Re: Collective jumping
« Reply #15 on: March 19, 2018, 11:56:58 am »
Ok, so i missed what you were trying to say about the min. I reset that and things are acting better. I will let you know. I think that my be part of the issue not i just need to go through and reset up eveything.

Does the throttle neesd a low min too? Or just the collective.

Thank you so much.

mr_w

  • *
  • 207
    • LibrePilot
Re: Collective jumping
« Reply #16 on: March 19, 2018, 12:40:05 pm »
hi,
I have double checked what I did on my setup, and I have to correct myself, it is not V shaped but this:



This has given me nice positive and negative collective.. or at least I believe it does so :) (the thing actually flies).

Re: Collective jumping
« Reply #17 on: March 19, 2018, 05:37:09 pm »
Need both throttle and collective to have neutral set very low.  Reasons:

I think what we have learned is that for the default setup, when the throttle stick on the transmitter is above zero, but below neutral, then all the channels are set to their neutral settings.

It sounds like you had a low throttle neutral, but a middle collective neutral, so the first little blip of throttle made the collective servo go to it's neutral (middle), but adding more throttle stick made collective go to where you want collective to be.

karla

  • *****
  • 629
Re: Collective jumping
« Reply #18 on: March 20, 2018, 02:06:40 am »
@Brad,
I dont think you need to redo all the setup.
Just remember every time you need to redo the transmitter calibration you need to put the neutral back to min.
That change will cover it.
(but of course you still need to level the swashplate travel and set up correct angles for cyclic and collective :) )
Let us know how things develop!

@mr_w,
when the Collective Pass through is checked, in Vehicle, Helicopter, Basic settings, CCPM options,
then I don't think it matters how that curve look like. It will use whatever the transmitter is sending right?
I think the setting of the mixers (Collective, Pitch, Roll) just to the right, will have most impact on a good enough range of the collective.
« Last Edit: March 20, 2018, 02:29:27 am by karla »

mr_w

  • *
  • 207
    • LibrePilot
Re: Collective jumping
« Reply #19 on: March 20, 2018, 09:28:58 am »
@karla

Yes the curve actually matters, it is applied to collective channel. I suppose it would be helpful if there was title above the curve which says so.

It looked like for -1,+1 range, collective goes (with jumping) twice through all positions, so without digging further into the code, I just assumed that something in mixer is doing abs(collective) (or equivalent), and then applied this curve to scale -1,+1 to 0,+1.. which kind of fixed it. I am also pretty sure that this can be properly fixed somewhere else. Just thinking if this is common issue with heli setup? (this is my first and only heli with LP so far, but I have done plenty of multirotors and couple of wings).


karla

  • *****
  • 629
Re: Collective jumping
« Reply #20 on: March 20, 2018, 10:14:54 am »
Thanks a lot for verifying that the curve still has impact!

Its not really any issue when setting these Helis up (other than the default collective setting neutral to mid of min and max).
Since, once you check the Collective Pass through, then you will work on the transmitter only and adjust it to what ever that curve is looking like, until you get the desired movements on the pitch angels on the rotor.
« Last Edit: March 20, 2018, 10:52:12 am by karla »

Re: Collective jumping
« Reply #21 on: June 30, 2018, 10:17:28 pm »
@karla

Yes the curve actually matters, it is applied to collective channel. I suppose it would be helpful if there was title above the curve which says so.

It looked like for -1,+1 range, collective goes (with jumping) twice through all positions, so without digging further into the code, I just assumed that something in mixer is doing abs(collective) (or equivalent), and then applied this curve to scale -1,+1 to 0,+1.. which kind of fixed it. I am also pretty sure that this can be properly fixed somewhere else. Just thinking if this is common issue with heli setup? (this is my first and only heli with LP so far, but I have done plenty of multirotors and couple of wings).

My first heli too, and I have 3 other CC3Ds in multis.

I'm just about done setting up a 470LM with an Atom. I too had collective jumping. Through trial and error I found that mine wanted the collective neutral to be maxed. At any low setting it would jump to mid collective when the stick got down near the lowest...or max negative collective. I messed with curves for hours, nothing took the jump out until I tried maxing the collective neutral (actually max -5). My curve is -1, -.5, 0, .5, 1.

Tx is a Taranis. Now I have 3 modes on switch SD (could program for all 6, but don't see the need), 5 throttle settings on 2 switches 0, 25%, 50%, 62.5% and 75% (switches SA & SB set up by logic switches and special functions in Tx), SF as a master arm switch which not only arms the CC3D (set on Acc0), but over rides throttle output to -100 (or zero throttle) when not armed, and when armed it enables the throttle outputs from the 2 switches (again, using special functions and logic). Hope that made sense.

Now that the controls all work proper and react proper to frame movements I can level the swash for real with the tool then fine trim everything.....and then say a long, long heart felt prayer, lol.

After getting her airborne I'll be able to adjust those throttle outputs to suit her.

I want to say a big
THANK YOU
to all those that have come before me working out these heli issues.

Josh
aka The FlyNFool

Re: Collective jumping
« Reply #22 on: July 03, 2018, 10:42:38 pm »
 :o This thing is whacko. Don't know if I'm going to trust a $500 'copter to it.

So after doing a little fine tuning the collective jump came back, with a new oddity. When I get to that spot on the collective where it glitches, my rx (an X8R in D16 connected via Sbus) quickly reboots, of course sending everything haywire.

So tracking down why it was glitching and jumping the collective I ended up putting the collective input neutral back to same as collective low. Watching the input screen I could see it was not my Tx (Taranis) jumping it's output signal, which meant it was the Atoms problem in interpreting or processing the input signal. It simply does not like a full -100 signal. Changing the collective curve in LP proved no help, so I tried clipping the bottom of the output signal in the Tx output menu. -95 is what it ended up smoothing out at. No more jumping, no more Rx drops. Sooo -95 to 100 seems to have made it happy (er).


Any clue as to why the Rx would reboot when this anomaly occurs? What a serious problem. Makes me scared to put her in the air.

Re: Collective jumping
« Reply #23 on: July 04, 2018, 03:28:46 am »
Sounds like a bad 5V voltage sag when you put a big load on your servos.

Re: Collective jumping
« Reply #24 on: July 06, 2018, 02:29:29 am »
Hi TOC,

Servos were not binding, that I'm certain of. But none the less it seems that could be the case. I'm running 8.4v. Atom did not reboot. Must have been a heck of a load, where ever it went.

Swash plate tool came today, woot! Will work on it some more this weekend.

Thanks  :)