Hello Bill,
what do you mean by writing:
" Don’t forget to reset the parameters before the next autotune flight " ???
do we need to reset the parameters of the axis that was tuned ?
@ZvikaF for your first autotune in each axis, I would like you to reset all the Params that I had you set in the instructions. That way I am starting with a consistent configuration. After that then you can use it to tune in any order. Just remember to cycle power between axes because of the max accel variable that doesn’t get reset. I will fix that in future releases.
Holger,
I would hold off on using it on a tandem. The pitch axis would be tuned differently than how the autotune currently works. If you want to try using it on roll and yaw, I think that would be fine. Thinking about it more though. If you want, you could run autotune on the pitch axis but stop it after it completes the sweep of frequencies before it starts changing the angle P value. This would be helpful as I would see how the gain changes with frequency.
I have uploaded the firmware for your board type.
Also, maybe Felix can go over the instructions with you just to make sure the software you use to translate from English to German was correct.
Thanks again for your help!
Hey @bnsgeyer thanks for putting this together. It’ll be really interesting to see how it turns out!
Got a 280 size machine with a cube black.
Thanks Niall! Just uploaded the firmware for your board.
Wilco…
understand you want to have each axis done with the same startup parameters…
Already updated my 450. hovered a bit in the garden, flies OK, found that recording attitude was not set to high, had some compass issues, hopefully will do all test this weekend.
All,
@picoflug @Murdoch @Steve_Mitchell @ZvikaF @DrKnow65 @LoopZilla
I have updated the instructions with the things I mentioned above and you’ll see them in a track changes format in the document. I also converted it to a *.pdf format and it is provided as a link in the text of the release message (not under assets).
Regards,
Bill
Great .
Suspect you have some typos:
HNTCH_FREQ - rotor speed in RPM ? ( It used to be in Hz )
ATC_RAT_YAW_I 0.05
ATC_RAT_RLL_ILMI 0.08 ? ( I suppose you meant ATC_RAT_YAW_ILMI )
Do NOT fly yaw axis with the ATC_RAT_YAW_P gain set to zero*
thanks. You are correct. I will get that corrected
Updated document has now been uploaded
Hello Bill,
thanks for the firemware. I think, i understand the instruction. What do you mean with:
“● Don’t forget to reset the tuned parameters before the next autotune flight”?
Which parameters you mean? And when i set back them, is the effect from the tuning then away?
Holger
Ich schreib dir in deutsch.
Der Bill meint, dass alle Werte, die du mal selbst ausgeturftelt hast, sollen wieder geloescht und die Startwerte eingegeben werden. Dann sollst du das Autotune nach seinen Anweisungen vornehmen. Natuerlich werden dann die neuen Werte gespeichert. Ich hoffe, dass Felix mal reinschaut, der ist besser im Erklaehren als ich. Uebrigens hatte ich eine Augenoperation und darf nicht fliegen fuer einige Zeit.
Ich wuensche Euch viel Erfolg mit dem Autotunen von Bill.
Holger,
As Fred said, I want you to start with the baseline values for the tuning parameters that are listed in the instructions. However once you tune an axis, I do want you reset the tuned parameters back to the starting values. I want you to tune each axis once with the other axis tune parameters at the starting values. After you gathered that data, then you can experiment with having it tune pitch axis first, then roll and finally yaw and see how the tune for the roll and yaw change.
This is the first attempt at designing this tool and the testing is more about getting data on the characteristics of the different size of vehicles and testing my algorithms.
Thanks
Bill
Hello Fred and Bill,
thanks - I think I understand it now. Now waiting for good weather…
Fred: hope you can fly soon again
Holger
Hi Bill, @bnsgeyer
Had success with the Autotune in pitch and roll. Ran out of batteries for the yaw test. Will test yaw this afternoon.
Heli specs:
Heli Size, 380
Number of main rotor blades, 2
Main Rotor Diameter in meters, 0.843m
Tail Rotor Diameter, 0.195m
Takeoff Weight,1.900kg
Rotor Speed (RPM) =2220
https://drive.google.com/file/d/1OcyAdh5FHXdT7ss0UtNwhvXU-lyatVqM/view?usp=sharing
https://drive.google.com/file/d/1dFYwMoH5EqAyXrt4w6TPRUlVSe3mSYIh/view?usp=sharing
Afternoon Yaw test flight.
https://drive.google.com/file/d/1zfFoUUnDNZ2d6lnq8809db6S9xdE2kMe/view?usp=sharing
1 Like
Hello Bill @bnsgeyer
Had three windy flights today, the helicopter was drifting too much, had to stop the autotune and continue in midair, hope you can get data from the logs.
Pitch was quite OK. I have no direct computer telemetry, only YAPPU on the radio, was looking at the messages but could not see Autotune complete, is there one ?
waited a bit after what seems like end of maneuvers, landed in auto tune, disarmed, hopping it did finish 
Roll tune was done on the second flight, the tuning process seemed more agile, at one point the helicopter started Pyro-flip, luckily was high enough to to bail out. resume few times at the same flight and landed after it seemed that the tune had ended.
Third flight did the yaw autotune, disabled the internal compass due to magnetic reading problems,
As the heli was flying all over the area, had to stop and resume the tuning midair, hope it gives some valid data.
Heli Size: 450
Number of main rotor blades : 2
Main Rotor Diameter : 0.72 m
Tail Rotor Diameter: 0.16 m
Takeoff Weight: ,0.98 kg
Rotor RPM : about 2200, check the RPM reading
My manual tuned parameters:
https://drive.google.com/file/d/10rAXYOZXtDTguzu_m0psESWy7jQPHcY-/view?usp=sharing
My Initial tune parameters:
https://drive.google.com/file/d/1DMIMteat07uFdCKQg-Cj76D1Lv2Aerke/view?usp=sharing
First flight - Pitch
https://drive.google.com/file/d/1uNJ2lArg6gB2K47-tMwOfJym8GODNneT/view?usp=sharing
Second Flight - Roll:
https://drive.google.com/file/d/1376TwotOOGXX2fqJSiQqby8yo_Si4MUT/view?usp=sharing
Third Flight - Yaw:
https://drive.google.com/file/d/1TMVT3K_qbnUycoDlntfaENMcIr7ktJr4/view?usp=sharing
P.S: Mission Planner parameter change ATC_RAT_PIT_D range warns: 0.001 - 0.80
@ZvikaF so we’re you able to use your aileron/elevator stick to keep the aircraft from drifting.
That’s odd. I will check that. Did you enter autotune from althold or stabilize? If you enter from loiter then you won’t be able to control drift and the position controller will reposition the aircraft between oscillations. And will yaw the aircraft to keep it pointed into the wind.
I appreciate the data but I would like you to do it again during a low wind day. I will look at this data but I won’t be able to look at it until tomorrow.
Thanks!!!
Hello Bill
Thank you for your contribution and hard work,
I entered from stabilize mode. check second flight at about 12:36:41
Had some time to review the data and some of your source code, seems that on packet ATNH Tune step is always “4”, but data shows clearly the oscillation command.
Looking to see if YAPPU telemetry code can be changed to alert me on the Autotune completion.
Will try again in calmer day.
Steve,
Had a quick look at your data. The flight characteristics of your aircraft is different than my 600. I will have to look at it more and potentially change my tuning strategy to be smarter and detect differences in the frequency responses to decide how to tune. Any reason why you only had Rate P gain tuned for yaw axis. I would suggest tuning your rate d gain and then run the yaw autotune
Regards
Bill
Bill,
I thought that might be the case after seeing the ATC_ANG_RLL_P reach the 10 limit you had set. It was my fault (finger trouble) for setting the yaw D gain param to zero and not double checking it before testing. I have to also admit that I’m using the default tune param’s on the tail. It has worked well, so never bothered to tune it. Here is another yaw autotune flight log with D gain set to 0.0025.
https://drive.google.com/file/d/1vgF2VF_ZJ7JBGVzPPBbBytB_p87WawVp/view?usp=sharing
ZvikaF,
The YAPPU telemetry displays “Autotune: success” and in the next line below that is “Pilot override x4”. Unless you are really looking for it, it’s hard to see. I found you can hold a a very small amount of cyclic to reduce the drift without the pilot override kicking in) and the tune happens way faster. As Bill said, low wind day is best. Had one tune take 14 mins and then the next less than 5mins. The less input you do the faster it will tune.