If you’ve got an FrSky receiver, you’re likely to have either a PPM or SBUS receiver. So, what receiver are you using? You are likely to be using one of the following brands: FrSky, FlySky or Spektrum. We’ve also created several little guides that should help to take you through these processes: Otherwise, none of this will work… If you are unsure how to do this, check out your receiver and/or radio manual. Before following the procedure I’m about to set out, please make sure that your receiver is bound to your radio transmitter, and correctly connected to your flight controller. Note: Airmode might be off, a higher value on low could get more stability.Next, we’ll move on to setting up your receiver on your quad. Its not even close to tune a quad in any way, harder as hell… My mini thats going on 3.2 and my baby thats still on 1.7 but custom built is a big difference, and I would assume that all wound become even bigger with 3,2 on both in terms of tuning.Ĭould be rubbish but my thoughts so far on the switch and the new filters. It has become more personal in a way and you probably could see the “gear” used in a another way. All is to personal preferences tho and ive only tested with 7″ props with 2405/2250 motors so the tiny tests ive got the time fore I would say that with BF in general the tune will give a bigger difference between setups than before. Yaw is a bit special on a tri so harder to tune with the new filters. In general BetaFlight is more aggressive on the PIDs than CleanFlight was, I guess the tune would be around 25-30% above Bf on pith and roll (Not all numbers tho, they will be different depending on setup). To reduce the instance of LFO, Increase the default tri_dynamic_yaw_boost setting for all tricopters. Change the default Servo PWM frequency from 300hz to 250hz as this is the correct PWM frequency for the stock BMS-210DMH Servo. ![]() Keep in mind that this all assumes that the servo has been installed and setup correctly in the first that being said I’d suggest the following changes to Triflight: Finding the right balance is key to removing Tail-Wag / HFO / LFO. To high of a value and the servo will tend to over-shoot / over-react to subtle yaw adjustments (HFO). Too low of a value and the servo will act as if it’s delayed, causing LFO to apear. For all intensive purposes, it appears that the yaw boost controls the servos reaction speed. A value of 360 caused the Tail-Servo to twitch periodically. ![]() With the default value of 240, Oscillations were nonstop at around 3-5hz. The yaw boost aparently can have a strong effect on LFO. Once the WAG / HFO / LFO have been tamed, it should be possible to push the Yaw terms higher. Setting too hight of a Yaw P-Term results in High Frequency Ocsillation (HFO), while too much Yaw I-Term can cause LFO. Thus the motors dont start to overheat and so only the relavent data is being passed. The term / filter settings within the PID tab are mostly for clarity. Before I get into it, here are my PID’s, filters and specific CLI settings. Check the attached diff file for further information on the settings used.Īfter much inhouse tuning and test flights I “may” have figured out the issue of Low Frequency Oscillation (LFO). ![]() Obviously they may or may not show correctly in the log file. Keep in mind that the PID’s used in the flight were not saved as I didnt know how at the time. Note: All Notch filters are disabled, with just Dynamic active. My current Lowpass setting are as follows: I beleive that I have bad filtering setting that are causing the LFO to dominate regardless of the PID settings used. Did allot of trial-n-error with this but didn’t see much in the way of real improvements to the Oscillation. I’ve made a log of triflight 3.2.2 (Matek F405 build) and placed it in Google-Drive for your viewing pleasure.Īs of today I am finaly able to adjust my PID’s in-flight. This is most noticable when flying strait or hovering but the effect all but disapears during turns. Allot of Low Frequency Yaw Oscillation, regardles of weather of flight mode.
0 Comments
Leave a Reply. |