[hitewill]

I have output tab set at what the cool kids are setting theirs at, 0 min 40 neutral and 2000 max,  my throttle #s on the input page 1000 min, 1040 neutral and 2000 max.   Everything spins beautiful when using the "test motors" check-box.  when I watch the readings under Data Objects>Actuator Commands, those #s follow right along with the #s on output page.   Everything goes to hell after arming with radio though, watching Actuator Command when my throttle stick is just up from min, those #s read as "channel 0 40us, channel 1 is around 900us and climbing same for channels 2 & 3, those 3 channels continue to slowly increase until I push the throttle stick back down to 0.  Ive tried the manual way of calibrating the esc's using "home and end" as stated in the  How-to section, I have gone back and started from scratch and rebuilt using the setup wizard enough times and the only weird thing that happens is if you try and redo the radio setup wizard AFTER dshot has been picked and saved as soon as I close the "changing to always disarmed dialog box"  the motors start turning at around 1000 or so and the only way to stop them is pull the battery or check the "test motors" box and manually 0 the 4 sliders.   it seemed to me for whatever reason or error on my part 3 of the 4 esc's fail to calibrate a neutral.  I have tried using the blheli_s and did a fresh flash and write.  Im out of ideas short of picking up some new esc's and see if that fixes it, Just thought I would see if any of you guys have any words of wisdom. 

DYS XSD 30A

[f5soh]

Hi,

If you talk about 0 - 2000 range for output i assume you are using Next branch and DShot output ?
DShot output do not need calibration, just neutral adjustment using the "Test outputs".

Wizard do not support the Dshot for now so you need to setup manually in Output tab and/or adjust the Dshot rate in System > UAVOBrowser (search for Dshot)

When armed and you apply Throttle the behavior you describe seems ok, the quad is stabilized and PID loop try to correct/compensate bank/yaw errors.

[hitewill]

The librepilot revision (HEAD detachee sur bitbucket_vladimir / LP-479_dshot):50fce728
Built on Apr 25 2017

so if that's the next branch then yes.   I will take it out for a fly maybe I'm stressing over nothing.  when doing the Mag calibration.  Can it be held in the required positions or does it have to totally still and stationary like with the accelerometer calibs. 

AHH I just totally visualized your response, when I give it throttle the fc is trying to stabilize itself in what it thinks is level, or stabilized, but without props its not getting the results it expects so the rpm's continue to climb trying to stabilize itself.  I will do a fresh setup and calibration just in case.  Then go for a test flight

Thank you

[Mateusz]

Replying to your question about Mag. Mag calibration is only needed if you intend autonomous flights with GPS and requirement is additional external Mag. When it comes to calibration, it is continuous from the beginning, so it does not matter how you place it, these positions are just guides to cover all mag directions, you can also ignore them skip to last 6th step of calibration and rotate it slowly in all possible directions than save. That's what I do. Note that for autonomous flight (INS13) you need all sensors with no exceptions giving good data. Other sensor fusion algorithms like Complementary filter is for racing and uses just gyro and accel. Remaining options are for diagnostics or for  hardware that has missing sensors. You should make copter flying perfect in Complementary before attempting INS13.

[f5soh]

The librepilot revision (HEAD detachee sur bitbucket_vladimir / LP-479_dshot):50fce728
Built on Apr 25 2017

This is already a old branch.

Here is latest Windows build, LibrePilot-16.09+r423-g1a4bb0c_i686.exe :
https://drive.google.com/open?id=0B5d_9OcqQj8BNEVya0NxQVo1NDQ