We have gizmo moving around under control now. Unfortunately it is still too unstable to really control.
Monday we dove straight into flight testing. It didn't work at all. We pulled out the multi-meter and oscilloscope to troubleshoot. After spending a considerable amount of time trying to find what was wrong someone noticed the fuses on the motor leads were toast. We all went out to our cars to grab our spare fuses. Several cars are using the mini fuses now, but we managed to find enough 30-amp fuses. Flights were more as expected than the crash a few weeks ago. Gizmo got into a slow roll and a definite +x translation whithout command. I was able to land it quickly enough. We took video, I'll post links as soon as the video is put up. We did two flights, with similar results. I did some more test flights today. Knowing that there was a tendency to roll (we define roll as rotation around the vertical axis), I tried compensating for the roll while keeping one of gizmo's gear on the ground. I was able to control it fairly well though the response was a bit slow. That tells me I need to up the control constants a bit. I also tried full flight. It did an immediate translation but appeared to not get any worse, indicating that the response was not fast enough, but that it did zero out the rotation. It also pointed out that it is sufficiently small enough and unstable enough that I am not quick enough to really bring it under control. I ended up shutting down the motors just as I noticed that I was entering into pilot induced oscillation. All of which occurred really fast, each flight was probably 1 or 2 seconds from the time the gear left the ground. Todo items: - buy a good digital video camera. It would be useful to watch video rather than my internal memory. - Tweak the control constants - add higher level control code. We are only controlling rotational rates right now. We need to add in a PID loop for rotational position/translational velocity. Dave -- David Masten <[EMAIL PROTECTED]> _______________________________________________ ERPS-list mailing list [EMAIL PROTECTED] http://lists.erps.org/mailman/listinfo/erps-list
