New type of yaw control

[AtomicHemi]

Hi All,

I have flashed the new control 8 onto my baby trip and my V3.5 integrated naze

Only had chance to of hover tests on both, with the baby if i do a little punch out and chop the throttle the tri does a 90 deg left yaw, tail tune was completed not sure where to start

for the v3.5 just a bit of advice check your board alignment in the configurator and check that when you tilt the copter the animation does the same

On the v3.5 I am having some trouble with the bench tail tune, I have finally installed a feedback servo but when doing the bench tune the servo goes to the left back to centre then I get a long beep, might be my fault as I do not have the template and have set the end points by comparing to the baby tri and trying to line up the baby tri tilt tool

at the moment it hovers but has a lot of pitch/roll oscillation

Thanks

[wynnsquad]

Just flashed v8 to my tricopter V4 and it seems pretty nice. I am having the same behavior that @biggestRCEfan is having where sudden yaw input leads to significant pitch deflection. When airmode is off, the pitch deflection due to sudden yaw is much less, but as a consequence, when I punch out, I get un-commanded pitch deflection. Kinda interesting. Anyway here is my diff and I’m working on acquiring a datalogger.

# diff
# Triflight 0.7 Beta 2 - new control test v8 / RCEXPLORERF3 3.1.7 Sep 10 2017 / 11:15:56 (ca5d4df)
name RCE_V4
servo 5 1079 1919 1483 40 40 100 -1

feature -RX_SERIAL
feature RX_PPM
feature CURRENT_METER
serial 0 1 115200 57600 0 115200
aux 0 0 1 1600 2000
aux 1 28 3 950 1300
aux 2 1 0 1000 1250
aux 3 2 0 1375 1675
aux 4 31 2 1650 2000
rxfail 4 s 1100
rxfail 7 s 1925
set align_board_pitch = 180
set gyro_lpf = 188HZ
set moron_threshold = 32
set deadband = 10
set yaw_deadband = 10
set tri_tail_motor_thrustfactor = 110
set tri_tail_servo_speed = 307
set failsafe_off_delay = 50
set failsafe_throttle = 1350
set failsafe_procedure = AUTO-LAND
set baro_hardware = AUTO
set pid_process_denom = 1
profile 0

set yaw_accel_limit =  10.000
set p_pitch = 44
set i_pitch = 26
set d_pitch = 25
set p_roll = 47
set i_roll = 30
set d_roll = 25
set p_yaw = 200
set i_yaw = 20
set d_yaw = 33
rateprofile 0
rateprofile 0

set rc_rate = 85
set rc_rate_yaw = 85
set rc_expo = 30
set rc_yaw_expo = 30

# 

[biggestRCEfan]

Wynn, this is not what I have with V8. What you describe I had with the basic 0.7 beta 2. With V8 mine is near perfect.

[lauka]

@wynnsquad: try lowering the tri_yaw_boost. It defines the maximum yaw output. With smaller value it will generate less movement on the pitch axis.

It’s also possible that your servo angles are a bit off.

[Ambrose]

Hello everyone, i have an eagle tree vector rcexplorer v4 and i’d like to know if anyone has any recommended pid values, gain values or tuning advice to make my tricopter fly and hover as stable as possible. also the tail keeps drifting the entire copter to the right without any input from rudder. Any help or advice is welcomed. Thanks

Ambrose.

[lauka]

@ambrose: what version of the FW are you running?

It should not drift in hover. Have you checked the centering of your channels?

Any more experiences from the v8? Thinking of making the beta 3 out of it.

[wynnsquad]

@lauka Thanks for the tip. I’ll try lowering the tri_yaw_boost parameter and report how it goes. I noticed that my tri_tail_motor_thrust_factor and tri_tail_servo_speed were both quite a bit higher than RCEfan’s (possibly because I’m using 6V on my servo?) so maybe that explains why my yaw is acting so much more aggressive in comparison.

@biggestRCEfan, sorry about that, looks like I misread an earlier post!

[fomoco_44]

@lauka
I’ve been flying V8 on my MiniTri but have been battling a tune issue so I hadn’t really got around to posting feedback until now.

My flight this morning:
Sync @ +18.145
https://1drv.ms/f/s!AtWLDB57HPnrgzY9W8GjFhwb2ooI
https://1drv.ms/v/s!AtWLDB57HPnrhEM3nscrl-dlDjuW

I still have a lot of oscillations on the yaw axis which I think is also showing up the roll+pitch, I’ve already dropped the yaw P to half of what the V8 defaults are, and was planning to half the yaw D for the next test.
From my experience, the V8 code feels great from the yaw control point of view. Yaw authority at low throttle is great, and flips and rolls are much more predictable than previous versions. I Like!

Setup:
Mini Tri frame with integrated F3FC, V8 firmware
Stock EMAX motors and AIKON 30a ESC’s, set to Multishot
4s 1500mah LiPo
GEMFAN 6040BN props
BMS-A206 digital servo @ 6V(with feedback) – https://hobbyking.com/en_us/bms-a206-mg-digital-mini-servo-3-2kg-0-05sec-22-5g.html
Blackbox logger + MinimOSD

# diff
# Triflight 0.7 Beta 2 – new control test v8 / RCEXPLORERF3 3.1.7 Sep 10 2017 / 11:15:56 (ca5d4df)
name MiniTri
servo 5 1008 1885 1458 40 40 100 -1

feature CURRENT_METER
feature BLACKBOX
feature AIRMODE
map TAER1234
serial 1 128 115200 57600 0 250000
serial 2 1 115200 57600 0 115200
aux 0 0 0 1775 2000
aux 1 4 0 1775 2000
set motor_pwm_protocol = MULTISHOT
set serialrx_provider = SPEK2048
set ibat_scale = 350
set align_board_pitch = 180
set gyro_sync_denom = 2
set deadband = 2
set yaw_deadband = 2
set servo_lowpass_hz = 140
set tri_tail_motor_thrustfactor = 107
set tri_tail_servo_speed = 579
set baro_cf_vel = 0.985
set pid_process_denom = 1
set blackbox_rate_denom = 6
profile 0

set d_lowpass = 90
set vbat_pid_gain = ON
set d_setpoint_weight = 60
set p_pitch = 55
set i_pitch = 40
set d_pitch = 35
set p_roll = 48
set i_roll = 35
set d_roll = 30
set p_yaw = 30
set d_yaw = 45
rateprofile 0
rateprofile 0

set rc_rate = 130
set rc_rate_yaw = 130
set rc_expo = 20
set rc_yaw_expo = 10
set roll_srate = 75
set pitch_srate = 75
set yaw_srate = 75

On a side note, is there a limit(s) or validation on the in-flight TailTune that prevents it from saving??
I have a Quanum Trifecta with an F3FC on V8 and Savox SH255MG servo(with feedback) which i can get to tune inflight no problems, but as soon as I land(and disarm) it the tail servo wobbles violently and I get the 2-Beep failure notification…..even though I was getting the success beeps in-flight. Thrust factor remains at 54 so I’m positive that the tune didn’t complete properly.
I was thinking of enabling the servo lowpass to reduce the servo wobbles from the structural mode issue, as per this:
https://github.com/martinbudden/betaflight/blob/master/docs/Mixer.md

Any thoughts?….I was hoping to get the Trifecta working so I could help fault-find the MiniTri and also confirm the benefits of the new V8 code.

Another side note…..for the Trifecta the servo mechanisim is via gear drive, so I’ve had to use the CLI smix command to reverse the servo direction. This works fine for normal flight, however during the bench tune the servo direction is still reversed. Not a big problem, and the tune still works and saves correctly, just thought I would let you know about it in case it was an easy code change.

[mcstan]

@lauka

Could I get this build for KAKUTEF4 because I cooked my last f3FC and I wanted something a bit more simple to work within the terms of soldering.

Thanks

Stephen

[mcstan]

Hi I managed to install triflight beta 2 to the KAKUTEF4

I set the servo to motor 6 pin seen that was using a different timer

so I used the below command

resource servo 1 c08

Then I solder the feedback wire to the RSS pad and then had to set the feedback to look at it

set tri_servo_feedback = EXT1

and I had to move motor 3 pad to motor 4 pad

resource mtor 3 A02

all seems to work!! 🙂

now if I could get that build of this version that would be great @lauka

[lauka]

@mstan: good work! 🙂 I will include the changes to the beta 3 release.

[lauka]

@mcstan: Did you have a typo in the set tri_servo_feedback = EXT1? I guess that should be the RSSI? The Kakute doesn’t seem to have the EXT1 channel set.

[mcstan]

I did a spreadsheet when I was working it all out

so i knew what timing was on what channel

triflight was the beta 2

spreadsheet below

https://docs.google.com/spreadsheets/d/1U9GsH-BYLhi9PXOBwyAU9qemBFHNsGiePXEKDfhY73M/edit?usp=sharing

I can only guess EXT1 was the marked RSSI pad on the FC, my one was the all in one version

I did do the servo speed test and it seems to do it. IE did the crazy tail wag like a happy puppy 😛

[lemarec]

Hello

I Have a Holybro kakute F4 AIO V2,
Is this controller working with triflight and where did I set the servo ?

Thanks for any help

[mcstan]

Hey

I only had the version one… I did look at getting the version 2 but didn’t see a reason too yet.

you could try mapping it too UART 5?

I don’t know what the resource map looks like on Holybro kakute F4 AIO V2

[Author]

Posts