Copter-3.6.10-rc1 available for beta testing

when i select geofence, i can select also select the actions in case the drone will risk go over the virtual fence… can be rtl or land, can be add also the option hold mode.
Dont mind about min and max parameter for the height, I’ve a wrong type

Motor Test does not work in Copter-3.6.10-rc1,

I see the same thing. The FC disconnects from Mission Planner when a Test motor is attempted and the error is generated.

@winstonsaz, @dkemxr, OK, thanks for the report. I’m not seeing the problem with my vehicle but that doesn’t mean there isn’t a problem of course, just that I can’t reproduce it.

Can I ask which flight controllers are you using? I may need to ask if you can provide a dataflash log (after setting LOG_DISARMED = 1)…

I’m using a PixRacer on the vehicle I tested.

1 Like

I also use Pixracer.

1 Like

I’ve tested it on my pixracer and it seems to work. I’ve noticed that it does cause some kind of delay in the main loop which leads to the EKF becoming unhappy but the same issue appears to be present on 3.6.9 as well.

Could either of you provide a log?

@dkemxr @winstonsaz @rmackay9 - I was not able to duplicate this issue with a Pixracer using MP 1.3.7105.26478 on Native Windows. I wonder if that error is related to running is virtually on a Mac? Are you willing to share your parameters so I can try to duplicate this issue locally?

1 Like

I noticed a change in attitude after having installed copter-3.6.9 and Copter-3.6.10-rc1 in my Pixracer, then I went back to copter-3.6.8 and now everything is fine.

@winstonsaz, Is this a log file of the motor test issue? I’m quite keen to get to the bottom of this…

I believe the attached is from an attempt at Motor test with Logging disarmed enabled. As before as soon as Test Motor A was pressed it disconnected from MP and posted the communication error.
2019-07-16 21-41-41.bin (272 KB)

1 Like

@dkemxr,

Excellent! thank you very much for the log and it clearly shows where the issue is. It’s with the IDWG (independent watchdog) which is incorrectly triggering during the motor test. @tridge and I will come up with a fix and include with 3.6.10-rc2. Thanks!

EDIT: by the way, when the watch dog triggers it will write these “MSG” lines into the log the next time the vehicle is booted up.

MSG: Baro: skipping calibration
MSG: Restored watchdog attitude -1 0 111
MSG: ArduCopter V3.6.10-rc1 (4cd315aa)

@winstonsaz, no need for the log file now.

@Randy, sorry for the delay, I just read the message.

1 Like

@winstonsaz, np, thanks… a new release candidate will be out (probably next week) with a fix.

1 Like

@rmackay9, I’ve tested it on my pixracer and it seems to work. I’ve noticed that it does cause some kind of delay in the main loop which leads to the EKF becoming unhappy but the same issue appears to be present on 3.6.9 as well.

This is also due to the same cause?

@winstonsaz, the EKF becoming unhappy is apparently a long standing issue. We’ve got some kind of delay happening in the motor test which is upsetting the EKF. I plan to fix this but it’s not related and has been there for a while.

1 Like

I’ve found and fixed the MotorTest issue and pushed out 3.6.10-rc2.. Thanks again for finding that!

not had the chance to test the new build rc-2 on the Zero yet will do it tomorrow if i can but i noticed the there is no tone when disarming the flight control safety switch not a huge deal but nice to have the tone.