Copter-3.6.0-rc5 available for beta testing

Copter-3.6.0-rc5 is ready for beta testing. The list of changes can be seen in the ReleaseNotes and also copied below.

Changes from 3.6.0-rc4

  1. Support new boards CubeBlack and HolyBro Pixhawk4-mini
  2. ChibiOS small enhancement and fixes:
    a) make board serial number match NuttX
    b) relay fix for Pixhawk boards by defaulting BRD_PWM_COUNT to 4 (same as NuttX)
    c) MatekF405-Wing default compass orientation fix
    d) Pixhawk4 default compass orientation fix
    e) Omnibus F4 Pro I2C fix
  3. Bug fixes:
    a) Telemetry baud rate fix
    b) RC failsafe fix to RTL in all modes except Auto when FS_THR_ENABLE is continue-with-mission
    c) waypoint navigation related safety check that PSC_POS_XY_P is non-zero
    d) aerofc-v1 update bootloader utility fix

There will certainly be an -rc6 to resolve some known issues including:

Thanks in advance for any testing people here can do!

If possible please raise issues as a new Topic in this thread.

1 Like

Any reason to expect different compass behavior with this release? MP reported “compass not calibrated” after updating from Rc4 (chibios both). And the 2nd on-board compass on the PixRacer is not recognized when doing the calibration. I think this can be seen in the device ID’s comparing them to the Rc3 release which is the last I have a saved parameter file from. They are identified differently.

1 Like


Seems like buzzer sounds for pressing safety switch are gone.

Pixhawk 4


mRo radio does not connect using ChibiOS without the wire removal.

Frame: HEXA
fmuv3 001A0034 33365109 36353335
ChibiOS: e7b6334c
ArduCopter V3.6.0-rc5 (2301b196)

Hi I like to install using load custom firmware can you give me the direct link to the directory for px4-v3 rc-5? I am lost. normally it in the latest directory? but the firmware does not say RC-5 after loading unless I use MP load beta. Where is MP pointing to?

I think MP temporarily downloads them each time, should be here :

Ok that did it. I see RC-5 now. Thanks!

Firmware site, path: /Copter/beta/PX4

ChibiOS rc-5 on the mRo in gps mode is jerky see video: GPS switched on at 00:13

" "

logs ChibiOS and Nuttx nuttx is fine.

1 Like

Downloading logs in RC-5 gives false flags errors. Nav check, Bad ARS, and Pos vert in HUD.


That is a bit of a concern. There was a change related to the compasses. It was meant to be a non-functional change but perhaps something went wrong. Txs for the report. I think we should be able to reproduce with a pixracer but we may need a dataflash log file from you…

Txs for the report. re EKF errors during dataflash log downloads, I think that happens in Copter-3.5 as well. It’s potentially something we should fix but we may not fix it in Copter-3.6.

Flight log attached from a brief hover as it wasn’t stable in assisted modes.

Edit: I flashed back to 3.5.5 Nuttx as it wasn’t flyable on 3.6.0Rc5 Chibios and the compass’s are properly recognized and calibrate OK.


Nice video and logs of the problem. I’ve created a new thread here to discuss.

1 Like

@Quadzilla many thanks for reporting the issue with the mRo X2.1. I’ve reproduced the issue and fixed it in master and for copter 3.6.
I’ve done a build here for you to test:

1 Like

Thanks so much will test it!

Quick test on my quad using Pixhawk classic flight controler few increible! ChibiOS rc-5 Thanks!



Thanks for the report, I’ve added it to our list to investigate.

Hi happy to report that the Green frog is AOK using ChibiOS copter-mRoX21-beta-test.apj you made me! and is working really nice !! floats like a dream!! Big thank you and Randy as well!


1 Like

Great, thanks for testing so quickly! This will be included in Copter-3.6.0-rc6 which will be out within a week I think.

1 Like


Quad nose dived in takeoff and crashed.

IronMan 650 Sport
ESC XRotor Pro 50A
3.6.0-rc5 Pixhawk 4
stabilize mode.

When I started the takeoff all seemed stable, but at about 3 feet the quad took a nose dive and landed on its back. Ended up with only one broken antenna.

Actually I have a hunch that the cause is due to mechanical failure but I’m not sure. I am not proficient enough in reading the logs so any input would be appreciated.

2018-07-12 2007-48-42.bin
2018-07-12 2007-48-42.log
2018-07-12 2007-48-42.log.param