Align Multicopter Align G3 tuning

[Shelipso]

I posted this initially in fb but will continue to report here. I had a rough start on my m690 but will not go through all the problems. Suffice to say that I was not one of the lucky ones.

Had a flight today with my M690/GH3 and GH4/12-35. I changed the props to T motor 15x5 timber self tightening to potentially reduce the vibrations caused by the folding props. The new ones are well made and m690 flies smoothly, motors not too hot and hovered well in wind. I think I will keep these props for now.

I am now trying to tune my Gimbal PIDs. I am using the voltage booster and use a three colour combination dampener on each corner (red,black,grey) . My thought was to try to dampen vibrations at a wider range of frequencies using the three different softness (pls correct me if I am missing something). The recommended stock PIDs take my gimbal and the camera into an instant shake. So even on the ground, I could not use the recommended setting. I lowered all gains to about 50 and for this round ended on these numbers:
Roll 50/90/100
Pan 50/80/80
Tilt 40/70/70

The result is much better but still not good enough.There is a little bit of jello which generally kicks in on moves. It feels like a pan vibe so I will now focus on pan/roll. I am intending to lower the gains on FC a little bit too.

While not perfect, this has been a significant improvement on my first flights. Fingers crossed.

 

[Shelipso]

And here the response I gave to a question on FB which asked if I had tried to balance the folding props...

... to be honest no. I do not have a setup to balance folding props and in recent months have flown my other setup with self tightening props. I decided to focus on the gimbal and m690 itself with the props I am confident of. The other issue I have with the folding props on m690 is how they work with the m690 type of connection of the arm to the main frame. It uses two bolts in the centre of the tube. My theory is that the resulting arrangement is prone to lateral vibrations. Dji s900 or Zero UAV or other arrangements that use a clamp around the tube provide a better dampening in the horizontal plane than the connection detail used by Align. Consequently with m690, removing prop unbalance is even more critical than what you would need on other systems. I decided that it was easier with a single prop. All said, this is only a theory and I don't want to unnecessarily divert the discussions here! Thx.

 

[Shelipso]

Just to keep it in the records, I have progressed the tuning and it is now much better. I added a thicker layer of foam between my OSD/TX and the frame to isolate the TX fan vibrations. That does reduce vibrations on the frame a little bit. I also used slightly softer dampener combination. Now using two reds and a grey on each corner. Finally and most importantly, I arrived at a better PID setting for the Tilt. Here are the final numbers at the end of this round. The video is now almost vibration and jello free. I have to adjust the shutter speed and potentially use a ND filter to get it better but for now I am done with PIDs.

So for summary:

With a G3 and GH4 Camera with 12-35, I have stock PIDs on the APS-M and the following PIDs on the gimbal:

Roll 70/90/70 (P/D/Pw)
Pan 50/80/60
Tilt 80/90/40 (* this was the main area of change).

 

[Denny Rowland]

I now have the high voltage conversion working OK. As a general comment I would say that with the 2.2 firmware the P values need to be reduced a bit from the recommended settings.

 

[binlagin]

Would love to see some unedited footage.

I cannot get rid of the tilt vibrations that seem to plauge almost all footage i've seen

 

[Denny Rowland]

Your balance may be the problem. There is a need to take up the slight movement in the side arms. Don't let them sag against the stops. What lens are you using? The secret to getting it to perform correctly is in the calibration. As a general rule I run the start up three times before it settles down. It should rotate in the yaw axis without any cross axis movements. I have now also started using the 360 pano mode with 24 shots and it is working great. Tilt settings with the 20mm lens are P 30 and I 25. Power 60. With the camera running at 96 fps it will show any flaws, there is no need to post stab. only colour grade. Post stab. won't cure vibration or jellow.
In my experience the vibration that you may be seeing is from the tilt motor (too much P and I).

I am using very hard mounting rubbers, (Not standard) Jellow usually comes from secondary mass oscillation. ( model and gimbal assy. resonate together) Too soft will always be a problem with powerful motors.

I have had a different problem, as of yesterday. Whilst filming at a great height I started to descend and the motors all stopped. It fell about one hundred feet before I got them running again at full power. This did not fully arrest the descent and the model crashed into a bush causing slight damage to the retract units. I heard the motors screaming on the way down so I can only assume that it was a classic case of vortex ring. A 5Kilo Quad has a high prop. loading. I have known people say this can happen but it was the first time I have seen it myself. The worst part of the event was when a highly dangerous snake tried to tag me as I rescued it.

 

[Denny Rowland]

You may also find this useful.

 

[Denny Rowland]

Align have also released a new APS-M signal wire set. About time too as the original ones were crap. I have now proved beyond doubt, that my recent crash was due to a faulty cable.
When you peel back the rubber sleeve you will find a nice stress raiser in the form of a dollop of glue to hold the cables in place. However due to the location of the controller and the power unit, this cable is impossible to fit without some bending at the fracture point. View attachment 24590
That red wire is the main supply voltage to the APS-M and the Rx, it gave an intermittent signal and that caused the shutdown.

 

[Denny Rowland]

Since Align asked me to return the damaged parts of my M480 and gimbal I have heard nothing at all from them.
Shades of early DJI problems all over again it would seem.
On face value this product as such, is not safe to put in the air at this moment in time....