After looking at the transition test video, I had forgotten about the boom extensions for the front motors. I was thinking they might be in the front fuselage.
I also take back what I said about no yaw control. If you adhere to the motor rotation scheme for the Quad + image above, then you will have yaw control without needing either individual tilt capability on each wing or a rear motor that can move left or right like on the CL-84 R/C VTOL.
Q_TILT_TYPE should still be 0 (Continuous). You then need Motor Tilt (SERVOn_FUNCTION=41) for the transition output port on the PixhawkâŚsay AUX1.
What do you mean by not getting the wing to rotate automatically? If the wing does not transition up and down when changing flight modes from QSTABILIZE to FBWA then the transition servos are not set up properly. If you post a configuration file (.param) then it will be easier to analyze.
Ok, so I was out over the weekend due to travel and thesis writing, but Iâm back in the lab now.
Attached is my current param file. I switched the flight mode from QHOVER to QSTABILIZE, but still couldnât get my wing to rotate (havenât even worried about the horizontal stabilizer yet), and my front vertically mounted motors wonât spin up either.
Also, if I have throttle on and all but my front two motors spinning while in manual mode, switch to FBWA, they keep spinning, as I would expect. However, If I then proceed to switch to QSTABILIZE, they still shut down and stop spinning, but weirdly enough, I still maintain aileron and elevator control (no rudder).
I am using the same Pixhawk outputs as you previously stated:
Main1: aileron
Main2: elevator
Main3: throttle
Main4: rudder
Main5: right wing motors
Main6: left wing motors
Main7: forward boomed motors
Main8: tail motors
Aux1: Motor Tilt (SERVOn_FUNCTION=41)
And within the Futaba radio, the following channel outputs are as follow:
Channel 1: aileron
Channel 2: elevator
Channel 3: throttle
Channel 4: rudder
Channel 5 - 8: throttle (should this be different?)
Channel 9: Aux 1
Channel 10: Aux 2
I suggest you change your Aux 1 and 2 to one of the first 8 outputs, that way if your processer fails in-flight, youâll still have a chance to fly manually by RC pass through (that only works on channels 1-8). Move your forward motors to channel 9, and your rear motors to channel 10.
Try changing servo function from 41 to 75 or 76. It shouldnât matter 75 or 76 since Q_TILT_TYPE =0. Otherwise I donât know whatâs going on. Maybe the mask advice I gave earlier is no goodâŚ
How is 5v getting to the Main Out/Aux ports? Do you have a BEC plugged in? Do you have Opto or BEC ESCs? The later requires that the middle red wire be removed before plugging into the Main/Aux ports so that multiple BECs are not shorted together. In summary, one of the Main/Aux ports needs to supply 5v to the servos and ESCs. Preferably, this is not from an ESC BEC.
Are you using a GCS like Mission Planner during these tests? I wonder what your HUD is displaying and are you really âarmedâ? In other words, have you held the rudder stick far right until you hear the arming tone? Sometimes, the rudder is reversed and you need to hold it far left to arm until you get the yaw direction sorted for Quad mode.
In your .param file, you have ARMING_RUDDER set to 0. Make it 1 or 2.
Ok, so it turns out I had a mechanical issue with the linear actuator mechanism. It is moving now (although still not itâs full range, so I will mess around with that), but my motors are still turning off when I switch to QSTABILIZE or QHOVER (although running with the recommendation to use QSTABILIZE instead).
I have perfect servo and motor control in manual an FBWA, but no motor control in QSTABILIZE.
Also, weirdly enough, I still have servo control in QSTABILIZE - this seems weird to me, shouldnât I lose servo control in this mode?
Also, I disabled that arming mechanism, so in theory, it should be automatically armed. Yes, Iâm using MP. I will enable rudder arm again and get back to you here in a couple minutes.
Check the channel 5 through 8 outputs in the mission planner status tab to see if there is anything going on. That will tell you if itâs an ESC problem or a code/setup/parameter problem.
they do beep, but were running in manual mode, so not sure what was going on there
@Naterater and @GregCovey, Iâm thinking it was a parameter thing now. I changed the ARMING_RUDDER = 2, and ARMING_REQUIRE = 1, and now everything seems to work the way it should! So next step is Iâll put it back together, run it through the modes, then take it outside, put props on it and test the hover.
Thank you both very much! I will post on here again with an update as to how things go!
So after some initial testing, I have two questions:
How can I increase the maximum throttle? I canât seem to find a parameter in the VTOL aircraft like the Copterâs THR_MID parameter. Currently, at full throttle, the aircraft will not takeoff. It is on the verge of taking off, but is still on the ground. It has been designed for a T:W of about 1.2.
In the hover configuration (QSTABILIZE), the pitch is reversed, but in FBWA and manual, the pitch is correct. Any idea of how to reverse pitch in just the QSTABILIZE mode?
If your ESCâs are outputting 2000PWM, or whatever they were calibrated at, then there is nothing you can do to increase thrust. I have a feeling that you have a mechanical/electrical problem - not an ardupilot problem. A 1.2:1 T:W ratio isnât going to cut it - you lose 20% of your voltage during battery drain anyway, not to mention voltage sags from hover power that is usually significant. Most people are looking for at least a 1.5:1 T:W ratio on takeoff.
Double-check your board orientation, transmitter channel reversals, and servo reversals. Start with QStabilize. Then check FBWA. Finally, MANUAL. With each mode, check for the correct corrective action when you move the airframe and the correct action when you apply input to your transmitter. The responses will tell you what to reverse - SERVOn_REVERSED or RCn_REVERSED.
Ok so everything is working, I put larger motors on, but now I have to tune the gains in hover. It is currently all over the place whole in QSTABILIZE. My controls are in the right direction, but it does not want to stabilize itself. Is this likely gains or something else? It isnât a simple oscillation, just seems to get worse and worse. Iâm reading up on this problem, but was hoping one of you might know.
Your description is too vague so maybe another video would help or post the .log file from the test.
The two parameters that will have the greatest effect on your initial hovering are Q_A_RAT_PIT_P and Q_A_RAT_RLL_P. If they are 0.2 then try 0.3, etc.
Also, I use Multi-rotor ready ESCs on my QuadCopters. SimonK or BLHeli firmware makes the ESCs respond much faster to Flight Controller inputs. Hopefully, you have chosen this path.
Usually oscillations are a result of gains being too high, but thereâs no perfect way of knowing. http://ardupilot.org/copter/docs/ac_rollpitchtuning.html#ac-rollpitchtuning is a good reference, but you wonât be able to use the copter GUI for tuning â youâll have to use the full parameter list or tree.
Ok, so itâs been a while without an update. I test flew the aircraft (thought I had the hover gains fairly honed in), but in the middle of the transition, it decided to make one sharp pitch up, then a sharp nose down and crashed. I will post a video sometime in the near future. I think it may have been gains?
Also, what do you both think? After fixing the aircraft, I am thinking about trying to put it in the VTOL configuration (semi-permanently), then flashing the copter firmware and performing an auto-tune on the gains. After I get the PID values, I would manually put them in the Q parameter list. Have either of you done something like this?
Right now, I am having to just do nothing but sit down and write my thesis as Iâm defending next week. So I probably wonât be attempting anything more until next week sometime.
Sorry about your crash. Without looking at the video, my suspicion (based upon the frame) would be to guess that your booms may allow a see-saw action which disrupts the transition.
A good practice is to first confirm a reasonable hover capability, then gain sufficient height so that if something goes wrong on the transition to FBWA you can switch back to QSTABILIZE for your emergency bailout back to hover.
As for using Copter PIDs for a QuadPlane, I am not certain if they are equal.