen this once before, and we had 
                                    kind of suspected it because we only used 
                                    normal bottle caps, instead of the longer 
                                    thread ones. Due to the coupling and washer 
                                    there just isn't enough thread left with a 
                                    normal cap. We will go back and replace the 
                                    caps for the new extensions so that does not 
                                    happen again. The dummy payload will be 
                                    replaced by a new payload pod that will 
                                    house the flight computer, camera, parachute 
                                    and staging mechanism actuator. The staging 
                                    release is mounted just above the pod. The 
                                    pod will also be used to support the 
                                    sustainer and protect the payload contents 
                                    from the spray during separation. We have 
                                    also replaced the central pipe with a longer 
                                    one to account for the extra bottle and the 
                                    new pod being attached. Staging MechanismAfter much consideration we decided that 
                                    the staging mechanism will be based on a 
                                    Gardena release mechanism rather than a 
                                    crushing sleeve. 
                                     The reasons for going with the Gardena 
                                    mechanism are: 
                                      The sustainer can be released exactly 
                                      when needed, rather than when the pressure 
                                      drops inside the main booster as with a 
                                      crushing sleeve.All our existing nozzles will fit the 
                                      sustainer. We know that the Gardena mechanism can 
                                      hold relatively high pressures up to at 
                                      least 220+ psi. We want to be able to 
                                      retain the design once we start 
                                      reinforcing the bottles. The crushing sleeve method doesn't 
                                      lend itself very well to the construction 
                                      of CD nozzles, which we plan to revisit 
                                      especially with the sustainer. We have manufactured most of the air 
                                      supply components of the staging mechanism 
                                      that allows pressurised air to go from one 
                                      of the booster segments into the 
                                      sustainer. The air hose fitting in the 
                                      bottle cap looks a little complicated, but 
                                      it includes a non-return valve, as well as 
                                      allowing the air hose to swivel so it 
                                      doesn't twist when we need to take the cap 
                                      off to fill the segment with water. We 
                                      removed the spring from the Gardena 
                                      mechanism and we will add a small latch 
                                      that will keep the release closed until 
                                      the flight computer determines when the 
                                      best time is to release it. The release 
                                      will be actuated by a small RC servo 
                                      motor. The force to pull back the Gardena 
                                      attachment will be provided by a number of 
                                      rubber bands.  
                                       We will publish the full technical 
                                      drawings once the system is completed and 
                                      we know it works. 
                                         |