9XR throttle too sensitive

Hey folks,

I have a 9XR that i just started using to fly my APM deadcat quad with.

The problem is that the throttle stick is way too sensitive for me.

I used to fly the quad with a cheaper radio set and the throttle stick seemed great. I had too move the stick 'miles' to get it to change speed - which is what i like.

Now with the 9XR i only have to move the throttle stick a few mm and my quad is shooting up through the sky. Also, today when i was flying the quad it was a bit too high in the sky so i moved it down a few mm but that was zero throttle and it fell through the sky! I corrected it and all was fine - but i want to fix this.

I have calibrated the radio through mission planner. Just hoping someone with more experience could tell me how to make it less sensitive.

I was thinking, that at the moment the throttle channel mix on the 9XR is 100%. I guess this is the range it uses on that channel. What if i change it too 50%? Would that mean that the whole range of movement of the stick would only move through the lower 50% of the channel output. Effectively meaning i would be flying at MAX throttle at was previously 50% throttle?

Any suggestions would be much appreciated!


You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –


  • Same problem and I am using devo 7.

  • I had the same issue and found this video

    9XR Throttle Curve Using Expo

    It made a world of a difference. I haven't had any issues, but I am not an aggressive flyer

  • I think i may have this sorted out.

    I am experimenting with throttle curves, in particular the expo throttle curve on the 9XR which should hopefully give me a stable hover.

    I am also putting a sonar module on the quad to help in altitude hold mode.

    By moving the COG back the quad should be more stable also, as the deadcat wants to fly forward because of motor orientation.

    • Hi,


      I wouldn't really recommend doing it this way. It's probably ok in modes where the throttle is under manual control (i.e. stabilise) but not ideal in alt-hold, loiter etc.


      In those modes your throttle stick is used to request a particular rate of climb / descent, rather than a throttle output. Therefore you shouldn't find the "throttle" to be too sensitive in alt-hold. If you are, then you need to tune alt-hold mode on the APM.


      To fix it being "too sensitive" in stabilise mode you can also start by reducing the throttle accel P parameter. I did this on my F450 as I also found it to be a little too sensitive with the defaults. I think it's better to get the copter responding "correctly" (i.e. not too much) to small throttle movements, rather than change the radio to make it easier to make even smaller throttle movements.


      If you do use radio expo then this should work ok in stabilise mode but in alt-hold mode you will find you have to make bigger stick movements than you want in order to change altitude from your hover position: in these modes you probably want a linear relationship between your stick position and the vertical speed of the copter and putting expo on the radio will prevent that.


      Therefore the other thing you can do on the 9XR radio would be to set up a custom switch so that your throttle expo is only active when you are in stabilise mode, and it's a linear curve in all other modes. I've done this on the roll/pitch stick with my 9XR, to apply expo in stabilise / alt-hold modes to make accurate flying easier at close quarters, while not having expo in loiter and other modes.

      • Perfect! Thanks for your input. Much appreciated. I will change the p accel value, possibly make two throttle curves, and i am also getting a sonar. That should sort out the throttle issues. But the deadcat still wants to fly forward all the time so i am hoping that next chance i get to fly, having the COG further back will limit that.

        • I've never used the sonar - we have one, but didn't find it necessary. altitude hold is accurate to within a few inches without it so it would only be necessary if you want to follow variations in the ground surface at a fixed distance. I believe they're pretty tricky to get working well, with the "noisy" sonar environment caused by the prop wash. I also believe that they wouldn't have any impact on manual throttle modes (stabilise) though am happy to be corrected there.


          Regarding flying forward, generally this shouldn't happen regardless of CoG, so long as the I term and Imax in your rate roll / pitch PIDs are big enough. It's still a good idea to get CoG as central as possible, don't get me wrong, but if it's flying forward then the either the CoG is drastically wrong or something else is amiss, because the APM is working to keep the quad "level" (even against a CoG error), which in windless conditions should mean not drifting.  

          Does the HUD in Mission Planner show level when the quad is sitting on a level surface? If not you need to re-do accel calibration. Is it possible the motors may not be mounted square (not pointing straight down), such that when the quad is level the thrust is not vertical? Is your rate roll / pitch I term at least as big as the corresponding P term? If those things are all ok, then the quad should hover more-or-less in place (except for wind) even if the CoG is a bit off.

          I also had a problem with my 9XR while I was getting used to it, whereby I had it set to save trim when I flicked the trainer switch (whatever was the current stick position was set as the zero position). Might be worth checking you haven't got that switched on too and inadvertently saved an offset neutral position on the pitch channel (see what RC input the APM is reporting on the RC calibration screen when the stick is at centre).

          • I have heard that many people have problems with deadcat style of configuration. It has the front motors further apart - so the APM thinks the motor aren't where they are. Arducopter configuration doesn't have an option to specify the new location of the motors does it?

This reply was deleted.


Shivchand Jaysaval liked Shivchand Jaysaval's profile
Aug 25