Last week I was able to test out some new belt configurations. After checking out the test stand in tests #9 and #10, it was clear the test stand was working well, but the belts were still having issues. It's apparent that the flight rotors require a lot more torque than the test rotors and there is more slack in the belts due to this.

The first attempt (Test #11) was adding a pulley on the slack side of the double sided belt. It seemed to help a little bit, so the next attempt was to try tightening the belts a little bit. This did not improve the situation. The literature suggests that having the belts too tight is as bad as having them too loose. This is because if the belts are too tight, the pitch on the belt gets stretched and they no longer mesh well with the pulleys. I don't know if that's what happened during the test, but after the test was over, the two sets of propellers were no longer aligned with each other.

Another potential issue was how well the props were balanced. I hadn't balanced them thinking that they were probably close enough and the weight of the propellers was less than the pulleys. After Test #12 I took the time to balance the propellers. Only two of the propellers needed any weights and those two didn't need very much. I'll do another write up on balancing the props later. However after balancing them and running it, there was no perceptible change. The last change made was removing the springs on the double sided belt tensioner. I was curious as to if the springs were contributing to the oscillations. Again, no visible changes in the oscillations.

So overall thoughts on testing so far.

• The frame does not flex significantly like it used to. Some of this has to do with the way it was tied down before, but some of it also has to do with the changes to the frame that have been made.
• The flat idlers are working. This are significantly lighter than the steel HTD pulleys.
• While the belts are moving around significantly horizontally, they are staying in place vertically. Those parts of the issues seem to be solved
• The relay board is working well. I was concerned with using mechanical relays with this amount of vibration, but so far they've worked.