Tridge keeps the special builds for tiltrotor here. Welcome to the fun!
Tridge keeps the special builds for tiltrotor here. Welcome to the fun!
3.6 has tiltrotor support, but there is limited development atm from what I understand as quadplanes are presently the priority. I’d expect that to get better in about 2months time.
My initial test flight of the APM Plane TiltRotor firmware had some issues when transitioning from QStabilize mode to Manual. The idea was to trim it out in Manual mode and then switch to FBWA with TRIM_AUTO enabled. When switching to Manual, the CL-84 immediately lost tricopter stabilization and had no forward speed into the wind. Although the crash looked catastrophic, the CL-84 landed in some tall weeds and only broke a piece of the horizontal stabilizer and elevator. It was repaired easily.
Once repaired, I did some ground testing with the props off. It seems that when changing modes from QStabilize to Manual, the main motors instantly stop compensating for hover when the wing starts downward. Also, when in Manual mode and transitioning to QStabilize mode, this change appears to immediately start compensating for hover when the wing starts upward…which is good. I think we need the copter stabilization during the transition from QStabilize mode to Manual mode. Is it possible that this exists if switching from QStabilize mode to FBWA mode?
Greg, I’m sorry to hear about your crash, glad it was an easy repair. That’s a scary moment though. Remember our recent discussion about why Manual mode is needed as the only way to positively shut down the quad motors? The danger is that you can never switch to Manual mode unless you are on the ground or safely in forward flight. This is a tough habit to relearn for long time Plane pilot’s who instinctively switch to Manual when something goes wrong, or for Copter pilots who used Stabilize as their “manual” mode. It’s a switch that needs to be quickly accessible in case of an unexpected RTL on the ground, but not very accessible when flying.
It sounds more confusing than it needs to be. Perhaps the legacy operation of Plane is getting in the way and some rethink is needed for QuadPlane, TiltRotor, and TiltWing.
Ignoring the Manual mode issue for the moment, I suspect that a similar result would have happened if switching from QStabilize to FBWA mode. This is something I can test on the ground but recall that my previous ground testing showed a bug in the code where the motor control “sputtered” when changing from QStabilize to FBWA. The workaround was to go from QStabilize to Manual mode.
If the QStabilize to FBWA mode test shows that the main motors instantly stop compensating for hover when the wing starts downward, then the current solution will not work for TiltWing or TiltRotor. We need operation more like the FireFLY6 transition.
i do not what correct function for tilt of motors.
i found define in arduplane 3.6 source code 41 - tilt motor servo and setup it for rc10-function
without arming_require = 1 vertical motors not starts, i do not use safety sw, board safety disabled by params
Tridge’s code uses channel 12 (AUX 4) or RC12_FUNCTION = 41, unless you have changed it. If the Flight Mode is a Q-mode, like QStabilize, you need to arm the plane with right yaw stick. If you are in a Plane mode like FBWA or Manual, the plane arms itself. At least that is how I remember it from work.
Switching to FBW is supposed to maintain altitude using the Qmotors. It will not stop. If it sputters there must be something wrong.
I did some more ground testing and switching from QStabilize mode to FBWA mode exhibited the same issue as when switching to Manual mode. However, switching from QStabilize mode to FBWA mode also exhibited the motor sputter issue.
I also tried changing Q_ASSIST_SPEED from 0 to 10 and it had no effect. The quadcopter compensation ends on the start of the transition from QStabilize mode to FBWA or Manual modes. On the return transition to QStabilize, the quadcopter compensation appears to start immediately. This caused my CL-84 to drop like a brick because it has high wing loading.
At this point, I think I will call it quits on the CL-84 testing and put back the stock controller as it flew quite well as a sport tiltwing. This will free up a complete Pixhawk system for another project. I am thinking about a second QuadPlane project that can lift more payload than my Bix3. The Bix3 is a good experiment plane to get comfortable with QuadPlane features but I would like match my FireFLY payload capacity of 2.5lbs which is also the same capacity of the ridiculously expensive, poorly designed Precision Hawk Lancaster. Perhaps the Volantex Ranger EX is a good candidate with Skywalker size.
Greg, pity that it did not work with the tiltwing.
We have just build a tricopter tiltrotor plane. First flight as copter was succesfull with tricopterfirmware (3.4 rc1, see video)
We still need to reduce take-off-weight and enhance the pid values for roll before installing tridges code.
Rolf, that looks really good!
3 motors and 3 ESC’s are much better than 5, I say.
Best of luck!
Nicely done. Does the rear motor have a servo for yaw?
The rear motor servo is essential to compensate the torque around the yaw axis
My CL-84 worked fine with Copter v3.3.3. If you try Tridge’s beta code for tiltrotor next, it will be interesting to see if you observe similar issues.
I bring back this dicussion because I don’t really get how the Q_parameters work and how to assign the outputs
I have a project of building a quadplane with a servo to control the angle of the 4 motors.
Here is the params I started to configure but that’s just the beginning and nothing is pluged on my Pixhawk except the RCIN and some servos to see how the outputs are answering
15-09 Param list beginning Quadplane bis.param (12.6 KB)
The Parameters I Changed are :
_ Q_ENABLE = 1
_ Q_TILT_MASK = 15 (1111) in binary (because my 4 motors are tilting)
_ BRD_PWM_COUNT = 6 (to have the AUX5 ;))
_ V_TAIL_OUTPUT = 3 (Because I have a V-tail)
For the RC _FUNCTION I chenged them so I get this configuration :
MAIN5: Front right motor, CCW : RC5_FUNCTION–> 33
MAIN6: Back left motor, CCW : RC6_FUNCTION–> 34
MAIN7: Front left motor, CW : RC7_FUNCTION–> 35
MAIN8: Back right motor, CW : RC8_FUNCTION–> 36
AUX4: Servo for motor tilt angle : RC12_FUNCTION–> 41
AUX5: Servo for motor tilt angle reversed : RC13_FUNCTION–> 41, RC13_REV–> -1
I’d like to set my own PWM for the tilting angle and also have a transient mode with a 15°degree angle for exemple.
I don’t understand exactly how the “tiltrotor_slew(float newtilt);” is working in quadplane.
Can I create a QHover15deg mode that could be the same as QHover but with tilting motor angle at 15 degree ?
Hope you guys can help me
@GregCovey I’ve been feeling guilty about not following up on this thread. I thought I would have more time before the OBC started, but I’m leaving for OBC in the morning and still haven’t had time to work on tiltrotor some more. Darn!
Anyway, I think I know what is causing the motor “pop”. The PWM value is just changing too fast. It needs some slew rate code to prevent it going from nothing to full throttle in zero time.
I still haven’t setup my Cl-84 for flight. I hope to do that when I get back from OBC in early October. I’d really like to give it a test flight.
Also, you’re right about MANUAL mode. As soon as you switch to MANUAL all quad stabilisation is disabled. The intention is users would switch to a mode like FBWA first, but we need to fix the pop issue before that becomes workable for the CL-84.
No problem. I have converted my CL-84 back to the stock FC and it flies great as is. My QuadPlane projects are keeping me busy now. My goal is to compare the QuadPlane concept for mapping against other more expensive solutions currently in the UAV commercial market.
I wish you the best of luck in this year’s OBC!
I have been flying my CL-84 for some time. My focus has changed from tilt wing to tilt rotor. Havent seen any posts here about tilt rotor with Pixhawk hardware and Mission Planner. I am currently using
a KK2 board on my DIY Osprey, that works after a fashion. I would really like to used the functions of the Pixhawk to control my Osprey. Any developments in this area that I have missed?
We have mounted larger wings, improved the tilt-mechanics, installed tridges code and done the first copterflight succesfull:
The next steps will be testing the transition.
The first part of the forward transition was successful. We chose 30 degrees as initial tilt angle, because the ratio of lift force (85%) and forward thrust (50%) seemes to us to be optimal. The plane accelerated quickly, stable as on rails and without any change in altitude. Thanks to Tridge for this firmware-version. Before complete switching to the FBW-A flight, I intentionally terminated the transition to see to see how the copter behaves at higher forward speed: In the attempt to slow down speed in copter mode, he still juggles up. We will have to adjust the PIDs even more sharply before the first complete transition.