Arducopter 3.2 Stable: Important Bugs List

This thread is intended to present any known bugs and other issues with the 3.2 stable release. It should not become a general chit-chat thread, nor is it a problem resolution or investigation thread. But if I’ve missed something, please feel free to post to highlight the issue. Only significant problems which would cause a crash or other dangerous operation will be shown, to keep the list short.

[ul]

  1. Do not use DO_Set_Home mission command. This results in a bad operation within the program, and definitely lead to a crash. https://github.com/diydrones/ardupilot/issues/1677

  2. Sometimes the Landing Detector has a false positive and will shut down the motors even while flying at altitude. https://github.com/diydrones/ardupilot/issues/1678 This error is very rare, but has happened a few times. The error is recoverable if the user immediately increased the throttle stick position to full, or switches to Stabilize and raises the throttle. The copter will begin flying again.

  3. RCMAP do not use the correct RC_Min and Max values. https://github.com/diydrones/ardupilot/issues/1661
    There is a workaround, ensure all your channels have exactly the same Min and Max in your Transmitter, then redo the radio calibration and ESC calibration. Now even though the RC_Min and Max params are not working right, it has no effect as they are all the same anyway.

[/ul]

I am reading MAVlink messages from APM autopilot and the pressure reported is half of the real pressure thus like 504hPa instead of 1008hPa - did anybody see this problem ?

Pawel.

bad baro : you have a 5607 instead of legacy 5611
It’s common on chinese pixhawk 2.4.8, dangerous to use these boards

Thank you, important information… this is APM, not PixHawk - is there any workaround ?

Pawel.

The only way is if I’m not wrong to make modifications in ardupilot code : bebop is 5607 so it’s already implemented. Then build your firmware yourself. I saw one user who is making it on arduplane in rcgroups forum but still had problem with baro. Replace 5607 with 5611 is a better option but buying a new board with good sensors should be better …

Why is MS5607 “bad”, apart form having worse pressure resolution then MS5611 ?
Is the programming interface different ?

If I’m still not wrong, ardupilot is opensource (openhardware and opensoftware). They provide schematics to hardware which un apm and pixhawk boards include 5611 barometer.
5607 is cheaper and have less precision/accuracy than 5611. 5607 have half resolution on comparison to 5611. Firmwares are build with taking consideration of default sensor provided for apm or pixhawk. If you don’t have sensors required by project, manufacturer didn’t respect schematics of openhardware design.
If you have this cheap component on your board, you could have doubt in other parts IMO…

Hi, Was this MS5607 Baro issue ever resolved in ArduPilot ? I see it works with PX4 after changing a parameter but can’t find any solution for Ardupilot …