Strane wobble isseu

[SRXer]

I have a stange wobble isseu in my Swift 550 fbl with gt5.

The heli has had numerous flights 200+ and i am very satisfied with the gt5 performance.
Lately (last 20-30 flights) the heli shows some strange behaviour.
Few flights flying very fine straight crisp etc. (within the range of low headspeed flying 1650rpm).
The other moment (next flight) i notice some reactions which i did not do (elevator), or high speed ending up in pitching up or down.
If I bring the heli down for hover i see a continously increasing wobble on elevator from just a bit progressivly increasing to +/-20 degrees with a 1hz frequency over a period off 4-5 sec. No need to The heli is a real pita to land.

I thought it was vibration related so I changed all bearrings.
I thought it was PID related but as soon as i brought the overal gain channel from 150 to 0 (!) while hovering no change in wobble.
I thought a crash should damaged the gyro but why 75% off the flights good 25% sh*t?
I thought it is servo related but the servo's are doing fine in a vibration test (i dont have a spare set to swap and test).
There is no relation to wind as far as i kan see.

3 things seem to be repeatable:
1) if the wobble occurs, land, power off and on, problem gone.
2) i am under the impresion the wobble starts after visiting some menu's.
3) it is always a complete flight or the remaining flight after visiting some menu's

I am open for every suggestion, please HELP

[gwright]

Make sure you have the vibration filter turned all the way up (goes 0 to 5, 0 is max, not 5, but try both in case I'm mistaken). Increase sensor deadband to max.If those don't stop the behaviour, then your I gain is too high, so :
increase the I-limit (max is OK),
If still no joy, then decrease the I gain on the swash. Takes a big change to do anything, so first go halfway between where you are now and 0. If better but not resolved, go half again, if totally resolved, you can play with i gain by increasing halfway from current point to previous problematic point, and get it as high as possible with no wobble.

I max out the deadbands, they're very very fine adjustments (sensor and tail are the two available), max out the I limit, then crank I gain till it wobbles and/or porpoises in forward flight, then drop a bit.

[SRXer]

Thank you gwright,

I will test coming weekend and report back.
It is late here next time i'll also report my settings prior and after the test.

Kind regards for your reaction.

[Laurens]

How are your head dampers? Which FBL head?

[SRXer]

Hi Laurens,

after some misfires i try to answer questions:

dampers:
swift16/550 original and
trex hard dampers 80shore? not the stonehard dfc dampers.
Both same results.

Head
metal upgrade head from swift for flybar,modified to flybarless.

remember over 150 flights all ok, then wobble problem started.

So now my setup
Only the groups followed by settings in the program order.
SETUP
97,111,96,5,standard tx,dx7,-,-,-
STICK MENU
40,45,0,0,22,5,5,10
SERVO SETUP
760,0.05,333,120,0.112,200
SENSOR SETUP
n,r,r,r,r,5,2
SWASH
electro,117,90,90
SERVO MENU
mid 20,5,-6,6
rev. n,r,n
swash 115,115,115
a-b 175,180,250,250,250,250,250,250
P/I MENU
swash 77,60,100,20,85,15,4
tail 108,80,7,0,-2.22,-15,-25,0

any comments please let me know.

regards

[Laurens]

I wouldn't actually know by just these facts. I think its better to visit someone with more experience to see if he notices anything. I would suspect something mechanical.

[SRXer]

Testflying today.
Open field, 10:00AM, 29⁰celcius, 1-2 bft, clear sky.

Flight 1 original settings as posted before, result: increasing elevator wobble as describet in the first post.
Flight 2 Swash I gain to 40, result: same as flight 1.
Flight 3 Swash I gain to 20, result: same as 1 but also slow reacting and disconnected flight.
Flight 4 Swash I gain to 50, Swash sensor deadband to 25, result: Wobble gone but now a continues 5mm up down movement on the tail (5-10Hz).
Flight 5 Swash stick deadband to 10, result: same as 4.
Flight 6 Swash sensor deadband to 15, result: Wobble back.
Flight 7 Swash sensor deadband to 25, Swash P to 97, result: nice flight.
Flight 8 Swash P to 107 result: good flyable but tends to fast elevator wobble sometimes and has a increased tendency to pitch up or down at FFF.
Flight 9 Swash P back to 97, result: nice flying is back again, at sharp sharp elevator moves a bounche also on tail stopts, tail up down 5mm not not very visible.
Flight 10 Swash feedforward to 75, Tail Dynamic to 22 result: tail OK swash almost gone.
Batteries empty.

Conclusion: the Swash sensor deadband did the trick, I could also ioncrease the P value about 20 points. (Thanks Gwright)

General thought
Swash deadband is used to cope with mechanical play? If so it could be that wear is the main cause of my problem, the heli is about 200 flights old had some crashes, the servo’s have play, if measured on the rotorblades in total about 0,8⁰ .

A other thing:
When I installed the GT5 I had the gain channel on 100 (on GT display) I could crank the Swash P and I levels to the max without any isseus. To find the Pand I max valeus I cranked the gain channel to 150 this remained that way all subsequent flights and the testflights above.
The need to crank the gain channel does this point to a (to) low mechanical gain? My servo throws are all 115 do you guys think a adjustment is needed here?

[SRXer]

This evening again four full flights.

Hot but no wind at all.

Swash D gain to 10,
swash feed forward to 70
input deadband tail and swash to 10: I am satisfied. no bounce in elevator,aileron and tail .