Hi everybody,
After a silly crahs testing arducopter 2.3 with our quad ( one of our four blades goes away in the take off at about 50 cms from ground), our APM 1280 with a Hotel 1.0 OilPam seems to be ok except a little big problem:
The artificial horizon in The Mission Planner or in HappyKillmore shows an inverted readout, it meens, when i tilt the board to the left, roll left, the horizont in the CGS shows it tilted to the rigth, roll rigth.
Its shows perfectly and stable the roll angle but in the wrong direcction.
Testing in the cli ( im using ArduPilotMega V2.23 software because of the "non supported in 1280" message in some cli tests on laters versions):
The first thing that called my attention was on Gyro tests, i take negative readouts from gyros, but positive from accelerometers (rolling left)
Gyro
0 0 0 | 0 0 -9
-1 0 0 | 0 0 -11
-2 0 0 | 0 0 -10
-2 0 0 | -1 2 -8
-1 0 0 | 0 5 -7
0 0 0 | 0 6 -6
0 0 0 | 0 6 -7
0 0 0 | 0 6 -7
0 0 0 | 0 6 -7
0 0 0 | 0 6 -7
0 0 0 | 0 6 -6
1 0 0 | 0 6 -6
2 0 0 | 0 3 -7
2 0 0 | 0 1 -9
0 1 0 | 0 -1 -10
0 0 0 | 0 0 -9
When testing IMU it shows the negative roll as espected rolling left:
IMU
r: -33 p: 0 y: 292
r: -33 p: 0 y: 292
r: -33 p: 0 y: 292
r: -33 p: 0 y: 292
r: -33 p: 0 y: 292
r: -33 p: 0 y: 292
and ADC shows more or less what i expected, some colums changing values during moves (Gyros) and other colums changing and staying in a fixed valeu during tilt (accelerometers) like in Raw Gyro test:
ADC
1687 1734 1339 1617 2055 2014 2477 2894 171
1687 1733 1339 1616 2056 2015 2478 2881 171
1687 1733 1340 1616 2057 2014 2478 2892 171
1690 1601 1352 1616 2050 2012 2504 2946 171
1690 1482 1340 1616 2049 1910 2405 2713 171
1703 1499 1330 1616 2037 1882 2486 2911 171
1710 1551 1321 1617 2044 1795 2374 2625 171
1688 1648 1338 1616 2046 1776 2405 2715 171
1695 1679 1330 1616 2045 1736 2349 2566 171
1687 1731 1338 1616 2045 1742 2378 2641 171
1687 1727 1337 1616 2045 1739 2372 2618 171
1687 1729 1337 1616 2045 1736 2366 2631 171
So, has anyone had this issue before?
Is it normal a positive readout from acc when rolling left vs the negatives readout from gyros?
If coming from a damaged hardware, could it be corrected by changing the roll sing in the code?
Im going to test with the analog filters enabled although i think it would not change the roll inversion.
Thanks in advance, and sorry about my poor spaniard english.
Replies
Hi again,
i've solder the analog filters jumpers and it seems to smooth a little the sensors graph, but nothing more changed.
However, after a hand test withouth propellers where the motors seems to response as expected, i decided to upload 2.3 code again and make an indoor fly test leting it hovering from my hand and trying to desestabilize it. It flyes perfectly and when i tilt it in any angle or direction its gets estabilized again whitout problems.
So i figure its must be a telemetry problem. In fact, in this last flyght i've noticed that when i give throttle up, the telemetry get freezed in my pc and its apears again when throttle its cuted. Xbee modules show rx and tx flashing normally during that.
I'm going to use another pc, rewire everything, use separate batteries for motors and APM and test it again.
I will update the tests.
Thanks.