Hey! Did a first test flight today - saw an weird issue and cant seem to figure it out. I was flying north and the drone did random yaw right turns. When I flew east i did not see that issue.
If i recall, then the wind was east-west.
Flight log - Unique Download Link | WeTransfer
There seems to be a big torque split between the CW and CCW motors. This will cause yaw and general tune issues. Get the blade tips perfectly lined up to reduce the split between the motors. Millimeters mater for this, so be as precise as you can.
The log also shows points where one of the motors is hitting min values. Any time a motor becomes saturated (min or max) you will loose control or have a moment of instability. I suspect that this problem will be addressed once the mechanical issue of the torque split is addressed.
There are also very high vibrations on the Y axis. X and Z look good. Make sure there aren’t any wires or cables bouncing on the controller or creating issues there. Look at how the antennas are mounted and that they aren’t flopping around causing problems.
Set:
ARMING_CHECK,1 #If it won't arm with "1" then figure out the problem.
PSC_ACCZ_P, 0.24
PSC_ACCZ_I, 0.48
There are some very unorthodox tuning parameters, but not worth addressing until the mechanical issues are fixed first.
Edit: I just re-read that this was a first flight. Reset the tuning parameters to default. First flights should be done in stabilized or Alt-hold. POS-HOLD mode is generally replaced by loiter but shouldn’t be used until things are set up better.
What tuning guide did you follow? I could take a guess based on some of those unorthodox parameter settings.
@Tangoviktor stop wasting your time and start using AMC.
Thanks. Do u think that the distance between propellers is off by couple of millimeters on the ccw and cw or could or be that since they are folding props there could be some play in the prop assembly between the shimmy rings? I did have to loosen them so that they would spin more freely
Most of the tuning i have done are pid tuning only and rest are default parameters. Only the atc parameters are gotten from autotuning without a gimbal and then the gimbal weight was added.
What are the unorthodox parameters?
Rate PID’s. And we usually see the RPM parameter misconfiguration after someone has watched a series of YouTube videos containing mistakes. You won’t see that in any Arducopter Multi-rotor tuning guide.
Are you sure of the Motor Range setting?
Autotune did not produce those values.
When I level the motors I actually go from the base of the motor shaft, not the prop tip. Takes a bit longer but it’s most accurate. Might be worth it if you have folding props because they add another variable in the pivot that will make leveling difficult. Folding props however won’t cause the torque split, that’s the frame itself.
Search on the forum and there are plenty of comments from people using folding props and them having issues with vibrations.
So what i found out today is that the arms both have an inwards tilt to them. I am trying to understand where that came from, but it seems, that front right motor might be up to 2mm higher at some point.
Just one of several issues on this craft.
What about this log? this is the same airframe, but should have a different tuning.
I will try to do the methodical configuration tomorrow. This drone has custom hardware so it has no cables other than motor cables inside the arms. Could it be that the MJF gets printed crooked?
If i fix the alignment issue and do a new tuning, what other issues do i have on it?
I don’t see much different in tuning. You still have out of whack Rate PID’s. One result:
The Motor Ranges are probably not right.
You still have RPM parameters configured. This is a misconfiguration that I suppose you got elsewhere.
Error rate on the ESC’s is high:
I would start over with the Methodic Configurator and use the reset to default option.
Thanks. Will do that. What causes the esc errors?
Could be a few things. Disable all the RPM parameters you set.
What ESC’s are they and what protocol?
I use dshot600 and bluejay. Esc hardware itself is custom
Try a different Dshot rate starting with the next higher.
The ESC errors are gone. Most likely because you disabled the RPM parameters. Or it could be the loop rate.
The tune isn’t that great and the Motor Ranges are at default. And perhaps the same place you got the RPM parameters the Batch Logging parameters are misconfigured. In fact with that Flight Controller don’t use batch logging at all. Set these
INS_LOG_BAT_MASK,0
INS_LOG_BAT_OPT,0
INS_RAW_LOG_OPT,9
I would set the Rate P&I values equal for Pitch and Roll. Set the Motor Ranges properly. Make a hover flight to collect data to review the notch filter. Re-configure that if required and then run Auto Tune.
Surly you got some of your tuning process’s from Rossers’s videos?
I did look at the video, but i did not follow that.
I have not specifically controlled the RPM parameters. The only think I did was enabled BLheli passthrough.
I also tired out setting the motor range - all of the motors behave simmilarly at 5% of
So thats why it seemed that default values worked for me.
I also noticed that the log i posted is missing a lot of the flight. Here it is again.
Then why do you have MOT_SPIN_MIN at default which is 15%? You didn’t follow the procedure.
