Debugging the tricopter mini racer

[LeoTheHuman]

The math is very simple. Just divide thrust by the cos() of a deflection angle and you will keep the vertical part of the thrust intact while tilting the motor. This will lead to non-linear yaw response (less with a small stick deflection, much more with higher deflection), but that is what we usually want anyway. (:

[shufty]

I totally see now what David meant about getting the Tri to be as locked in as a mini quad!

Struggling to get the little bugger to stop waggling it’s tail in fast forward flight and just feel more locked in than it currently is!

Any ideas on reducing the tail wag?

[Terje]

@shufty

Try configuring the hardware this way:

Left front motor: CCW
Right front motor: CW
Tail motor: CCW

Load the triflight_0.1_betaflightPick_NAZE.hex firmware onto your Naze32 board, load @David CLI settings.

That should take care of the tail wag 🙂

– Report back on the results!

[ZoomnBoom]

If one tilting rotor is a good thing, two has to be better. With 2 tilting rotors (ideally on a Tcopter so the tilt angle is symmetrical) you could control frame/camera angle, maintain low drag at high speeds, and you could effectively brake to get around corners.

To make the most of it, would require flight controller support, but as is, you could just use a V tail mixer to mix the flight controller yaw servo signal with an additional transmitter channel for tilt. Then put the tilt on a three position switch or slider and off you go.

[LeoTheHuman]

If you make symmetrical tricopter with 120 degree angle between all arms and servos on all motors you will be able to fly in any direction keeping any angle you like (true 6 dof multirotor). (: Although it would be absolutely cool, it won’t really help in racing…

[RCExplorer – David]

Cool videos Terje 🙂 Looks like your getting a hang of FPV flying 😀

Still haven’t gotten around to sticking the blackbox in the copter yet. Been raining and I’ve been really busy. Hope to get it up again soon.

[shufty]

Hey Terje. that’s exactly how my mini tri is already setup 🙂

[Patrick]

Hi, know the Betaflight 1.10 only from MiniQuads with the BorisB additions. Does the Triflight/Betaflight recommended also for the normal Size Tricopter V3? – Better than the old 1.9 rc explorer naze version?
Thank you

[LeoTheHuman]

Afaik, changes for @David’s board layout were not integrated into cleanflight. Have someone noticed any problems when loading betaflight as is into the integrated board?

[shufty]

Works fine, just had to make sure I changed the board align pitch to 180.

[Terje]

@David (or anyone who has done it successfully) could you describe how to connect the blackbox data logger to the integrated Naze32 frame and setup the logging?

[Michael]

So 1st post here. But I wanted to reply to Leo about problems with the betaflight. I flashed my my naze frame this past weekend several times. The only thing that would happen was a bunch of twitching and noise. The servo set up screen gave me a bunch of odd numbers and the receiver screen did not recognize any of my radio inputs I changed the receiver type and still nothing. I have been flying my mini for almost two months now, and I have configured it twice. It is by far the best trike I have built. I fly los due to nausea when I tried fpv. I get car sick if I am not driving. So lots and lots of aerobatics. The 1st pic is my 1st version and the second is the current version. Any way I would love to get rid of the tail wag for when I do actually give her the beans. Under full throttle she shakes like my Dachshunds do when I get home from work. But most of the time I do not see anything. The only changes that I have made to Davids 1st recommended set up have been to increase the rates and adjust the rudder pid’s.

[Michael]

Looks like I need to adjust my photo file size. Sorry about that.

Attachments:

1. 

   IMG_20150709_1124241301.jpg
   

2. 

   IMG_20150817_2221123541.jpg
   

[Terje]

Just loaded Betaflight 1.10 to a test Naze32 board.

Tested the servo tab on the tricopter and the flying wing. Changes are not saved.

This is the same freaking bug Cleanflight had in the 1.9.0 version! The bug is also in the Triflight 0.1 version since it is based on the Cleanflight 1.9.0 version. The bug is NOT in the version @David released. The bug has been reported to Dominic and he has closed it.

[lauka]

I implemented the throttle based servo control on Sunday, and it actually worked quite well out of the box. The copter was able to maintain yaw heading when hovering and throttling up, even when I set I term of yaw PID to zero. The problem is that I couldn’t yet figure a good way to couple this correction with the PID controller. Currently it is independent from the PID control, modifying the output of the controller. This is confusing to anyone trying to tune the copter in blackbox as this correction can’t be seen in the logs. This also adds extra step to the tuning process. I’m quite busy currently as my vacation ended this week, but I will continue the experimentation and testing of this at some point.

@Terje, Dominic closed the issue on this commit. The commit has been in Cleanflight master since June 10, therefore it is included in Triflight and Betaflight. I suspect that this has been broken again in the newest configurator.

[Author]

Posts