Hi,

I recently discovered a post on RCG here back in May 2015 from a new member about using a single Pixhawk for VTOL control on the BirdsEyeView FireFLY6. It lead to the PX4 page below describing some development firmware for the Pixhawk. Does anyone know if there have been updates to this development or how to obtain the firmware for testing?

VTOL Firmware for Pixhawk

PX4 Page to BirdsEyeView FireFly

Thanks.

Views: 11306

Reply to This

Replies to This Discussion

Chris and Tridge,

That's awesome news! In addition to my FireFLY6, I also have the HK CL-84. I flew it stock for the first time last Saturday. If Vimeo wasn't locked out at work I would post my video. It flies fine but only turned well to the right, not to the left and the transition from forward flight to hover is quite sloppy. I don't even think that the ESCs use SimonK or similar. The CL-84 is quite "barebones" compared to the FireFLY6 with AvA so I am thrilled to hear that you are adding FireFLY6 to APM:Plane!

I'll plan to replace the stock controller in the CL-84 with a Pixhawk and make a project out of it. It sounds like a great new venture for APM!

Regards...

This is my near stock Canadair CL-84 VTOL from Hobby King video. The only modification I have done is to add a steerable nose wheel. It flew great and transitioned fairly well. I could hear the high-pitched whine of the speed controls in flight. Although I could turn to the right very well, the left turns were very shallow requiring plenty of space.

The video shows my second flight of the CL-84. The first flight was a bit better but I wanted to transition back to hover straight down the field for the camera so it fought the slight cross wind for a moment. It's a fun model and should be even better with the new APM:Plane feature!

Canadair CL-84 VTOL Flight 1 from Gregory Covey on Vimeo.

great! Really glad to hear someone else has a CL-84 to test. I have mine put together but haven't finished fitting a pixhawk to it. I'm not going to fly it with the stock controller, instead I'll go straight to ArduPilot.

One thing that is a bit of a disappointment is finding that the tiltwing servo is a retract servo. So you can't actually control the tilt to any angle, it is either fully up or fully forward, with the change happening over about 4 seconds. That is a shame as it means a lot of the smooth transition logic I've added for the firefly won't work for this model. I'm going to create a CL-84 simulator that has the same characteristics and then add a Q_TILT_TYPE parameter which will allow you to tell ArduPilot that the tilt mechanism is binary not continuous. Then I'll change the tilt control logic so it knows about the rate of tilt and thus what angle the wing is at during transition. That will allow it to properly compensate in the tricopter code for the tilt.

The main issue I see with that is if the plane hasn't gotten above its stall speed by the time the wing tilts fully forward then it will stall. Do you have any idea what the stall speed is? It is a pretty small wing and the model is quite heavy for its size so I'd guess 10m/s or so, but perhaps you have a better idea?

Also, how long does the battery last if you are mostly doing forward flight? In your video you fly for about 2.5 minutes, do you know how much battery it used? What battery capacity did you use?

Do you know if the stock controller use the rear motor for yaw control in forward flight? Or does it use differential thrust on the forward motors at all?

Cheers, Tridge

Hi Tridge,

My CL-84 video was edited to remove several of the forward flight passes. I flew for about 4 minutes and used 50% of my 2200mAh pack. This seems to be in line with the consensus I have read from other CL-84 owners. The hover-only time on a 2200mAh pack is about 4 minutes and the forward flight time about 8 minutes with shorter hovers to start and land.

I agree with you on the tilt servo so it would be very interesting if you can compensate for it in the firmware. Otherwise, we can probably swap the servo out. The Chinese are amazing when it comes to these intricate mechanical mechanisms like the wooden tilt control in the CL-84.

On my FireFLY6, I see the main motors go from hover to just a bit forward which starts the forward motion of the wing and then the motors finish the 90 degree forward rotation with some sort of turbo boost that you can hear before it settles down to fly. The wing has been designed and the parameters set to make it almost impossible to stall...which is good for newbies. I have been communicating with their autonomous survey support in Florida (Erich F.) to change some parameters for my preference. The parameters are in their Full Parameter List but not always documented with text or units.

In hover mode on the CL-84, the rear motor provides both back end stabilization and rudder control as it can rotate left or right. In forward flight mode, you have ailerons and motor speed differential on the mains. The rear motor is completely disabled in forward flight mode. Some people have added rudder control for better authority. Some guys have re-calibrated their ESCs to fix the poor turning in one direction. Apparently, the motors were spinning at different RPMs.

I don't think the stock ESCs will work with APM but perhaps I am wrong and it just changes the PID values. In my video, during the forward flight transition to hover you can hear the sloppy side-to-side effect from the non-SimonK ESCs.

As far as the stall speed, I honestly don't remember trying to slow it down. I'm used to flying the FireFLY6 in Alt. Hold mode so the throttle is set to about mid. On my CL-84 flights, the throttle was a bit above mid and I just left it there because it flew pretty solid. As you mentioned, the wing loading is a bit high for a parkflyer. Some guy from Germany posted this, "its high wing load of 38 oz/sq.ft (or 117 g/dm). This high wing load should lead to a stall speed of roughly 30mph (or some 50km/h)".

Regards...

If the stall really is 50km/hr then that means it has to accelerate from nothing to 14m/s before the tilt mechanism finishes rotating. That is pretty fast accel. It will be running the fwd motors at roughly full throttle during the last stage of the transition (it has to in order to equalise with the pitching moment of the tail) and maybe that is enough?

I don't have differential on the fwd motors for yaw in the ArduPilot code yet. That will be worth adding, and is something people with twin motor fixed wings have asked for before.

Thanks for the info!

Tridge,

I have a spare Pixhawk that I can start mounting in my CL-84 this weekend. Your "Tiltrotor support for plane" page in Ardupilot seems to reference a wing setup with elevons like the FireFLY6. Let me know when you have a build for the CL-84 testing and some output documentation if it is different than the normal plane configuration. In the mean time, I could load copter on it and play with modes like Stabilize, Alt. Hold, and Loiter. Just adding a Pixhawk with BARO and GPS to the CL-84 will be fun!

I finished swapping out my stock ESCs for RC Timer ESCs with faster SimonK firmware. The main two ESCs were 45A and the tail ESC was 30A. All three were overkill in size so little to no cooling is required. The 4s Lipo voltage makes the current draw small.

Since the new ESCs were Opto-isolated, they had no BEC. It turns out that the stock CL-84 electronics is fed from the BEC on the tail ESC, so I needed to add a 5v regulator which I planned on doing for the eventual Pixhawk controller upgrade.

My initial hover test in the backyard was quite successful. The CL-84 held position nicely and was easy to control. I'll get another video of it flying before I replace the flight controller and make it a real autonomous VTOL.

My stock tail prop broke when I was putting it back on so I replaced it with an Aero-naut 6.5x4 carbon electric prop. No drilling was needed and it works great!

Notice that I also modified my elevator to use five Dubro #119 small nylon hinges. This made the elevator much smoother and eliminated the binding I had with the stock setup.

Looks great!

I think we should move the discussion of the CL-84 with ArduPilot over to the discourse server thread:

http://discuss.ardupilot.org/t/tiltrotor-support-for-plane/8805

I feel a bit bad hijacking this discussion of FireFly6 with PX4 to be a CL-84 with ArduPilot discussion :-)

Sure, no problem. This thread has changed directions several times as part of discovery and as the available technology changed. My first main decision here was to choose the APM stack. Next, the BirdsEyeView AvA feature appeared which provided the use of a single Pixhawk with the APM stack.

Now that VTOL is becoming a feature of APM:Plane, it will provide a new level of possibilities for the hobbyist, researcher, and professional.

Hi,

I'm new with Pixhawk and I want to use it in a Y6 tiltrotor aircraft I just assembled. 

I'm using QGroundControl v.3.0.2, where I downloaded the Y6 tiltrotor BirdsEyeView configuration. I went to all the sensors calibration process, and during the radio calibration I allocated a RC channel to the AUX1, which is supposed to command the transition. When I start a bench test without props, I can see the hovering phase acting properly, but when I switch the radio channel for transition, nothing happens and the aircraft is still acting as in hovering (all 6 motors running, no rotation of the forward motors beam).

I got a pitot tube and I set the transition airspeed to zero, just to be sure that no failsafe mode was messing with the transition, but still nothing happens when the transition AUX1 channel is switched.

The pins connections I'm using are:

AUX1: tiltrotor servo;

AUX2-AUX3: Elevon servos;

MAIN 1  to MAIN 6: 6 motors.

Have you got any suggestions for me?

Thanks,

-Paolo

it is my problem too, i stuck in this step unfortunately ...

Paolo said:

Hi,

I'm new with Pixhawk and I want to use it in a Y6 tiltrotor aircraft I just assembled. 

I'm using QGroundControl v.3.0.2, where I downloaded the Y6 tiltrotor BirdsEyeView configuration. I went to all the sensors calibration process, and during the radio calibration I allocated a RC channel to the AUX1, which is supposed to command the transition. When I start a bench test without props, I can see the hovering phase acting properly, but when I switch the radio channel for transition, nothing happens and the aircraft is still acting as in hovering (all 6 motors running, no rotation of the forward motors beam).

I got a pitot tube and I set the transition airspeed to zero, just to be sure that no failsafe mode was messing with the transition, but still nothing happens when the transition AUX1 channel is switched.

The pins connections I'm using are:

AUX1: tiltrotor servo;

AUX2-AUX3: Elevon servos;

MAIN 1  to MAIN 6: 6 motors.

Have you got any suggestions for me?

Thanks,

-Paolo

Hi Paolo and babak,

When you say, "I downloaded the Y6 tiltrotor BirdsEyeView configuration", I am not certain what this means.

If you meant that you downloaded a Pixhawk configuration file then you do not have the required custom FireFLY6 firmware for the Pixhawk.

If you meant that you downloaded the custom FireFLY6 Pixhawk firmware, then you still need the AvA key to activate it.

Lastly, if you are using the older "Classic FireFLY6" firmware, then you need the Transition Bridge module and a second flight controller; one for hover mode and one for forward flight mode.

I hope this helped...

Reply to Discussion

RSS

Groups

Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service