So far I was able to get any plane flying by using the AP defaults (and then Autotuning). But since this plane is a bit out of the ordinary (which is why I chose it), I was wondering if some pros could recommend settings that should probably changed right away, to ensure a smoother maiden flight.
For the XFly Eagle Twin, keep the maiden setup close to defaults rather than adding tuning changes upfront. Check CG, control directions, and failsafe on the ground first, then change one parameter at a time after the airframe is proven.
Good point, thanks. That’s the kind of input I was looking for. Re: Boomerang - the plane is popular with LOS pilots for being able to do such tricks. But I don’t think I want to try rapid rotations in FPV.
I think I will go with 1/10 reduction of differential thrust gain for a start, because you’re right, its effect should be massive with the engines so far out. Also they’re quite powerful. I didn’t dare trying full throttle indoors for fear to mess up the whole room.
Ah yes, ESC Calibration is also still on the to-do list. I tried accessing the stock ESCs via BLH but no luck, they’re really just oldschool plane ESCs. So far I always replaced stock ESCs with BLHeli ones, for telemetry etc. But since single BLH/AM32 ESCs a hard to come by these days and the stock ESCs reputation isn’t that bad, I was planning to go with stock this time. It’s been ages since I did a manual ESC calibration, so there sure is enough potential for error…
Getting closer to actually flying this, I did some thrust tests outside and discovered differential thrust (rudder) is reversed. Now before I reverse the channel in the transmitter, how can I make sure yaw stabilization isn’t reversed as well?
In general all reversals are only set on the FC itself not on the transmitter. Otherwise you will have problems if you switch between manual and auto modes.
Well, I almost never fly in manual, but I always have the mode set just in case, and somehow I always made it work correctly in all modes, just by reversing transmitter channels and servos until everything moves correctly. But I see the point - so the better place to reverse would be RCn Reversed?
Stil wondering how I can verify correct direction of yaw stabilization by differential thrust - since there is no rudder to watch for movement.
Are you sure about this? I’ve read multiple times in this forum that it doesn’t really matter when channels are reversed and that this sort of thing is a matter of personal preference. Is there something specific to this design which makes reversing in the FC preferable?
What a gloriously impractical airplane. I want one.
Not sure where you’re reading this, but motor direction and assignment must be explicitly known to the autopilot, or else any mode other than manual mode has a chance to fail due to misconfiguration. Mission Planner’s motor test feature is the best mechanism to ensure the outputs are configured correctly.
In Mission Planner under Setup > Basic Hardware > Servo Output, you can view all servo outputs without the servos or ESCs actually having to be powered. It should work if the FC is powered only via USB. Then you can (only temporarily, for testing purposes) safely increase RUDD_DT_GAIN and KFF_RDDRMIX and apply full thrust without motors are actually running; this will let you see if differential thrust is working as intended.
In FBWA, when rolling to the left, the rudder should move to the right. If the response is reversed, you’ll need to swap the left and right throttle output. If the response to stick input is reversed, you can reverse it on the transmitter channel or at the FC input (as long as the transmitter has been properly calibrated, including the center position, in this case it doesn’t matter where the signal is reversed).
You should think about whether you even want to fly with differential throttle. As @LupusTheCanine wrote, I’d be extremely careful too with that. Set it too high and you’ll be on the ground faster than you can blink.
I was thinking of the common issue with the elevator channel needing to be reversed. I agree that the motor direction needs to be known by the autopilot.
Thanks, I’ll try that testing method tomorrow evening, before hopefully finally doing the maiden the day after - weather looks promising.
Due to the expected extreme effect I’ll reduce RUDD_DT_GAIN from the default of 10 down to 3, hopefully making it barely noticable. I’d still like to have some differential thrust if safely possible, as this plane is also known for hot landings…
.
