The workshop build is still taking up most of my time at moment, once complete however it will give me more space to work and build.
In the spare moments afforded to me by the rain we have been having here I have continued the development of the AW609 which I began last summer.
After lots and lots of hovering and lots of less than graceful 'landings' I dialled the hover in quite nicely - the project came to a sudden stop however when I lost the prototype during transition testing.
The problem was trim. I would use the Tx to provide some trim in yaw during the hover. This is achieved by tilting one rotor forward a small amount, and one back. It works quite nicely in the hover giving you the ability to null out any spinning. The problem occurs when you transition, the yaw trim is still being commanded by the Tx but as the flight computer starts to mix yaw and roll it powers up one motor (to continue the required yaw trim as yaw in fwd flight is provided by the motors). This results is a yaw but even more problematic it also commands a roll as the motors are not fully tilted fwd yet.
In short, you cannot use trim from the Tx when using a flight controller. For a bicopter design this is tricky as any discrepancy between the two motors will end up giving you a yaw. This can happen during the build, or after a transition, as I found the 9g servos struggled with accuracy when tilting the motor nacelles back to vertical. One solution will be to change the PID values such that the flight controller itself nulls out the yaw in the hover. Another would be to do away with the individual tilting of the nacelles and go with a single translating spar that tilts both motors at once. So long as they are build accuracy the model would simply weathervane into wind.
I would like the model to be controllable in yaw so will continue with independent nacelle tilting for now.
Some simple changes I have made so far include a new hatch, new motor nacelles and a redesigned horizontal stab. The stab was originally an all flying tail but it lacked rigidity. The hatch is a huge improvement in access at the expense of some rigidity, however it was painful to try and tweak any PID values when you had to remove the wing each time. Tech update nacelles move the thrust line of the motors. I identified a need to try and have the thrust line coincident with the spar around which the nacelles rotate. This is so that the motors are not fighting the tilt servos, and to line up the hovering CG with he FWD flight CG since when hovering the CG needs to directly under the thrust line.
I would love to hear your thoughts on this project or any others, you can post your comments below.