@bnsgeyer thought to make a video to help users have a view of what it looks like:
SID_AXIS: 7 & 8
Interestingly pitch axis was spot on, while roll axis showed tune can be improved a little.
5 Likes
@Ferrosan, thank you for the video!
I have a question about SID_AXIS.
@bnsgeyer recommended SID_AXIS: 1-3 (Input [Axis] Angle), @Allisterās lua (at least v1) used SID_AXIS: 4-6 (Recovery [Axis] Angle), and you used Rate [Axis].
As long as it includes the axis of interest, are any of the SID_AXIS parameters acceptable for the System ID flights?
Good catch. Yes, I used 4-6 but I was testing on a multi rotor. Looking at @bnsgeyerās comments you should use 1-3 for a helicopter. Itās a quick edit in V2 of the script. I can update one if you need it.
Added some notes in the script so itās easier to edit for however you want to use it.
2 Likes
What is the easiest way to not use the āmagnitudeā part of the script so that it can be set maually with a tuning nob?
You could use transmitter based tuning
I donāt think youād want to adjust the magnitude during a sysID run, but certainly between runs would be okay.
In the script you could leave it, or comment out the lines that change the magnitude.
Thanks for fast reply Allister.
Sure, the plan was to adjust it between tuning sessions. It is nice to be able to quickly land and adjust the magnitude if set to high.
I tried to comment out lines 78, 84, 90 and also lines 27, 30 and 33. I am not sure it worked.
That should work fine
That might return an error when the text message is sent because the message uses that variable.
If youāre not in a hurry Iāll re-create my original version that does not have the magnitude control in it. Itāll be a few hours before I can get to it though.
I just posted a new version in Github that says NoMag. That should be what you want. I ran it in the sim for a few seconds, but obviously this quickly it hasnāt been flown.
1 Like
That was quick 
Thanks Allister. I will give it a go tomorrow
line 83 and 89 still mentions magnitude. should these lines be in the script?
I deleted the lines and it looks like it works now, on the bench at least (did not work with the lines)
Sorry about that. Should be fixed now on Github.
1 Like
I tested the āno magnitudeā script yesterday and it worked fine combined with a sepatate tuning nob.
1 Like
Bill, would you recommend doing this for the yaw axis as well, or leaving the tail at initial suggested tuning parameters (ATC_RAT_YAW_P: 0.18, ATC_RAT_YAW_D: 0.003)?
No, leave the rate P and D gains as they are. Otherwise, you wonāt be able to control the yaw axis.
2 Likes
Bill, I have yet another question: In your video you mention setting ATC_RAT_X_FLTE to zero if possible.
How can we tell if itās safe to take it to zero, or if not, what value to lower it to?
Thanks again!
@bnsgeyer I finally got back to this. OMP M4. Iāve taken a stab at pitch and roll parameters from the SysID flight. Can you have a look and give me some feedback before I fly these parameters? Thanks!
M4_40_Roll.param (575 Bytes)
M4_40_Pitch.param (579 Bytes)
Hi Allister,
I think you did a pretty good job on Pitch. I took a slightly different tactic. Using the Notch filter, you can reduce the peak and still be able to increase the Rate P gain a little more. Here is what response looks like and the gains
pitch_allister.param (646 Bytes)
I thought roll was tuned pretty well too. I made a few adjustments. I would pull back the FF slightly to 0.11. That will bring down the rate controller below the 0 db line but more importantly it pulls down the magnitude line down for the attitude controller. in many cases the attitude controller magnitude will stay above the 0 db but just want to keep it down. The other thing I did was increase the Rate D term to 0.0003. it helps flatten out the magnitude.
1 Like
Thanks! Iāll give these a try and let you know how it goes. I appreciate your time!
1 Like
On my largest and lowest head speed heli is the only one where my sysid runs look and react like the example video to tuning.
I have gone through a round of using default params and getting the sysid logs then making changes and now flying again on those changes with another sysid run set.
here is a link to the sysid run with all 3 axis to check https://drive.google.com/file/d/1vM3PoAvhJVtVKY8oTZtEEpAmwWbc8cqJ/view?usp=drive_link
This is a 380 size heli that weighs about 1.5kg with the bat I used for the runs.
I have already made some adjustments and saved another set of params based on this check to fine tune it. I ended up pushing ANG P down even more.
My other 2 helis are both smaller and the sysID run logs for those are much more wonky but I wanted to ask if this was even headed in the right way to check that I am understanding basics at least.
I am finding I am having to lower my ANG P not raise it on all of my helis (from default 4.5). Is this indicative of some problem? Can it be lowered all the way to zero? For copter it appears the min is 3 but I am able to set it lower. I donāt know if there is an issue with this however.
Other things that im not sure about are having no or low D if its a problem.
Is there a sorta of min or max frequency where we just dont care about the tune beyond or is it mostly just based on the continuity?
1 Like