Hello azaz44,
good work you made ! But this shouldn't be necessary in my opinion, why should some channels have different PWM-Center ?
That really can't be true, there must be something else to blame, the whole thing is the same for all channels! There is no logical reason why there should be differences here - I don't really understand why Tomas doesn't take a position on this either, there is obviously a contradiction here and only he could tell us what could be the reason, he obviously is responsible for it a few channels the correct zero point defined by firmware - Tomas, we absolutely need your support and you can't ignore it!

Regards
Matthias

you've tried adjust the stick with the screws?
In radiomaster tx16s mk2 max I've h ad the same problem and it depended on the fact from imperfect mechanical centering.
After Re-calibrate the radio and only after in spirit you should have perfect 0.

Hello Franz88,

Horus X10 S Express (ACCESS):
The idea of calibrating is not bad, but I had calibrated it several times before and there are still differences -
What I can say is that aileron, elevator and rudder is at 1500 us without subtrim or center PWM...that works as it should !
Throttle and pitch is different, with different subtrims at the moment - throttle has 4,6 % center subtrim and pitch has 1,5 % center subtrim regardless of whether center subtrim is the right way or the wrong way...and that's really funny
Gyro has also an different center subtrim, here 2,7% regardless of whether center subtrim is the right way or the wrong way...
Where these differences come from is beyond me, especially since the center subtrims are not identical -

I'll definitely try to recalibrate again, if it doesn't work then it's weird -

Additional information:
In my Radiomaster TX16S (ACCST) (Edgetx 2.83):
No center subtrims in any channels with one exception in Gyro (2,9%) - otherwise everything fits here-
But what is also different are the different min. and max. values for reaching 100% in Spirit Settings...

Regards
Matthias

this difference I believe it occurs why these radios with edge tx (and betaflight) they do not have 1000 usec precise at minimal sticks and they not have 2000 usec precise at maximum sticks.
Other brands yes.
but you have so many ways to customize with this radio.
First try the calibration
But you are sure which you are not confusing the values ​​of the mixer with the values ​​of the sent signal?
If you view the usecs what values ​​are displayed at a minimum, at center and the maximum?
The spirit reads the received usec from various channels and she doesn't care mixer and other functions of radios
Make a video or photos where you can see the usecs

Hello Franz88,
in my Radiomaster are the AG01-Gimbals (hall-effect) in my Horus are also hall-effect gimbals, so there should be no difference - Horus-gimbals have an 12Bit overall resolution resolution - AG01-Gimbals I think they are also 12bit...
And no I have nowehre an additional mixer which can distort the values and only linear curves at the moment (-100 0 +100) -
Yesterday I have calibrated my Horus again, we will see latest at the weekend -

Regards
Matthias

As for me, I don't want to complain. Spirit is a great FBL, I love it, and there were maybe some reason what it's done this way. Or maybe it was just a bit messed up a bit at the beginning and it could not be changed later? This happens, and fixing such thing now, when there are many users with configure models, is not an easy topic.

For me it's fine if I find some logic in it, and some set of numbers which work and seem to make some sense. So I have confidence that it works repeatedly, same all the time, and same in every model. Then I can clone models in radio and have confidence that clones will work same way, with different Spirit units and different receivers (I have 7 helis to setup on Ethos, a lot of work..).

I also need some confidence in using "Special functions", so for example when you control "cyclic rotation speed", "cyclic gain" etc. with channels. Many parameters in Spirit are integer numbers, like Flight Style, rotation rates, cyclic feed forward etc. And with integer number it's risky and any small offset might change a lot. Maybe you setup everything and get "cyclic rotation speed" 12 in one model by sending some value from the radio. But maybe it was just a tiny bit close to become 13 and tomorrow it becomes 13. Or in another model with another RX it becomes 13. This because of having no idea what these numbers depend on. So far I was checking each parameter in each model to make sure mapping from radio values to integer values works, but this is a lot of work.

I think with investigation I made yesterday I'm close to understand how it works. I was able to get numbers which seem to have some "rules" instead of being random numbers. At least min/max ranges are symmetrical, and there are similar settings for some groups of channels. I'll spend a bit more time on this (it will pay off when setting up all helis later), to maybe try to come with unified PWM for all channels (except of pitch and gyro), and also find some consistent value for throttle. And I'll retest to check if results are same in other models.

And as of stick calibration, this is out of question in my checks. I force channel values in radio when calibrating outputs, to take analogs out of equation. So I force the value to be -100 or -50 or 0 or whatever I need.

Analogs are a different story and of course they have to be calibrated right.

And BTW I use spring loaded throttle in my radio. So my throttle / pitch is centered. Which is why having some weird offset in pitch isn't really great for me.

go here

above exit the values usec at a minimum, centerx, max stick.
Can you show me the 3 positions with a photo for example take the values ​​for aileron and elevator.
My radiomaster tx16s mak 2 max mount the ag01 and I REPEAT YOU that I had to move the screws why some sticks they were not centered mechanically.
This type of stick it has the adjustment screws on the front and rear.
But these inaccuracies with re-calibration they adjust.
changing from mode 2 to mode1 i had the same problem in the center stick and after I solved
This radio does not send 1000 usec at min .
Let's go step by step and for now let's see the values ​​from that menu that I showed you

I improved stuff and tried to unify it even more. Here's final version. It makes things as consistent as possible. I don't think it's possible to get it better. It seems pitch needs 7 us offset comparing to Rud/Ele/Ail, no matter what. I also checked throttle and it seems it needs different PWM. So there are 3 different PWM values and 3 different ranges.

I tested it in 2 helicopters
- FrSky Archer RS + Spirit GT
- FrSky Archer GR8 + Spirit Pro

And results are consistent. I have more helis to test and setup but I think this set of numbers will work everywhere.

Hello,
now it works for me too without any problems - I calibrated again my Horus and made then nothing other then jjuk said, set the midfrequency to 1520 and then make your min/max points and at last if necessary a little correction of the mid frequency for example from 1520 us to 1522 us to get "0" in the middle-
Now all values are 100% reliably even after multiple starts -
For me makes it now sense because most servo midfrequencys are at 1520us - there is nothing wrong !

Thanks a lot to jjuk, that was the point, nothing more !

Regards
Matthias