PDA

View Full Version : KK Tricopter help needed.



Hemmi
13th April 2012, 01:14 PM
I have built 2 tri-copters based around the hobbyking quadcopter controll board v2. Both have the same problem. When I make sudden stick movements the tricopter oscillates badly. It almost sounds like the motors are cutting off and back on rapidly. If I'm high enough I can usually recover it, but if I'm close to the ground, it causes a crash.

For example, I hover the tri about 4 feet off the ground. Then rapidly increase throttle to shoot up, the tri then starts bouncing on roll and pitch rapidly(enough to almost flip).

However if I'm very easy on the sticks with very low rates, it flies super smooth.

I have programed the ESC's according to the kkmulticopter instruction. I have set throttle range, etc. Changing the gyro gain only seems to affect stability, however it does not change the crazy behavior.

The boards are vibration dampened with foam padding, and the props are balanced. I have the same gear in my hexa running Arducopter and they work flawless

I am using the following:

firmware XXcontrol_KR_TriCopter v2.5

ESC's Turnigy Plush 30A

Battery 4000mah 30-50C

Anyone seen this problem before? Please help

Motors KDA 22-20l

Props 10x4.7

Flying Monkey
13th April 2012, 02:08 PM
Sounds like the firmware you're using might be the culprit. I think the version you're using has both P and I loops. So the roll pot controls the gyro gain on both roll and pitch, and the pitch pot controls the I loop. Maybe try adjusting the pitch pot only, start with zero. I never had any success with that version so I went back to the Tricopter v1.6, I think its the one by kaptain kuk. v1.6 works just great.

Exyator
13th April 2012, 02:12 PM
What timing settings do you have on your ESC's?
Try increasing your timing to medium/high if you have it set to low.

Hemmi
13th April 2012, 02:19 PM
Sounds like the firmware you're using might be the culprit. I think the version you're using has both P and I loops. So the roll pot controls the gyro gain on both roll and pitch, and the pitch pot controls the I loop. Maybe try adjusting the pitch pot only, start with zero. I never had any success with that version so I went back to the Tricopter v1.6, I think its the one by kaptain kuk. v1.6 works just great.

Flashing now, I will report what happens

Hemmi
13th April 2012, 02:34 PM
Yes it seems when I adjusted the roll gain the yaw would be messed up. and if I adjusted pitch the roll would be out of wack. So on version 1.6, what is the defauts for first flight? all pots at 50%?

Flying Monkey
13th April 2012, 03:11 PM
Yeah 50% is pretty good. If it oscillates turn the pitch and roll down a tad. Yaw can usually be higher, like 70%.

Hemmi
13th April 2012, 04:12 PM
Ok I took it out on a 20% full battery, so not a great test. After having to reverse the yaw gyro and uping the gains to about 70% all axis. It flew super stable. better than I ever achieved on the other firmware. I did however have another episode where a quick full throttle caused it to go crazy. The roll and pitch wobbled back and forth like 45 degrees till it crashed. After resetting it flew super stable again. However it is only happening on extreme stick movement. nice gentle movements and it flies like an eagle. This is all well and good for FPV, but sometimes I like to scream around LOS. Like I said I have 2 tri's that have this same issue. both use the same motors, but different ESC's. My hexa used arducopter APM with the same motors and ESC's and it has none of these issues.

I'm sure its something simple. But for the life of me I can't figure it out. It acts like the ESC's are cutting because they cant draw enough power. But I use the same motors and ESC's and battery in my Hexa, and it is pulling twice the power.

Hemmi
13th April 2012, 04:20 PM
take a look at this vid. Not mine!

This is kinda what happens, but while flying if I use extreme throttle movements. but as long as I'm easy on the throttle, it flies super stable .


http://www.youtube.com/watch?v=-4SL1wnyZMg

Hemmi
13th April 2012, 04:25 PM
Oh one other strange thing from v2.5 to v1.6. It now takes more throttle to hover. Maybe I need to redo the throttle range calibration. But I don't have instructions for how to do it with v1.6

Flying Monkey
13th April 2012, 04:28 PM
Okay, now I'm stumped...

Does it do this when the pots are set at 30% ?

Flying Monkey
13th April 2012, 04:32 PM
Throttle Calibration

With the tricopter unpowered, move the yaw pot to 0% (counter clockwise). Move the throttle stick to 100%. Power the tricopter, after the first set of beeps move the throttle stick to 0%. It beeps again, unplug it. Move the yaw pot back to 50% or so and repower it and you're set.

It might be that they just needed a recalibration with the new firmware.

Hemmi
13th April 2012, 04:38 PM
I can't seem to even control it that low. Maybe I can lower my dual rates and get her to be controllable with that amount of gain and see. I just can't see why this is a throttle issue. Hard rolls and yaws don't do this. I can hover it and roll 45 degrees each way. the same with pitch. But if I pitch it a bit forward and slam the throttle up ( for that blistering, take off like a raped ape!). It then wobbles and I loose it. But If I fly around easy with no extreme throttle changes, its as smooth as glass.

Hemmi
13th April 2012, 04:42 PM
Throttle Calibration

With the tricopter unpowered, move the yaw pot to 0% (counter clockwise). Move the throttle stick to 100%. Power the tricopter, after the first set of beeps move the throttle stick to 0%. It beeps again, unplug it. Move the yaw pot back to 50% or so and repower it and you're set.

It might be that they just needed a recalibration with the new firmware.

Thanks for your help. Maybe I'm using to much gain. At 50% it was hard to keep it in a hover. Maybe I need more dual rates and less gain. I will keep at it and post my findings.

Flying Monkey
13th April 2012, 04:46 PM
With my turnigy 9x I use the following settings for smooth flight...

Pitch/Roll Endpoints: 60%
Pitch/Roll Dir: 40%
Pitch/Roll Expo: 25%

Trying to fly it with 100% endpoint and 100% Dir is pretty challenging. Fun for aerobatics but HARD!

Hemmi
13th April 2012, 05:20 PM
Ok I calibrate the throttle range. It helped a bit, still a little sluggish on the throttle, seemed faster on v2.5. I tweaked the gyros to about 30-40 percent. dual rates are set at 35% no expo (yet).

Now here is the good news!!!! No more forking wobbles at full throttle. I was tearing up my backyard..kinda scary a large 33inch tricopter with Ginsu blades whipping around a small yard at 20mph.

My only issue now is that the throttle seems soft compared, but this I can live with. Thank goodness I made this a folder, as all the crashed have resulted in no damage, ever less prop breaks now.

Anyway I think for now I can call this issue solved. Here is a pic. So thanks for your help Flying Monkey! Now for the FPV gear :)

9029

Flying Monkey
13th April 2012, 05:39 PM
SWEET!!

Now to get the gains up a bit further I'm going to suggest my vibration dampening method... http://fpvlab.com/forums/showthread.php?4251-Vibration-Dampening-amp-Isolation-Solutions-Guide

The Sorbothane is near $30 for a 1 foot sheet but its enough to last many, many multicopters. It's kind of a funky way to mount the motors but it definately helps. I've had my gyro gains up to 80% without oscillations and the footage from my point and shoot camera is nice and smooth. I'm still refining the method a bit but one thing that has changed since I wrote that guide is that the engineers from Sorbothane wrote me an email and let me know that the sorb needs to be smaller than the plates compressing it. It shouldn't squeeze out the edges. I haven't updated the guide yet though cause I haven't had time to test this on the thrust stand.

I live in an apartment, so yeah, when the kids start chasing my flying cuisinart it scares the crap out of me! Time to land unfortunately.

kalasatu
13th April 2012, 05:57 PM
My quad uses Mike Barton's xxcontroller firmware and I had the same problem. With gentle flying my quad was rock solid. With aggressive maneuvers something caused motors to flip the quad. I played with PID values and now the quad is a bit sluggish, but at least it stays upright.

Same HK v2 control board with 30A plush ESCs.