Help to review crash log

Hello & Happy Easter,

I’ve been working on a big hexacopter (30" props) for some months now.
Firmware Version: V4.0.3, Pixhawk Cube Black , 2 X Here2 GPS,
AUW APPROX= 28 KG
The frame has flown several times and Auto Tune successfully completed with GPS modules connected to GPS 1& 2 Ports i2C serial.

After extensive reading on advantages of UAVCAN vs SERIAL, I decided to CAN was the way to proceed.

NODE ID: 20 CAN 1 and manually changed NODE ID: 21 CAN 2. (When connected to SLCAN only one GPS is displayed, however under HW ID both
UAVCAN modules displayed as COMPASS 1 & COMPASS 2

After encountering many compass calibration issues, BAD GPS etc last month, I finally managed to get no error messages on MP & thought frame was ready to be flown and re tuned.

COMPASS_TYPEMASK: 54271 allowed me to successfully calibrate both external compass. Today I discovered 63487 is most appropriate???

Once airborne flight modes used - STABILIZE & POS HOLD - No problems
Selected AUTO TUNE and the frame pitched forward into the ground. Yes, it was spectacular & Rome wasn’t built in a day.

I’ve spent the morning learning how to review data logs etc, however I would very much appreciate some professional expertise.
The two items that are of most concern are:
Large compass offsets
FAIL= MAG_ENABLE not found.

FYI, I’ve got a quad with same setup which is solid. The only difference being
GPS 1 = SERIAL
GPS 2 = UAVCAN

Can Pixhawk cube black use CAN 1 & 2 simultaneously or 1 x CAN - 1 x SERIAL only?

Thanking you in anticipation

The one item “of most concern” is nothing. Auto Analysis reports that because there is no MAG_ENABLE parameter in 4.0.

Go through the tuning page here as I noticed your ATC_RAT_RLL_FLTT was still at default.
That will definitely flip a large copter.

Were you flying a breeze?
If not then your copter is a bit out of balance.
Otherwise the power level to the motors looks spot on.
Vibrations look excellent.
Even Desired/Actual looks great, until of course it flips over.

So go through that tuning Wiki carefully and try again.

Thank you for your time and advise.
I will recheck the PID settings again.
When first auto tune carried out all PID settings where at default and the copter (all be it a little aggressive) successfully completed the auto tune more than once.
Test flight was conducted in calm conditions.
Any thoughts on “large compass offset params?”

Thanks dkemxr, I was not aware of that.

Hi dkemxr,
What are your thoughts re “large compass offset params?”

Again, what’s showing in the Auto Analysis is not current. The maximum compass default offset has been revised upwards to 1800. Your offsets from the parameter file are not that high and unless you are getting compass errors I wouldn’t worry about it.

Hi dkemxr,
Thank you for clarifying compass offset.
So basically the crash was due to incorrect PID values being preloaded prior to Auto Tune?
As previously mentioned flight in both stabilize and pos hold where good prior to auto tune being engaged.

Hi Mike,
Apologies for badgering you for advise.
Do you believe wrong PID’s where responsible for auto crash?
Also when both CAN1 & CAN2 Ports are utilised should they both appear in UAVCAN menu listed in Mission Planner?

Thanks once again

Have you been through the tuning wiki and made adjustments?

Haven’t played with CAN yet so I cant answer that question.

Thanks Mike,
No haven’t made adjustments yet as the frame requires some repair work.
I’ll keep in touch once adjustments etc are made.

Cheers