LibrePilot Forum

Users => Vehicles - Helicopters => Topic started by: Filippo94 on April 07, 2016, 12:17:59 pm

Title: Non linear Servo Response
Post by: Filippo94 on April 07, 2016, 12:17:59 pm
Hi guys, I started the basic setting of the CC3D for the first time, while waiting all the components of the Tarot 450 helicopter are at their places. I've set up the main frame, and the servos, connected them to the CC3D, which is also connected to an Orange receiver. Everything seems to work fine, stabilization works, inputs work, servo direction is correct, plenty of response when moving the board.

I get though a wrong servo response from the collective channel. Everything in my radio has a linear 0-100 curve, the GCS reads it correctly, the Curve settings panel has linear curves, and even the servos reach correctly maximum and minimum of their range while checking with [Test Outputs], but when I change the blade pitch with the transmitter, I get a splitted curve that seems to very roughly go to 100-0 in the upper and the lower part of the stick.

I've recorded a video about it, sorry for background noise ::). (https://www.youtube.com/watch?v=2lDNM_x1CtY)

I've checked all the possible things around the software, but it's my first time with it and it's possible I missed something. I can not find an explenation of this behaviour though :-\.
I attatch my current UAV too, be aware it's incomplete and untested.

Do you have any ideas on how to solve it? Or maybe it's the way it's intended :o?

Thanks in advance :).
Title: Re: Non linear Servo Response
Post by: daveapplemotors on April 07, 2016, 01:34:16 pm
Your UAV TX values in Hardware Settings:

 <field values="1073,1891,1886,1901,946,1902,0,0,0,0" name="ChannelMin"/>
            <field values="1106,1486,1478,1488,1487,1481,0,0,0,0" name="ChannelNeutral"/>
            <field values="1920,1060,1059,1068,2029,1056,0,0,0,0" name="ChannelMax"/>
           
Try replacing the 1106 with 1075 for throttle neutral. And possibly change the pitch neutral to 948 instead of 1487. We must set neutral close to minimums on throttle. My collective and throttle channels have very similar numbers.
Title: Re: Non linear Servo Response
Post by: Filippo94 on April 07, 2016, 02:48:53 pm
Having the neutral pitch under the min value gave some strange results, I've lowered the min to 500, now it's the correct way finally. I'll have to further tweak the values though, it's really asymmetric.

Thanks for your help :D

About throttle value, I did not test it yet (only the min value to turn it on in the output tab), as I have the engine plugged off. I have not yet the main gear bearing, so the engine is not engaged. is it safe to push it to the max without any load?
Title: Re: Non linear Servo Response
Post by: daveapplemotors on April 07, 2016, 03:06:46 pm
Sorry,
I thought pitch minimum was 846 and you lowered neutral to 848 right?  But that was channel 5 and I goofed. Try channel 6 and put hte neutral value very near minimum. Do that with throttle, too. And no need to put neutral below min ever.

And yes, a brushless motor can run full throttle with no load. Usually I run my 250s at about 70% throttle (set in TX after final Manual Calibration of TX with defaults).


BTW
Set your manual rates to 450 or higher to see some movement later on. 220n is way too low.
Title: Re: Non linear Servo Response
Post by: Filippo94 on April 07, 2016, 03:23:21 pm
Thanks for the suggestion, now I can surely see better how the cyclic behavies, and if there are colliding parts.

I've attached current radio values, seems to work pretty well. I can't measure pitch, as the main shaft is not correctly positioned yet (the main gear is going to change its height maybe), but the range seems even on both sides looking at the servos' angle.
Title: Re: Non linear Servo Response
Post by: daveapplemotors on April 07, 2016, 03:29:23 pm
Collective neutral is still lower than minimum in your screenshot.
Title: Re: Non linear Servo Response
Post by: Filippo94 on April 07, 2016, 03:39:06 pm
True that, forgot to change. Going to put it a hundred higher than minimum.
I thought the neutral value is the center of the signal, where indeed there's no positive or negative input. Why it has to stay close to minimum?
Title: Re: Non linear Servo Response
Post by: daveapplemotors on April 07, 2016, 03:45:42 pm
I don't know why it has to stay close to minimum. But only on Throttle and Pitch. Try neutral value a little too high and you get swash jitters.
Title: Re: Non linear Servo Response
Post by: Filippo94 on April 07, 2016, 03:48:38 pm
Yeah saw that, 100 higher than min causes this. I'd have then to put it as close as possible to this threshold, without causing any jitter, right?

Why? It works the same with gyro compensation, battery usage, flying style: As close to the edge as possible, without touching it ;D

Edit: Hmm, putting it even at the same value of minimum still causes jitters...
Edit2: Actually it cuses to show again that interruption in the rensponse curve
Title: Re: Non linear Servo Response
Post by: daveapplemotors on April 07, 2016, 03:53:20 pm
Yeah. Close to minimum stops jitters.

We heli pilots live life on the edge:)
Title: Re: Non linear Servo Response
Post by: Filippo94 on April 07, 2016, 04:04:50 pm
Hell yeah!!  ;)

Further testing seems to show that the collective min value has literally no usage, I've set neutral to 1000, and from 1000 to 500 the min value does no differences as far as I can tell, not even causing deadzone...
Title: Re: Non linear Servo Response
Post by: daveapplemotors on April 07, 2016, 04:16:12 pm
Cool, thanks. I have never checked collective because they looked like throttle values to me and I kept them equal to  throttle values.

Only throttle is important to have close to minimum, right?

Good on you.
Title: Re: Non linear Servo Response
Post by: Filippo94 on April 07, 2016, 04:34:54 pm
Not sure, the only setup I did on engine is from the output tab, where I've set a mid value where the engine starts spinning. And of course set your values. The throttle response is pretty good, maybe a bit discrete (I can hear light "zones" of speed at higher values, but I'm really unsure about the quality of my electronic). Will further test once I'll have a complete heli to test with.

Thanks again for your help ;D!