I recently was testing the copter 4.5 dev version on a 6 kg AUW multirotor. I’m already using the same firmware on other birds of same setup and they are flying stable and good.
I had done the auto tuning as per the process and did couple of flights as well. The bird responded very well. The crash happened when I did an auto flight.
The bird took off weel and did well till the 4th waypoint which was also the last. From there it started descending down in a roll. The logs show ErrYAW as high and the motor 1 saturated. I would like to get some insights into this issue as this is something that I was least expecting, and I currently have 4 birds that run in the same firmware. The link of the param is given below.
I’m going through the blog now. Thanks for the info. But I would like to really know the issue here. This crashed happened on 4th flight after auto tuning.
Thrust loss on Motor 1. It’s not clear from the log why. Could be motor, esc, broken prop.
There is output oscillation before the thrust loss perhaps something overheated. Tuning could be better. You are using Dshot so I would suggest you flash the BDshot version of firmware and configure it to have ESC RPM available and then configure the Notch filter. Then run Auto Tune again. The notch filter should be configured before running Auto Tune in any case. Set this:
INS_ACCEL_FILTER,10
And read this Setting Motor Ranges as your MOT parameters are at default.
Thanks Dave for your suggestions. really appreciate. I’ll doing the settings.
I also have a doubt on the capacitor used. I use a 6S 30 Ah Li ion and the capacitors of 50V 2700muF on PDB and 50V 330muF at ESC side. Is there something I need to rectify on this side as well?
Did you ever find a solution? I think I am having a similar issue - the motors and ESC are fine with a very short cable between the battery and ESC with two capacitors. My YawErr is always at 1 and all the motors are pulsing between arm spin and full power which is causing them to get hot quickly. At this point I have no idea what to do so if you found anything that worked please let me know
I’ve been fiddling around a bunch to try and fix it and I am 90% certain the tune is not the issue as I followed the initial tuning flight documentation to the letter. I’ve just taken on Dave’s advice and looked at my motor parameters and I believe the issue is my MOT_SPIN_MIN was set far too high. I was initially expecting a however throttle of about 35-40%, but auto hover has put it at 25%. You can see in that older parameter list that the SPIN_MIN was 18% which I though would be half, but it turns out it was far too much and I believe that was conflicting with the tune. It is now set at 10% with the arm spin at 5% and using DShot 1200 to communicate with the ESC. These parameters have been effective in the past on my FPV quad so seem like sensible initial values. I did a quick hop to test and the motor output noise has gone way down, so soon I’ll head out to a safe location for a proper flight test. The weather conditions aren’t ideal today but I should hopefully get at least some data and will post it when I can.
I went out and did some small test hops in a sheltered area. It seems as though the reduction in MOT_SPIN_MIN has helped a little in the initial flight, however the instability quickly returns. Motors 3 and 4 (CW) are too hot to touch which clearly indicates they are drawing much more power than the CCW motors. I am assuming there is instability in the yaw control which is the root cause of the issue, however I am struggling to identify what exactly the problem is. In earlier testing the internal compass was problematic so I turned it off, and in operation the drone is expected to be around large metal objects so it cannot be relied upon for yaw data. If you have any other ideas, please let me know.
Here are some logs from today: 00000015.BIN (652 KB) 00000016.BIN (664 KB) 00000017.BIN (680 KB) 00000018.BIN (676 KB)
There were minor changes made between each flight. 15 had the new motor spin values, 16 decreased the INS_ACCCEL_fliter from 20 to 10 Hz, 17 had a small but not noticeable change to the min spin value, 18 decreased the yaw p gain and filter frequencies, then 19 increased the yaw frequencies back to where they were before. After all that the performance is still bad and I’m confused
I suggest you start over from default and follow a configuration and tuning guide. Some observations:
You have the Bdshot version of firmware flashed but your outputs are configured for PWM. You didn’t see this?
Don’t configure for Dshot1200 use 600 and configure for Bdshot. What ESC’s do you have?
Your Notch filter is doing nothing now as it’s configured incorrectly. Configure for RPM reference once you fix the above.
I didn’t see that, when I saw that I flashed the bdshot firmware I assumed that the motor outputs would be configured for dshot. I have since fixed the notch filtering based in what the documentation but I haven’t had the opportunity to flight test it yet. Looking through the logs there was some noise at about 5Hz on the rates which pushed me to double check. I’ll reconfigure the motor outputs tomorrow and hopefully that’ll get everything sorted. Thank you so much for the help!
