i am curious why are u adjusting the aux ports
could you elaborate alittle more about acro flight mode being different for the others firmware
Just trying to understand how this firmware works: are not the aux ports that control the ESCs?
Why are there PWM values and PWM_AUX values in the outputs tab?
Well I tried just briefly to switch to acro mode, but the copter was definitely out of control. I did not insist and switched back to stabilized. It is years now that I set a panic switch on the TX, when dealing with autopilots; it is so useful!
I can only imagine what would happens if a beginner tries acro mode and has to deal with such a situation ā¦
Iām not sure if it is only a matter of gains or the firmware isnāt able to manage an acro mode.
Acro mode in all the firmware I know, is a mode where the copter is held in its actual attitude and the pilot has control of the attitude and rate of rotation. It is also called rate mode for a reason. The copter does not have any other stabilization other than holding its actual attitude. I great improvement on that was Airmode, where the copter is held n the actual position even with motors not spinning.
Iām used to control multirotors in acro mode with high gains, but this seemed way out.
Maybe I will try to lower the rate controls, gain altitude and watch more carefully what really happens in acro mode.
Would be nice to have someone on the coders side explaining a bit more how this system works and what is really implemented in which release. Iām still dealing finding documentation.
i am yet to try flying in acro mode the weather here in ohio is being extremely crazy and unpredictable , its supposed to be warm this weekend will see if i can get out to the field for some PID tuning and with some luck my oscillation issues might get tuned out and i will then have the confidence to start beating on it
@chiloschista Hey Ric
Quick input on the PWM_* parameters: the Pixhawk has 2 output drivers: MAIN (IO) and AUX (FMU). PWM_MAX/MIN/DISARMED/RATE are for the MAIN outputs. On the Pixracer we only have 1 output driver which is the MAIN (but FMU) output. I suggest you try setting PWM_MIN/MAX/DISARMED/RATE like @greybush has shown it.
The reason you see PWM_RATE in the Default Group might just be a bug in the Parameter handling from QGC but you donāt have to worry about. The way it works is when you flash your Pixracer, QGC pulls of the meta-data from the firmware which contains the latest parameter descriptions. Try restarting QGC after flashing, that might help.
After you got this sorted out with the ESCs, please upload a log to http://logs.uaventure.com/ and share the link to the log here so we can take a look what might be wrong in Acro 
HTH
- ada
Hello Andreas,
thanks for your explanation!
So my last changes did change nothing: I had this suspect as nothing did change in flight.
I already tried to set PWM_* like @greybush, but result was motors spinning at max revs out of control. I had to disconnect the battery to stop them.
I tried again a momet ago, after a QGC upgrade, 3.1.3 now. Same result. ESCs seems not calibrating.
By the way new QGC has MOT_SLEW_MAX now on top of PWM outputs tab.
Should I switch to master?
About Acro flight mode I tried again holding the multirotor by hands. The feeling is that the quad does drift to the right, it does not hold the attitude.
I tried to trim it but it was drifting anyway. Seems it is missing some sort of deadband.
Ric
Master shouldnāt be necessary I think. Did you restart after changing the PWM_ values? Which airframe configuration do you have set?
Note: Acro has no attitude lock, so it will drift.
Yes, I tried everything I was able to think about. Out of curiosity I flashed master, but have the same results. Am wondering about my ESCs.
The frame is a TBS Discovery and I set it that way.
What? Acro has to have attitude lock, it has to stay where the pilot put it. Itās not leveled, nor absolute like stabilized, but locked. Otherwise the pilot will spend his time trying to control a cake of soap.
I just received the PixRacer R14 sold by mRo, but it doesnāt come with the 3M damping foams like the Pixhawk.
Does the PixRacer still need those foams to isolate vibration or can it be mounted directly onto the frame center plate?
Also, can I use v1.6.0rc1 to test it or should I use master?
Searching a little bit I discovered that Emax BLHeli Oneshot125, are declared that way but seems to be some strange, not standard firmware: crappy ones?
Ok I can leave with that. That multirotor is meant for night flying and eventually inverted flying.
What I miss right now is a working acro mode and a better user friendly flight modes selection possibility.
Ric
one thing over time i reallized with this FC is that when u flash from one px4 version to another , problems a rise for the previous version of firmware settings and info will carry over setting to the next flash , example if you flash from standard to master px4 you should have noticed that some of the calibrations for your compass and maybe a few other things you may not have had to setup this up. this a problem and can be fixed by flashing arupilot before reflashing back to px4 master version.
maybe this is why you had issues, its worth a shot to check it out ,post your results
@chiloschista If you have a suggestion on improving the flight mode selection maybe open another thread for a discussion?
About Acro: first of all, not everybody wants attitude locking. If your copter is tuned well it actually holds attitude pretty good purely rate controlled (which is the core meaning of acro). But it looks like one of our devs is going to add this soon so you might be lucky 
Again, if youāre still facing problems with Acro, please post a log so we can give you advice, Iām sure it only needs tuning.
- ada
not sure i understood the problem but i think current PX4 Acro mode already have attitude locked on roll/pitch?
currently not, not such a big deal to add though
This PR implements full ACRO mode attitude control (attitude control vs. rate control): https://github.com/PX4/Firmware/pull/5498
It flies well, and feels like perfect rate control, according to both me as an intermediate pilot and the flight logs.
Warning: long post ahead!
Hello @Mark_Whitehorn,
thank you very much for that (I thought you were behind it = : D). Will see if I can do a flight during the weekend.
@AndreasAntener again about Acro mode, I can agree with you if we speak about a fixed wing. A fixed wing does not require attitude lock because it has aerodynamics helping on that side, e.g. static margin, dihedral etc. But did you ever try to fly a fixed wing with a negative static margin or a negative dihedral? A plane with tail becomes crazy, but the pilot could eventually still control it and do crazy aerobatics. Not very comfortable though! A tailless plane becomes unflyable by its nature, but it still has a little bit of help by aerodynamics. I was able once to land a plank with a completely wrong CG and therefore negative static margin.
I guess a multirotor without attitude lock can be compared to a fixed wing (better a plank) with a negative static margin and negative dihedral. It does not stop rotation, but continues it forcing the pilot to counteract all the time.
As an example imagine doing a 4/4 roll: with a fixed wing, with aerobatics settings, say very neutral static margin and dihedral set to zero, the pilot has to roll and stop four times, without any couteract (Iām not speaking of pitch control!). The plane will stop rotation when the pilot stop the roll input. I expect the same from an aerobatic multirotor, but I guess without attitude lock it will become impossible.
Is there around anyone flying a multirotor in Acro mode (I mean flying not levitate = : D), any video?
Sorry to be long, but am trying to be clear in what I mean.
Ric
@chiloschista Hi Riccardo, good to see you over here. I should have mentioned that my PR is MC only (and probably obsoleted by Andreasā work), but the implementation would obviously apply to FW also. The āsequencerā part is used to program attitude sequences which can be triggered by a TX switch. The 2 point roll was very cool to flight test on a multicopter. Iām looking forward to perfect 4 point and slow rolls with FW once Andreasā stuff is in.
Hey Mark, I didnāt make your PR obsolete I think ;). I only added Acro to FW. You could check with Daniel how to make the sequencer standalone, it could then be used for both which would be very cool :). Also check with him on attitude locking as he was interested in adding that.