Issue with flaperons mix. Servo jitter on D8RXP high speed, digital servo,

[GoogleCodeExporter]

Which board (stock / gruvin9x / sky9x / Taranis) are you using?
Taranis Build A with swapped SF/SH switches 

What is your openTx FW version?
opentx 2834
What is your openTx EEPROM version?
215

RX & servo
Frsky D8RXP with digital servo in high speed mode

What steps will reproduce the problem?
I fly fixed wings 3D stuff and I use flaperons mix as below. basically I've set 
my model up so that when flaperons are deployed, max servo displacement is 
limited by the amount of flaperon deployment and the aircraft will always roll 
on its axis. the mix fundamentals is something like:

CH02 (+100%)Ail Switch(SB↑)
 R (+90%)Ail Switch(SB-) 
 R (+80%)Ail Switch(SB↓) 
   (+100%)CH09 Slow(u2:d2)

CH09 (+10%)MAX Switch(SB-) 
     (+20%)MAX Switch(SB↓)

What is the expected output? What do you see instead?
my servo jitter when I apply flaperons, they stop jitter if I move the any of 
the sticks. By any (I tired aileron, elevator and rudder) see video
http://www.youtube.com/watch?feature=player_embedded&v=KNMMCwYnwWM

Please provide any additional information below.
can you please investigate what is the problem? I'm kind of inclined is some 
computational issue within the tx? maybe you can solve it by an updated 
firmware or suggest possible solutions? temporarily I can use  4 curves to 
setup the source for the 90,80,10,20 points but then I will need 4 curves but I 
really have two curves left and one Gvar free.  

I also wanted to ask if you can expand the firmware and make possible to use 
math to setup a source. for example one value for gvars can be used to to setup 
the flaperons.
GVAR FM0 = 90
FVAR FM1 = 80

CH02 (+100%)Ail Switch(SB↑)
 R (GVAR1)Ail Switch(SB-) 
 R (GVAR1)Ail Switch(SB↓) 
   (+100%)CH09 Slow(u2:d2)

CH09 (100%-GVAR1)MAX Switch(SB-) 
     (100%-GVAR1)MAX Switch(SB↓)

cheers

Original issue reported on code.google.com by cklea...@gmail.com on 2 Dec 2013 at 6:00

Attachments:

• CAP232_3D.eepe

[GoogleCodeExporter]

Can you see the jitter on the channel monitor?

To me this looks like a simple servo issue. Some digital servos try to keep 
their position so precisely that they start to oscillate when a large control 
surface compared to the servo's capabilities with little friction is attached. 
I've had that problem in several cases, usually touching and stabilising the 
surface with your hand makes it stop. Putting a friction pad to "brake" the 
control surface can help. 

Original comment by bernet.a...@gmail.com on 2 Dec 2013 at 6:34

[GoogleCodeExporter]

I will check monitor and servo and report back (possibly tomorrow). What
about allowing implementation of math functions as source in addition to
stick, switch, gvars? I.e ch9(100-gvar1)

Original comment by cklea...@gmail.com on 2 Dec 2013 at 8:14

[GoogleCodeExporter]

The mixer IS a math pad :)

100% CH9
-GV1% CH9

Done ;)

But if that's not enough more will be available thanks to lua scripting 
capability in the future. 

Original comment by bernet.a...@gmail.com on 2 Dec 2013 at 8:34

[GoogleCodeExporter]

Hi, many thanks reminding about the math and mixers. I will modify my model to 
use a gvar as you pointed how. 

Now about the servo jitter I checked the model again and I still think the 
problem must be something computational. I double checked that the aileron and 
rudder servo jitter when both sticks are at centre and only when I apply 
flaperons. the aileron and rudder servo jitter more if I apply full flaperons. 
they stop jitter if I move any of the sticks that are linked to ailerons and 
rudder.

I dont know why it stops when I move the rudder or why only rudder and aileron 
are affected. there is no gyro that might interfere, and the jitter did not 
change if I was holding on the control surface or not. The only thing that I 
could have thought is that I have a knifedge mix on the SG switch that moves 
elevator and aileron to keep the aircraft straight on knife edge, but if this 
mix was disabled or not did not have any effect on the jitter.

the only change from my original post is that the elevator stick does not stop 
the jitter as I previously thought. I looked at the channel monitor (since this 
a mixing issue) but I did not see the decimal number change but the channel 
monitor was flickering. I dont know if this is an effect because of centering 
issues, both of my gimballs were replaced under guarantee or if this has 
anything to do with R2771 / R2774. 

All I'm asking please have a look at the model attached in my first post and 
use one of your own rx and servo and check for your self.
Cheers
George

Original comment by cklea...@gmail.com on 3 Dec 2013 at 3:21

[GoogleCodeExporter]

Wow there's a huge amount of stuff in there.
Can't really take the time to understand everything, so could you please tell 
me exactly what switch combinations are OK and what others aren't. I've done a 
quick test with all 3 positions for SB but can't see anything wrong. 

It seems there's quite a lot of stuff that's useless too, for example the mixes 
on CH13-14-15, as well as 2 lines for Ele D/R (all 3 lines do the same, just 
have it once with no switch and it will be always active). 
Also why do you need all those 17-point custom curves in D/Rs to achieve expo 
and rate when you could just use the built-in expo? (e.g. the 80_100 curve, you 
could just put 80% expo and 100% weight in the D/R line, with the advantages of 
being able to easily edit it in the field and much less CPU load/memory usage).

Original comment by bernet.a...@gmail.com on 3 Dec 2013 at 11:31

[GoogleCodeExporter]

Hi there,  the problem is with flaperons mix (sb switch) ail on ch2, ch5,
elev ch3 and their max mixes.

The  rest is:
Ch13-15 with their curves are for ail elev mix, its activated by RS and
basically helps with harriers and because my elevator links are large the
mix reduces the elevator effect and increases loop radius. It just helps
doing nice and smooth loops (from memory ch13 curve is used, I left the
rest for easy implementation in the field.

The 17p custom curves are on purpose there.  I wanted to be sure that until
50% stick movement all three rates will result to the same servo movement,
and after that transition is smooth. Then If my expo is too much I can
scale as required with gvar5 and a knob. The benefit doing it this way is
that the servo movement relationship will be retained but also scaled.
 On 3 Dec 2013 23:31, <opentx@googlecode.com> wrote:

Original comment by cklea...@gmail.com on 4 Dec 2013 at 12:56

[GoogleCodeExporter]

OK had a look, conencted 6 servos, played with SB and the sticks, and can't see 
any jitter. 
Still believe it's your servos that in the right conditions will start 
oscilating, and for some reason this coincides with the flaperons being down...

Original comment by bernet.a...@gmail.com on 6 Dec 2013 at 9:41

[GoogleCodeExporter]

Ok, I will rig different servo (I recently bought a few savox mg255) I will try 
with those.  Btw what rx servo and rx refresh rate did you use?

Original comment by cklea...@gmail.com on 6 Dec 2013 at 5:00

[GoogleCodeExporter]

right, sorry it took so long. I tried a number of servo and only one brand 
(cheap Hobbyking MG digital servo) is basically having this problem. I'happy to 
close the issue. cheers

Original comment by cklea...@gmail.com on 8 Dec 2013 at 11:04

[GoogleCodeExporter]

Original comment by bson...@gmail.com on 9 Dec 2013 at 6:39

• Changed state: Invalid