Throttle center detent

[AirPix]

A couple of years ago I took up RC boats and found that it was very useful to modify the transmitter throttle ratchet to have a center detent. Boats use a reversible ESC and it helps to have the detent when initializing the ESC and when you need to sit still in the water. The detent was very simple to create-- open the transmitter case and with a razor blade cut away one of the teeth at the center of the geat sector that make the ratchet for the throttle.

I see now that the "50% throttle" is important with Multicopters since that is where the FC expects the throttle to be set in a hover. I'm thinking that a center throttle detent will also be useful since it gives a repeatable 50% point that you don't need to look down to see and set.

What do you think?

--Bill

 

[Electro 2]

.....since it gives a repeatable 50% point that you don't need to look down to see and set.....

Trouble is, the FC won't. Throttle mid-points are somewhat "rubber" in nature due to hysterisis built in to the FC and incedent winds "spoofing" the baro in the FC. Not a good idea. If you're flying LOS, watch it, if FPV, watch the altimeter on the OSD. These are the only sure fire no-climb/no decend indicators.

 

[AirPix]

That makes a lot of sense. The FC is an adaptable device using inputs from several sensors to arrive at the final "solution" and the exact value of the "throttle midpoint" gets prioritized to an approximation.

That explains why a careful set-up of the FC with the (Naza) Assistant Software is critical to reduce the number of "variables in the equation". And why flying a Multi is truly a hands-on (eyes-on?) experience.

--Bill

 

[deluge2]

Well this may be generally true with some FCs, but for DJI FCs with recent firmware versions, this is only strictly true when operating them in 'manual' mode. When DJI FCs are in other modes, i.e., Attitude Mode (ATTI) and GPS Attitude Mode (GPS ATTI), then the FC's automated flight control algorithms use throttle position and FC-derived altitude data to maintain altitude (altitude hold). The difference between ATTI and GPS ATTI modes is that in the latter, GPS data are used to monitor/control 'x and y' position (longitude and latitude) so that when hovering, the craft is stabilized in all 3 dimensions, despite wind direction. For some time now, DJI-supplied transmitters (basic units with their faults and limitations, to be sure) have both transmitter sticks set to self center in both axes of motion.

This change in DJI FC behavior relative to throttle stick position was introduced across existing (NAZA, WooKong) and newly-introduced (A2) FCs via firmware modifications. While reactions to this change included skepticism and hostility from some, overall the negative responses seem to have died down. Some still decry the absence of the (approximately) linear response of RPM to throttle that was the original behavior. What I like best about the new algorithm is being able to maintain altitude more reliably by simply allowing the transmitter throttle stick to self-center. This is especially helpful when filming with the craft high-up and at a large distance away, conditions where it is tough to accurately judge altitude. It also is helpful at lower altitudes, though one has to be vigilant when near the ground or obstacles. Overall, reliable altitude hold (which of course does not rely on GPS input) gives me one less thing to worry about.

Another advantage of reliable altitude hold at a midpoint that can be 'felt' (self-centering or center detent) is the related ability to reliably dial in predictable rates of ascent or descent by moving the throttle stick a small distance from the midpoint in the appropriate direction.

The altitude-hold at 50% throttle behavior of DJI FCs has led some using non-DJI transmitters (Futaba & etc.) to change joystick control configurations to provide self-centering and/or neutral notch behavior as the OP proposes. From various posts here and elsewhere parts are either included or can be purchased inexpensively from Futaba to provide a center-point notch or sprung return-to-center behavior for the throttle axis.

Steve

Trouble is, the FC won't. Throttle mid-points are somewhat "rubber" in nature due to hysterisis built in to the FC and incedent winds "spoofing" the baro in the FC. Not a good idea. If you're flying LOS, watch it, if FPV, watch the altimeter on the OSD. These are the only sure fire no-climb/no decend indicators.

 

[AirPix]

I hate to sound wishy-washy, but that makes sense, too. I'm still building my two big birds, but my "practice" quad is a Blade 180qx and I'm developing reflexes in a flight sim (AeroSim). I got the detent idea when I noticed that the Mini likes half throttle in it's Stability Mode and in the flight sim the modeled DJI/Naza quad has consistent "climb/sink behavior" at 50% throttle.

And the center detent is just a reference point-- you can always fly "a click or two" above or below if conditions warrant, and there is always the Trim knob on the Tx (though on some Tx's the throttle trim is strongest at Idle and weak above that). I may play around with the idea-- if it doesn't work out, that throttle ratchet looks easily replaceable (provided you can ge the replacement part, need to check).

Good discussion.

--Bill

 

[deluge2]

The above may be most relevant if you're planning on using DJI FCs. For these, as you stated earlier, the throttle position is really just an input to the FC. It is not directly related to motor RPM. In DJI's algorithm (non-manual mode), throttle position is more like a relative ascend/hold/descend indicator. Within the FC-imposed limits on max controlled ascent rate and max controlled descent rate (if only the latter were absolute...), the FC uses two primary inputs, altitude (from the FC altimeter) and 'pilot intent' (from the throttle stick position on the transmitter), to determine the 'desired' rate and direction of altitude change. It then modulates motor RPM on each arm to achieve the desired result. Of course this description ignores the 'other' things the FC is processing related to rates and directions of motion in the other two dimensions.

The bottom line is that on DJI FCs not flying in manual mode, the 'throttle' control functions as an altitude change control with inputs ranging from max descent (at low throttle), to hold (at center), to max ascent (at max throttle). *If* the algorithm is working properly, then you would only want to maintain the throttle stick at a fixed position off the centre point when you want to produce a fixed rate of ascent or descent for a period of time.

With DJI FCs and ESCs, this appears to 'just work' most of the time. I don't have personal experience with using DJI FCs with other ESCs, though when properly calibrated, these should work in the same way.

Steve

I hate to sound wishy-washy, but that makes sense, too. I'm still building my two big birds, but my "practice" quad is a Blade 180qx and I'm developing reflexes in a flight sim (AeroSim). I got the detent idea when I noticed that the Mini likes half throttle in it's Stability Mode and in the flight sim the modeled DJI/Naza quad has consistent "climb/sink behavior" at 50% throttle.

And the center detent is just a reference point-- you can always fly "a click or two" above or below if conditions warrant, and there is always the Trim knob on the Tx (though on some Tx's the throttle trim is strongest at Idle and weak above that). I may play around with the idea-- if it doesn't work out, that throttle ratchet looks easily replaceable (provided you can ge the replacement part, need to check).

Good discussion.

--Bill