1 April, 2020 at 10:55 #65259
Hey all, I have an OG mini-tri with the F1 board;
I took it all apart and built a regular sized tri with it, but decided I wanted to go back to the mini and do a bit of racing vs. my quad mates.
I did the servo feedback mod to my old servo and running Tri-flight 0.5 b3 (with what is the default baby tri PIDs) and it flys very well.
The only issue i’m having is a slight ‘drift’ when doing moderate acceleration or adding throttle to hover, the tri tends to yaw right (tail goes left)… its manageable through FPV but still annoying.
I have tried using the PIDs on the mini-tri product page, but they are from an older version of Tri-flight and have lines of code that are redundant.
So wondering if anyone has any tips or maybe even F1 mini-tri cli-dump for Tri-flight 0.5 b3?
I’ll get a video tomorrow! but for now; here’s a photo.
- This topic was modified 1 month, 4 weeks ago by Matthew.
Attachments:1 April, 2020 at 11:30 #65262Kevin_ErikModerator
Best way to see whats going on is to use a Datalogger then create a Step-Response plot. (PID Toolkit or Toolbox) A good example is the Nano-Whirlygig i tuned just recently: https://rotorbuilds.com/build/21545
This is using Betaflight 4.1.2 and motor filtering on an F4 flight controller. Scroll down through the images on that link and you’ll see the Step-Response plot for the copter. Ideally the plot should have a sharp rise and then flatten out at “1”. This indicates that the copters response to input matches what the Gyro senses. In other words, there is minimal delay or overshoot.
Anyways for the older Triflight theres a command called “thrust factor” that manages the copters Yaw response as throttle is applied. Increasing this value increases the amount of counter torque keeping the copters orientation locked in as throttle increases. That being said, the same can also be done with Betaflight 4.1 (or greater) by managing things such as TPA as well as the other factors that manipulate I gain. Unfortunately, you’d have to upgrade the flight-controller to an F4 to use the current version of BF. While that sounds daunting, the filter improvements alone make it 100% worth doing. (Motor filtering is awesome)
Be aware, that D-term and Yaw don’t work well together due to the D-term filter delay making the already slow Servo even slower. This delay causes undershoot, which is where the Servo will continually hunt to reach the correct position but is unable due to lag. Effectively the control inputs are coming faster than the Servo can possibly keep up, making the Yaw feel mushy and unresponsive.1 April, 2020 at 11:59 #65264
Very interesting response re; data-logging -> I haven’y delved into data-logging with multi rotors or this hobby in general as i like to keep it simple…
I’d happy update the board, but I have the AIO David made back in the day and until I snap it, i’ll happily keep using it, I’m far from “the latest and greatest” type of guy out there, and something that I can’t out-fly keeps me happy.
Yes, i recall when thrust-factor was implemented, I will have a look at it tomorrow, I sort of feel even though this happen when throttle is applied.. it happens not as a reaction to throttle directly…
I’ll have a look at the PIDs and thrust factor tomorrow, the yaw, especially at higher rates is more than what i need, so I’m happy to drop some I on the yaw and soften it if need be, even though it doesn’t hunt as such as it is.1 April, 2020 at 14:13 #65265Kevin_ErikModerator
Its cool, Davids gear is great.
However, your going to find it harder and harder as time goes on to find replacement parts or tech help. For that matter Tricopters are likely to become rarefied as fewer people fly due to the recent FAA regulations making it much harder to keep the larger / heavier Tri’s airborne. Which is in part why i designed the Whirlygig Tricopters to be small enough to fly anywhere.
***Just a few points on Yaw tuning***
– Filtering can make or break a Tricopter tune. Too much filtering adds delay that compounds any Yaw issues you may be experiencing. Ideally, you want as little filtering as possible, which is why data-logging is so helpful to figure out what is really needed. Then again, if all else fails, switch to Betaflight 4.1 (or higher) as the motor filtering is awesome-sauce!
– Servos response time can be drastically improved by increasing the input Voltage. While most Tricopters use 6V, I suggest using 7V if your external BEC and Servo can manage it. (MFR recommendations) Case in point, I run 7.4V on my Whirlygig Tricopters and its incredible just how much better it became.
Note: A word of warning, never, never and i mean never, power a Servo from the Flight-Controllers on-board 5V BEC. Not unless you enjoy replacing the Controller and Servo often.
– Dterm is designed to detect and prevent P overshoot yet requires significant filtering / processing to achieve. (Thus allowing P to be raised higher) This is why the response delay (lag) can be so detrimental to Yaw. In my experience, its best to fully tune the Tricopter’s Yaw is by using step response mapping (P & I) and then add as little Dterm (if any) to help flatten out the response. Obviously, the newer firmwares handle this much better and thus experience much less delay as a result.
Note: The worst ESC on the market is still 100 times faster than the best Servo. This is why Dterm is great for quads but not for Tricopters.
– Too high of a Yaw P or Iterm results in Servo overshoot. In a sense the Servo will rotate farther than needed, resulting in the Flight-Controller making a correction, then another and another etc. This can be seen as fast Tail wag (oscillation) and can occur both in flight and on the bench.
– Too low of a Yaw P and Iterm results in the copter not Yawing enough. Thus the Flight Controller may need to Yaw again and again to get close to its Target position. (Thing slow wag) Obviously, this takes time and will cause the Yaw response to both feel weak and mushy. This can also cause the copter to lose its orientation lock on its current position.
– Pterm sets how aggressive the Servo response will be. Ideally, you want the Servo to “snap” into position as soon as an RC input is given, hopefully without overshooting (wag). Best way to test this is the set your I-term to 5-10 and D-Term to 0, then hover over one spot while flicking the Yaw left and right. the copter should feel very responsive but not wag. If not, increase the P-term until it snaps or decrease it until wag is gone.
– Iterm is meant to keep the copters orientation locked until an RC input is given. Thus in a windy situation, the copter may float away some but still face the same direction. Ideally you’ll want to increase Iterm a few points at a time until you achieved this “Locked” state. Then go flying fast / strait lines to see if it still holds orientation. Here again, slowly add Iterm until you can throttle up hard with minimal Yaw deviation. Be aware that as you increase Iterm that you may need to reduce Pterm to prevent wag.
– During tuning it is advisable to disable all functions that can automatically retard or amplify your PID’s in flight. These functions / algorithms can make the copters behavior seam erratic and thus make it very hard to tune properly. Stuff like Anti-Gravity, TPA and the like should all be disabled until your happy with the tune.
– Once the Yaw is good, tune your Roll and Pitch just like you would a quadcopter.11 May, 2020 at 08:37 #65427
Thank you for your help -> for anyone interested… my issues were fixed by playing close attention to my tricopter and tweaking these parameters;
especially min throttle -> which I turned up to 300.
I am also currently selling my tricopter -> will ship worldwide.
- You must be logged in to reply to this topic.