start && end > -1) { if (start > -1) { var res = data.substring(start, end); start = res.indexOf('>') + 1; res = res.substring(start); if (res.length != 0) { eval(res); } } cursor = end + 1; } } } //]]>
Futaba HPS-H700 servos 18.07.2019 05:55:27 |
Registriert seit: 9 Jahren Beiträge: 7 |
Re: Futaba HPS-H700 servos 18.07.2019 09:33:38 |
Moderator Registriert seit: 17 Jahren Beiträge: 17.807 |
Quote:There is no benefit in performance in VBars. The reason is quite simple: the VBar's control loop is designed in a way that it runs optimally with the given frame rate which is 7 ms—142 Hz. If you try to increase the rate, it is not only useless, but counterproductive. The filters will not pass any inputs that have higher rates by purpose. We did lots of comparisons, which showed that higher rates are not making anything better, but worse. The reason is, letting higher frequencies out to the servos/rotor will only increase the current consumption of the servos dramatically. Neither the main rotor nor the servos can effectively handle these movements.
If you use narrower pulse width, you may gain a latency advantage in the range of a few %. This can safely be considered as no change at all. If pilots notice a difference, this is probably with other FBL systems, since most of them do not do much in the way of filtering. They more or less let the servo do this. In this case, it may make a difference, but not necessarily only to the positive.
Note that the radio is running with lower frame rates, so the reaction to the pilots stick inputs will not change on higher frame rates/narrower pulse width.
We don't see a need to follow such marketing hypes, if someone wants to make their servos run with VBar, they should support the standard pulse width.
On the tail, in our opinion, it's a hype as well. If you do a comparison with the same servo with 1.5 ms or 0.75 ms center pulse, you will not notice any difference. The technical gain is a maximum of 3 ms+1.5 ms = 4.5 ms latency. With 3 ms+0.75 ms = 3.75 you get a maximum latency gain of about 15 % (or 0.75 ms—thousandth of a second!). If a control loop really relies on that small phase margin, it's not usable. The big improvement on the tail was reducing the frame rate from 20 ms to 3 ms, which brought an improvement by 666 % (or down a full 17 ms), and _thats_ a difference. Also because the tail rotor can benefit from it, compared to the main rotor.
Any difference you can _hear_ on a helis main rotor is in a time domain which is a decade slower than the above values, and what you can _see_ visually would be another decade slower.
Quote:Neutral Pulse Width: Default 1520?s, optional 760?s
Re: Futaba HPS-H700 servos 19.11.2019 10:00:10 |
Registriert seit: 7 Jahren Beiträge: 12 |
Re: Futaba HPS-H700 servos 12.07.2020 12:43:30 |
Registriert seit: 7 Jahren Beiträge: 35 |