You could flash SimonK Firmware (if it is possible). When they have the same software there should no noticable difference.

I made very precise calibration and found some mechanical problem, as propeller's mounts on 4 identical motors have 2 various dimensions (Turnigy from HK)

After all is leveled and calibrated problem vanished

But anyway to build Pirate I have ordered 4 motors and 4 ESC and got two kinds of ESC and two kinds of prop mounts .... a little mess.

But it's ok, now.

@alex.. which rev is the one that enables data logging for AIOP2?

i'm trying to mod APNG to use data logging

Originally Posted by Paku

Short question on little copter drifts, as I encounter a little CCW yaw drift as well as little back drift.

1. Could it be a meter of sensor calibration ???
2. Should I or not, compensate with TX trims /I think no/
3. Longer story but could be it's the point. I mean my ESCs are not exactly the same. I owned on in the drawar so ordered 3 more - but as I am looking at them they are physically not the same. Could be it's the known issue with Turnigy Plush series moved from Atmel to Silab CPU .. but the drift is really small so they seems to work "almost the same"

Any way I have 3 pcs looking like that /big cap on the board/



while one older /like 3 months/ looks like this:



I was really disappointed when I discovered the difference ... but definitely they changed something in the mean time. Again ???

But finally I would expect that I can can put a bottle of water on one arm and it will try to be stable, so stability is more sensors but not hardware based.

You can flash Bl-Heli:

http://oddcopter.com/2012/07/18/flas...heli-firmware/
http://www.olliw.eu/2012/owsilprog-tutorials/
http://www.helifreak.com/blog.php?u=...categoryid=119

I fly turnigy 12A with BL-Heli -- Multiwii on a Megaboard... it flys GREAT !!!!
Nenno

I have configured the gimbal but no matter what I try I have some random twitches in both pitch and roll. The twitches have a 2-3 seconds frequency and the movement is about 1mm small.

Can this be the a code problem? Any ideas on what might be cause of this?

Things that I tried but didn't solved the problem:
1. tried several other servos
2. disconnected unnecessary hardware (FPV, gps, etc)
3. tried mpng 2.8 and 2.7

wrong selection of rc chanel look in hardware and gimbal and select it there...

Originally Posted by SecretM@n

wrong selection of rc chanel look in hardware and gimbal and select it there...

The gimbal only works if I select RC5 and RC6.
The servos are connected to pins 44 and 45 (aio pro)

Is this correct?

What is the control channel then, if CH5 is for gimbal ?!

Originally Posted by renatoa

What is the control channel then, if CH5 is for gimbal ?!

A8, A9, A10, A11 --> (roll, pitch, thr and yaw, not necessary in this order)

A12 --> RX ch 5 (fly mode selection)
A13 --> RX ch 6 (pitch gimbal control or PID tuning)
A14 --> RX ch 7

A15 --> unused
32 --> unused
33 --> unused

servos connected to pins 44 and 45
FC=AIO PRO
As I mention previously, the gimbal only works if I select RC5 and RC6 but it has some random twitches.

Originally Posted by nenno

I fly turnigy 12A with BL-Heli -- Multiwii on a Megaboard... it flys GREAT !!!!
Nenno

As for now, it all seems to work well but will try to flush them later to make stuff even better

There's an ArduCopter v2.9.1 release that "resolves the throttle failsafe bug found in 2.9 which could lead to the copter performing a Return-to-launch even if the there had been no radio failure." From what's described in the DIYDones thread and browsing the code, it looks like they modified a line in the 'read_radio()' function in "radio.pde" to go from this:

        if ((millis() - APM_RC.get_last_update() >= RADIO_FS_TIMEOUT_MS) && g.failsafe_throttle && motors.armed() && !ap.failsafe) {

to this:

        uint32_t last_rc_update = APM_RC.get_last_update();
        if ((millis() - last_rc_update >= RADIO_FS_TIMEOUT_MS) && g.failsafe_throttle && motors.armed() && !ap.failsafe) {

Should this change also be done in MegaPirateNG 2.9?

--ET

MPGN 2.9 R5 based on ArduCopter 2.9.1 available in the SVN.

Originally Posted by etheli

There's an ArduCopter v2.9.1 release that "resolves the throttle failsafe bug found in 2.9 which could lead to the copter performing a Return-to-launch even if the there had been no radio failure." From what's described in the DIYDones thread and browsing the code, it looks like they modified a line in the 'read_radio()' function in "radio.pde" to go from this:

        if ((millis() - APM_RC.get_last_update() >= RADIO_FS_TIMEOUT_MS) && g.failsafe_throttle && motors.armed() && !ap.failsafe) {

to this:

        uint32_t last_rc_update = APM_RC.get_last_update();
        if ((millis() - last_rc_update >= RADIO_FS_TIMEOUT_MS) && g.failsafe_throttle && motors.armed() && !ap.failsafe) {

Should this change also be done in MegaPirateNG 2.9?

--ET

Looks like Sir Alex already included this change in the code.

Originally Posted by Adilson

Looks like Sir Alex already included this change in the code.

Yes, that and the other changes from ArduCopter 2.9.1 are in there now. Super work as always by Sir Alex.

--ET

Don't know if this is a bug with 2.9 R5, but I can not calibrate my radio in Mission Planner 1.2.33. Reverting to 2.9 R4 restores the ability to calibrate my radio in Mission Planner.

Also, with 2.8 we were able to recalibrate the AIOP by holding disarm for >20 seconds. With 2.9 and the more detailed calibration in Mission Planner it seems this feature is gone. Is there any way to get the calibration with the sticks back? Comes in handy to recalibrate in the field without the use of Mission Planner.