Hello all,

i've just converted my next heli to Spirit... a Diabolo 700 (mod. 2018) and it did not run as easy to get it perfect as with my other conversions before the others all have been really perfect within 2-3 tuning flights... this one is still not.

i had to reduce cyclic feed forward to 4 and cyclic gain to 46 to get rid of (strong) cyclic oscillations @ 1650 and 1850 rpms (no problems @1450 or lower). no trouble with the tail - perfect from first takeoff. it's flying OK now, but there is still some elevator bounce-back... increasing nick filter to more than 2 to stop bounce-back again is leading to oscillations. the oscillations esp. with higher gains are not only visible in hover, but in flight as i.e. in a funnel too... the landing skids are visibly shaking and when getting stronger (mainly in hover) the complete heli is shaking heavily. when switching back to 1450rpms it stops within some 3-5 seconds. it seems to be more an aileron than elevator shaking clearly... when gain is too high i can start oscillations with small aileron inputs.

it's quite strange as it was flying perfect with Neo before since a year (with +/- standard setup, head gain was high... 84)... no change in technical/mechanical setup today - only Neo & Rex3 out and Spirit RS in. i've 5x checked over and over again the setup. Spirit RS arrived yesterday, firmware was already 3.2.

it is a Diabolo 700 (2018) with rigid (undampened) head, VTX717 main, RT116 tail, swash is perfectly level, blade tracking is +/- 0.1°, 13° pitch, 6° geometry is spot on, servos are BLS471 and BLS276 @7.7V, motor is Egodrift 4525HS 550, 18 pinion, Hobbywing 130HV, Esc gov, takeoff weight 4.92kg. outside temperature was ~8°C today.

logged Spirit vibration level is low. hover 6/9/13 @ 1450/1650/1850 and max. i've seen in 3D flight was ~40 (tictocs). setup file of first bank is attached... other banks not yet modified.

... any ideas on how to get higher gains or reasons why gains have to be that low? how to eliminate completely the elevator bounceback? what could be wrong with the setup?

Ciao
Michael

_________________
cheers
Michael

Hello,

it look like too high Geometry 6° value.
This could be responsible for the issue.

It should be maybe 1/4 lower from your description.
When you enter Subtrim (tuning) are all servo arms in the middle and pitch 0° at the same time?

For this model elevator filter 3-4 should be used.

_________________
Spirit System developer

Hello Tomas,

6° is perfectly 6.0° with rotor blades in longitudinal axis... measured on the blade above the boom or the blade in front.

all servos are perfectly centered (all pushrods are measured +/- 0.1mm same length) and 0° pitch in subtrim menu... swash perfectly leveled.

elevator filter (i called it nick filter in my post - from German name) 3-4 would be great to avoid bounce-back, but is causing oscillations if > 2

_________________
cheers
Michael

Thank you, the issue is really interesting.

I would recommend to make sure blades are not too tight in the blade holders.
Is unit mounted hard at the model? (no wobbles)? For example sometimes a cables or plugs are touching the unit. This can then make the problems.

Also antennas should be ideally mounted as far as possible from the unit - not too close to the unit.

_________________
Spirit System developer

Hello Tomas,

thx for taking care...

the blades are rather loose than tight... normally i fly them tighter when setup is finished - they are too loose for autorotation training in the moment.

the unit is mounted with 1 layer of supplied tape as recommended - see mounting situation in pics attached. antennas are loosely mounted to go downwards (in a bow) below the (extralong) Diabolo canopy. i've attached a picture (next post) of servo geometry too - just for you to check. it's all as per the manual of minicopter. any thoughts?

_________________
cheers
Michael

photo of servo geometry

_________________
cheers
Michael

Hello Tomas,

how should i proceed now? do you think antenna placement is a problem (putting tension on the unit or swinging/oscillating) ?

what else could i try to bring my (by far) best flying helicopter back to it‘s unique stability & precision? i‘ve just done another test flight and it‘s …

thx for your great help!

_________________
cheers
Michael

Hello,

for me it look like the cables are too close around buffer pack. Nothing should touch the cables near the unit.
Antennas are bent little bit more than they should, but this is likely causing not the problem but can damage the antennas.

From photos I would recommend to place buffer pack more upwards or to a different place where no touching is possible. But maybe I am wrong.

These are probably the only things that seems to be odd from the photos.

_________________
Spirit System developer

Hello Tomas,

you're right - from the photo (due to tele-lens) it really looks close with cables & buffer pack. i've checked and in reality there is 4-5mm between pack and the cables. i will anyhow set the pack 1-1.5cm higher as i can better access the cables then... was not really smart to mount it that close. in addition i will mount the antenna over the boom like with the other helis just for at testflight. just to 100% exclude that too... it's last hope

i really have no more ideas... the Diabolo was flying perfect one day before the switch to Spirit - so i would exclude any mechanical or setup problem (and i've checked again and again in the last days). i've done the exact same switch several times in the last weeks and esp. 3 weeks ago with Logo 600SX, equipped with the same swash and tail servos, same mounting, same cabling, resulting in +/- same geometry values (Logo only more pitch): geometry 6°/pitch/cyclic ring - Diabolo: 105/151/144 (11°) Logo: 109/182/142 (11°). and the Logo is flying like a dream with elevator filter 2, cyclic feed forward 6 and cyclic gain 60 (could be possibly much more - i've not tested as it is so great already). the Diabolo is a bit soft & not as precise as used to be and the tail bounce-back is a pain... i know they are completely different birds - but geometry is not that different.

last thing i really want to do is switch back to VBar Neo... loosing the great integration and all the other cool features together with my DS-12 Spirit Ed... and i have to admit i like the tail feeling with Spirit a bit more even i could not 100% exactly say why. but if no one is having a new idea this is most probably what i'll have as only remaining solution

_________________
cheers
Michael

Another option is to try a different Spirit unit you have to make sure it is not related to the particular unit.

Could you mount antennas so they are more or less directly, without a bending? At least for hovering tests.
It is enough to mount ends with a tape.

_________________
Spirit System developer