I've been having this issue for about a year now, and as of yet, haven't found a solution.
To start with, here is a video displaying the wagging yaw issue:
https://www.youtube.com/watch?v=7amHcd-Ws3A
This yaw wagging only occurs with a top mounted yaw mechanism. I'm sure it's got everything to do with the mass of the yaw mechanism initially moving the tail in the opposite direction of where the controller wants it to go. Simply mounting the mechanism upside down 'solves' this issue since the mass of the mechanism helps move the tail in the direction the controller wants it to move.
I've tested this with Arducopter 2.9.1b and 3.1.2 firmware.
I've also swapped servos in order to reduce internal servo gear slop as much as possible.
When I used the KK2 board on this tricopter, it did not have this wagging issue.
The only downside to using a bottom mounted yaw mechanism is the much increased chance of prop strikes, especially when landing hard, or in rough terrain or braking for a landing. So I really hope this issue can be fixed or that someone can lead me in the right direction.
Want logs? Just tell me what you want logged and I'll let the tricopter wag for you.
Here's my previous request for help from last year: http://diydrones.com/forum/topics/arducopter-tricopter-firmware-oscillation-issues-with-top-mounted
Replies
just went through 3x 4000mah 6s lipos trying to get rid of this tail-waging. No matter which settings I choose it is still there. Went from 1.0 stab yaw to 8.0 also played with P & I values extensivelly. This "tuning" costed me 1x CF 15:55 prop (the first broken one since March) as well as cosmetic damage to other 2.
In the end I "sort of" managed to get rid of the wagging by changing the arm length of the servo to min and increasing the arm length of the yaw assy to max. This gets rid of the wagging issue completelly, however it also decreases the speed of the yaw to the point of almost uncontrollable. Another thing I realised is that on my previous yaw assy I used a very fast servo 0.5s which gave 0 movement under load before starting countering, my current one is quite a bit slower servo 0.14s which I thought should work, but apperently is too slow: it gives about a good degree of movement before the compensatory action kicks in. So my plan is to get a fast 0.5s and see if this will solve my issue.
my previous design rce with 10mm cf booms with wooden cores powered by 380kv motors w/15" props @ 6s lipo didn't have this wagging issue, but lacked mounting space as well as all the wiring was bluntly exposed.
recently I moved to a design incorporating 16mm cf tubes and tarot hardware allows all wiring inside frame/booms.
my yaw mech has changed, now the rotation axis is directly in the middle of the boom, also I have changed the servo, servo arms, control linkages and the linkages connection points, so the airframe is totally different and now and I have the wagging issue. I have tried the full spectrum of "stab yaw p" with no results, turdsurfer, may i ask you if you got it figured out?
It doesn't matter where your yaw axis is relative to the boom, but where it is relative to the CG of the motor+prop. Ideally they should all be in 1 line.
I haven't made a proper yaw mechanism yet. I'm still using the rcexplorer v2.6 design. All I did was dial the stab yaw p down.
The wagging issue can be seen in the video in the original post. These 2 videos below were made after I tuned stab yaw P down, so you can see for yourself what the difference is:
The exact equipment on my frame as well as the arm lengths are in the video descriptions. It's basically a rcexplorer v2.6 frame but with hard wooden arms.
under the last video, in the description you are mentioning "too flexible" wooden booms, I suggest you get some 10mm square CF tubes with a round hole in the middle (8mm) and glue in a wooden dowel in it. I did some tests back in January, this combination was able to hold 10kg in the middle of 500mm boom without ANY deflection. 37mm boom weighted only 38g. also, there were no noticeable "twisting" either no matter how hard I tried to twist it. this is the best boom imho for rce2.6. the only reason I moved to 16mm CF tubes is an ability to hide ESCs in the middle of the body and all wiring inside the CF tubes, also allows for a more rigid yaw mech, as I found out the original David's nylon-hinge design was not "firm" enough for my 40mm pancakes swinging 15" prop at 6s and would develop some slop on the "hanging over" side. btw, nice flying, which RC link are you using?
I've got some extruded 10mm CF tubes already. I was also thinking of glueing wooden dowels into them to stop them from splitting before I use them.
The RC link is Scherrer NR UHF.
do it :) the difference between this kind of boom and even the 13mm (1/2") wooden ones we got in the USofA is very noticeable.
p.s. here is my old yaw (worked perfectly w/o any PID adjustments)
pictured when I was experimenting with hollow square CF extrusions.
servo is a turnigy slimwing servo. looking back at this picture I see that this old one had very "low" rotation radius.
F.y.i.:
An ideal yaw mechanism design should not shift net mass to either side when tilting. That leaves you with a design where the pivot point is as high as possible (near the prop).
I found a picture of one in this post:
http://diydrones.com/profiles/blogs/tricopter-based-on-arducopter-y...
Interesting! Should improve tail response even more...
Also I wonder if those code changes were ever implemented into the Arducopter? Edit: I read a bit further and it seems a quite lot has gone into the code for Tri's which is probably why they fly so well :)
My tricopter flies like a dream now!
I used the values as stated above.
Yaw stabilize: 1.2
Yaw p rate: 0.280
Yaw i rate: 0.038
Yaw d rate: 0.010
Works like a charm!
-
1
-
2
of 2 Next