[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.

[biggestRCEfan]

This will not change, I think. Triflight 0.7 requires betaflight or the latest cleanflight Configurator (tbc if cleanflight Configurator works). So if you want to run triflight 0.5 you need cleanflight Configurator v 1.2.4 frome here. To install follow steps elsewhere listed on the forum.

[biggestRCEfan]

The good thing about this hobby in modern times is that I found someone with the same problem and we are working the issue together. The one is in the US and the other in Australia. I don’t have to wait to ask someone at the club on the weekend.

I started with tricopters because they intrigued me. I own all three from David’s, after the first one being the FT Electrohub with wooden booms and naze FC. Built my first quad recently. Yes, quads are more agile and precise. And more robust.

The progress Lauka has made is fenomenaal. The V8 firmware is getting very close to the polished product we want.

Love this hobby!

[biggestRCEfan]

Wynn, here is my diff:

diff
# Triflight 0.7 Beta 2 – new control test v8 / RCEXPLORERF3 3.1.7 Sep 10 2017 / 11:15:56 (ca5d4df)
servo 5 1000 2000 1446 40 40 100 -1

feature MOTOR_STOP
feature TELEMETRY
feature CURRENT_METER
feature BLACKBOX
map TAER1234
serial 1 32 115200 57600 0 115200
serial 2 128 115200 57600 0 250000
aux 0 0 0 1700 2075
aux 1 28 1 1675 2075
aux 2 1 1 1300 1675
aux 3 31 2 1300 1700
rxfail 3 h
rxfail 5 s 1500
set min_check = 1010
set max_check = 1990
set debug_mode = NONE
set min_throttle = 1050
set motor_pwm_rate = 400
set align_board_pitch = 180
set gyro_lpf = 188HZ
set moron_threshold = 32
set deadband = 10
set yaw_deadband = 10
set servo_pwm_rate = 250
set tri_tail_motor_thrustfactor = 61
set tri_tail_servo_speed = 242
set failsafe_delay = 8
set failsafe_off_delay = 200
set failsafe_throttle = 1250
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 = 97
set i_pitch = 26
set d_pitch = 25
set p_roll = 105
set i_roll = 30
set d_roll = 25
set p_yaw = 180
set i_yaw = 35
set d_yaw = 35
set p_level = 20
set i_level = 10
set p_vel = 120
set i_vel = 45
set d_vel = 1
rateprofile 0
rateprofile 0

set rc_rate = 130
set rc_rate_yaw = 165
set rc_expo = 85
set rc_yaw_expo = 83
set thr_expo = 90
set tpa_rate = 0
set tpa_breakpoint = 1500

#

[biggestRCEfan]

Siclair, 0.5 is where we are coming from. Works well but there are other reasons to move to 0.7, including the simple reason that cleanflight Configurator (old version, needed lo run 0.5) keeps disappearing out of Google Chrome. In 0.7 with betaflight I am able to use Lua script on my Taranis to adjust PIDs and rates without hooking up the laptop. Another BF Configurator advantage is the diff function. Keen to know when triflight will merge back into BF, to access newer features.

Wynn, 0.7 beta 2 worked fine, except sudden yaw inputs caused jumps in pitch. Could have solved it with playing with the parameters you refer to but never tried. On Layla’s advice I went to V8 of the alpha code, which flies like a dream. Still playing with the PIDs on the yaw channel as it is not as crisp as for pitch and roll. Regarding low frequency oscillations I flew in strong, gusty wind (15km/h) on the weekend and hardly noticed any oscillations. Will post my diff later in the week.

[biggestRCEfan]

In a separate thread someone asked the same question and the response (from Lauka or Litter Bug) was that the F3FC does not support dshot.

[biggestRCEfan]

Just came back after I maidened my V4 with the 0.7 Beta 2 v8 firmware. Lauka was right, the jumpiness in pitch with sudden yaw inputs are gone. Had a blast flying LoS. Couldn’t do the in-flight tailtune as it was too windy.

Thanks Lauka, your work gives us much pleasure!

I now need to work on the yaw PIDs, as the yaw channel is a bit soft. Step inputs does not give the same sharp, crisp response as on the pitch and roll channels (my PIDs are the same as I used in 0.7 Beta 2). But first I need to figure out how to add yaw D gain to the lua script as for quads this parameter appears to be ignored and hence not included in the lua script. I am expecting that the D gain will have to play a roll in getting yaw response crisper.

[biggestRCEfan]

Servo feedback was added because it measures the position of the servo instead of the FC software needing to estimate the position based on other known parameters (which effectively what virtual feedback is).

So triflight is designed to ideally have the measured servo position feedback.

[biggestRCEfan]

Never flew with KK2 boards but you should look at stick scaling to adjust the sensitivity. Playing with P gain will change the dynamics of the response to stick change, but not necessarily the sensitivity (I may be wrong, but I did note that stick scaling is an option on KK2 boards so I expect the stick sensitivity is adjusted with the stick scaling feature).

I am curious: does the KK2 run a V-tail configuration defining the set-up of your props and the tail angles? How responsive is the yaw control?

Good luck with the r-build. Keep us posted!!

[biggestRCEfan]

Cheers Wynn.

Lauka suggested:

I would advise to try out the new yaw control branch.

Do you have airmode on? It should help with the pitch control

Will flash v8 of his alpha firmware he just sent out (I have 0.7 Beta 2). Will see if that helps.

[biggestRCEfan]

Thanks lauka, will try it. Yep, allways have airmode on.

[biggestRCEfan]

I upgraded to triflight 0.7 Beta 2 and increased the P gains on pitch and roll. It did appear to improve the low-frequency oscillation. Haven’t pushed them as high as Tabaris but will play more to see the best I can get. But first trying to get the tail dynamics sorted that came with the new firmware.

[biggestRCEfan]

Great, thanks Lauka. Will give it a try next weekend.

I took the jump to triflight on betaflight moving to 0.7 yesterday and maidened it today (stock V4 RCE tricopter). Main reason for making the change was to access PID tuning via Taranis lua script. PID changes work like a charm.

Flew LoS only and have only one issue: when flying smoothly I don’t have a problem, but when I release the yaw stick suddenly (or make sudden yaw inputs), the craft jumps in pitch – I think it pitches nose-up, as if it over-compensates for loss of lift due to the changing angle of the tail prop.

Any suggestions on what I can do to improve this? I increased and decreased yaw P gain but didn’t notice a significant change in behaviour.

tri_tail_motor_thrustfactor = 143
tri_tail_servo_speed = 307

This was on 4S.

Thanks for your good stuff, Lauka! I was hesitant to make the change to betaflight because I was quite happy with the handling of the 0.5, but betaflight gives more functionality and so far it flies really good.

Here is my diff:

# diff
# Triflight 0.7 Beta 2 / RCEXPLORERF3 3.1.7 Jun 4 2017 / 13:39:27 (1698ba5)
servo 0 1000 2000 1500 45 45 100 -1
servo 1 1000 2000 1500 45 45 100 -1
servo 2 1000 2000 1500 45 45 100 -1
servo 3 1000 2000 1500 45 45 100 -1
servo 4 1000 2000 1500 45 45 100 -1
servo 5 1101 1988 1553 40 40 100 -1
servo 6 1000 2000 1500 45 45 100 -1
servo 7 1000 2000 1500 45 45 100 -1

feature MOTOR_STOP
feature TELEMETRY
feature CURRENT_METER
feature BLACKBOX
map TAER1234
serial 1 32 115200 57600 0 115200
serial 2 128 115200 57600 0 250000
led 0 15,15:ES:AI:0
led 1 15,8:E:FW:0
led 2 15,7:E:FW:0
led 3 15,0:NE:AI:0
led 4 8,0:N:F:0
led 5 7,0:N:F:0
led 6 0,0:NW:AI:0
led 7 0,7:W:FW:0
led 8 0,8:W:FW:0
led 9 0,15:SW:AI:0
led 10 7,15:S:FW:0
led 11 8,15:S:FW:0
led 12 7,7:U:FW:0
led 13 8,7:U:FW:0
led 14 7,8:D:FW:0
led 15 8,8:D:FW:0
led 16 8,9::R:3
led 17 9,10::R:3
led 18 10,11::R:3
led 19 10,12::R:3
led 20 9,13::R:3
led 21 8,14::R:3
led 22 7,14::R:3
led 23 6,13::R:3
led 24 5,12::R:3
led 25 5,11::R:3
led 26 6,10::R:3
led 27 7,9::R:3
aux 0 0 0 1700 2075
aux 1 28 1 1675 2075
aux 2 1 1 1300 1675
aux 3 31 2 1300 1700
rxfail 3 h
rxfail 5 s 1500
set min_check = 1010
set max_check = 1990
set min_throttle = 1050
set motor_pwm_rate = 400
set align_board_pitch = 180
set gyro_lpf = 188HZ
set moron_threshold = 32
set deadband = 10
set yaw_deadband = 10
set servo_pwm_rate = 250
set tri_tail_motor_thrustfactor = 143
set tri_tail_servo_speed = 307
set tri_motor_acc_yaw_correction = 0
set tri_dynamic_yaw_maxthrottle = 70
set failsafe_delay = 8
set failsafe_off_delay = 200
set failsafe_throttle = 1250
set failsafe_procedure = AUTO-LAND
set baro_cf_vel = 0.985
set baro_hardware = AUTO
set pid_process_denom = 1
set blackbox_rate_denom = 2
profile 0

set p_pitch = 77
set i_pitch = 26
set d_pitch = 25
set p_roll = 85
set i_roll = 30
set d_roll = 25
set p_yaw = 140
set i_yaw = 45
set d_yaw = 33
set p_level = 20
set i_level = 10
set p_vel = 120
set i_vel = 45
set d_vel = 1
rateprofile 0
rateprofile 0

set rc_rate = 130
set rc_rate_yaw = 150
set rc_expo = 85
set rc_yaw_expo = 83
set thr_expo = 90
set tpa_rate = 0
set tpa_breakpoint = 1500

#

[biggestRCEfan]

@lauka, a few questions for you regarding triflight 0.7:
1. How mature is it to be used on the V4 (big tri stock electronics and motors)? I am still running 0.5 on Cleanflight and want to change to Betaflight but don’t want to do it if I am going to struggle to get it working properly (PIDs, servo feedback pin and other troubles I have been reading about).

[biggestRCEfan]

Take 2:

Are you flying accro or angle mode? In angle mode, neutral pitch/roll stick should give level flight and no/little drift. This can be trimmed by disarming, throttle max and roll/pitch sticks used to increment/decrement parameter called something like acc_x and acc_y. One beep for each increment. These values are saved when you reduce throttle to zero (if memory serves me well).

If flying accro, neutral stick (pitch/roll) should result in no commanded pitch/roll rate. Any disturbance (wind) will push it in a direction and it will drift. No wind, it should hardly drift at all. Look at your dead band as well, as the radio may jitter on pitch/roll commands even if you don’t move the sticks. This is normal with digital systems.

For stick sensitivity you could reduce rc_rate to reduce the amount of angle/rate commanded per unit of stick deflection. Do this before trying to adjust the PIDs, as the stock PIDS in my experience are pretty good.

For yaw drift, I would check if there is binding in the tilt mechanism.

[Author]

Posts