RC Groups

RC Groups
    Multirotor Talk
        Help! HK KK v2.1 - can't exceed 30% gain

#1 avt Oct 13, 2012 11:23 AM

HK KK v2.1 - can't exceed 30% gain
 
I'm trying to tune my first quad (Hobbyking's Q450), and am facing serious oscillations when the gyro gains are increased above 25-30%. At %25 gain (roll and pitch) it hovers more or less smoothly. But obviously there's not enough gain for it to balance itself.
From what I've read, this should be a problem of unbalanced motors and/or props.
I balanced my motors and props (at least I think I did it right), but the oscillations are still there.
When I hold the frame in my hand and throttle up the motors with no props on, I can feel some vibes through the frame. It's not monotonic, more of a sinusoidal pattern.
Could these vibes affect the controller so badly? Are these vibes the source of my problems with the quad (and perhaps in other parts of my life? :rolleyes:)?

UPDATE: Problem is pretty much solved after I flashed the ESCs!
After trying every possible solution, balancing every part again, I went to flash the ESCs (BlueSeries 30A) with SimonK FW.
Amazing improvement! I couldn't believe it would make such a change. It's like a completely different quad now. VERY responsive, VERY stable.
Thanks again for your help.

#2 guyg Oct 13, 2012 12:15 PM

Hi
can you write your settings here
as for P .gain/ limit
I Gain/limit for 3 chanels


Tx
Guy

#3 WiseDuck Oct 13, 2012 12:29 PM

And how the controller is mounted.

#4 avt Oct 13, 2012 01:12 PM

1 Attachment(s)
It's glued to the foam which is glued with a double-sided tape to the body.
Then there's a zip tie which holds it in place. I tried my best to make it less prone to vibrations as possible.

About the PI settings, I got this controller, flashed with KK v4.7. I'm not sure how to tell the exact P and I.
All I can say is that the roll and pitch pots are at about %25-30 from minimim position.
The yaw pot is at %50.

#5 oyibox Oct 13, 2012 01:45 PM

Quote:

Originally Posted by avt (Post 22990375)
Could these vibes affect the controller so badly? Are these vibes the source of my problems with the quad (and perhaps in other parts of my life? :rolleyes:)?

yes, yes and yes. sorry but you'd better try another frame :) or at least make sure you've tighten evrything

#6 2400RDR Oct 13, 2012 02:08 PM

As I understand it, in the 4.7 firmware, the roll pot controls P for pitch and roll, the Pitch pot controls I for pitch and roll, and the yaw pot still controls the yaw.

Suggested initial setup is roll pot = 50%, Pitch pot = 0% and yaw pot = 50%.

Here is a post from the Kaptain himself describing it:

http://www.rcgroups.com/forums/showp...5&postcount=73

Give this setup a try and see what happens. By the way, zero on the pots is fully counterclockwise. Also, you should realize that this board doesn't self level- you trim it to be as level as possible, but you still have to fly it. It looks to me that the zip tie you used may be defeating the vibration isolation effects of the foam. Try getting rid of that too.

Rick

#7 antiguogrumete Oct 13, 2012 02:30 PM

Same for me
HK Flight Controller KK2 Auto Level (1 min 37 sec)

#8 avt Oct 13, 2012 03:19 PM

Quote:

Originally Posted by 2400RDR (Post 22991319)
As I understand it, in the 4.7 firmware, the roll pot controls P for pitch and roll, the Pitch pot controls I for pitch and roll, and the yaw pot still controls the yaw.

Suggested initial setup is roll pot = 50%, Pitch pot = 0% and yaw pot = 50%.

Here is a post from the Kaptain himself describing it:

http://www.rcgroups.com/forums/showp...5&postcount=73

Give this setup a try and see what happens. By the way, zero on the pots is fully counterclockwise. Also, you should realize that this board doesn't self level- you trim it to be as level as possible, but you still have to fly it. It looks to me that the zip tie you used may be defeating the vibration isolation effects of the foam. Try getting rid of that too.

Rick

Thanks Rick, I just tried these settings. Got about the same results as before.
I also read that zero on the pots is fully counterclockwise, but for some reason this is not my experience. When the pots were at 50% and I wanted to lower the gain, I had to turn the pots CW, and not CCW.
I'm sure my gains aren't reversed. I guess if they were reversed I wouldn't stay in the air more than 2 seconds. I double checked them.
Also, when I tried to set the I pot at 0% (which should be fully CCW), the quad pitched down like crazy. I set it fully CW and the P pot to about 30% from fully CW, and it's back to normal.
I don't know what I'm missing here.

#9 2400RDR Oct 13, 2012 03:39 PM

Quote:

Originally Posted by avt (Post 22991772)
Thanks Rick, I just tried these settings. Got about the same results as before.
(EDIT)
I don't know what I'm missing here.

Just curious- did you do the ESC calibration?

Rick

#10 avt Oct 13, 2012 04:36 PM

Quote:

Originally Posted by 2400RDR (Post 22991905)
Just curious- did you do the ESC calibration?

Rick

Yes I did.

#11 kaz1104 Oct 13, 2012 07:04 PM

Some time ago I bought a "vigilante" frame from www.468.com, and after rebuilding this unflyable quad several times, the only variable left that I had not changed out was the frame itself !!. Seems the problem was, that even though I had balanced the propellers and motors vibration was badly messing with the flight controller. The solution was to cut some aluminium strips to act as doublers/laminate on the original arms. They bolt to the center frame with the arm mount and extend out to bolt to the motor with the original arm. It is now one of my best flying frames ... amazing change. This mod should also work for the Q450 frame in reducing, or at least changing, the resonant frequency of the arms of the frame. Flashed speed controllers will also generally allow you to use higher gain settings.

Mark

#12 avt Oct 14, 2012 02:07 AM

Quote:

Originally Posted by kaz1104 (Post 22993198)
Some time ago I bought a "vigilante" frame from www.468.com, and after rebuilding this unflyable quad several times, the only variable left that I had not changed out was the frame itself !!. Seems the problem was, that even though I had balanced the propellers and motors vibration was badly messing with the flight controller. The solution was to cut some aluminium strips to act as doublers/laminate on the original arms. They bolt to the center frame with the arm mount and extend out to bolt to the motor with the original arm. It is now one of my best flying frames ... amazing change. This mod should also work for the Q450 frame in reducing, or at least changing, the resonant frequency of the arms of the frame. Flashed speed controllers will also generally allow you to use higher gain settings.

Mark

Would you be so kind to post pictures of the frame, so I can see where the aluminium goes?
I'd appreciate it.

#13 kaz1104 Oct 14, 2012 02:39 AM

4 Attachment(s)
Quote:

Originally Posted by avt (Post 22995542)
Would you be so kind to post pictures of the frame, so I can see where the aluminium goes?
I'd appreciate it.

This is a different type of frame but the principle is the same. On the DJI or clone arms, the top of the arms are flat and the aluminium could go there ??

#14 avt Oct 14, 2012 02:54 AM

Quote:

Originally Posted by kaz1104 (Post 22995633)
This is a different type of frame but the principle is the same. On the DJI or clone arms, the top of the arms are flat and the aluminium could go there ??

Thanks Mark. I will go look for some aluminium plates and see if that helps.

#15 jseyfert3 Oct 14, 2012 03:13 AM

Quote:

Originally Posted by 2400RDR (Post 22991319)
It looks to me that the zip tie you used may be defeating the vibration isolation effects of the foam. Try getting rid of that too.

That was what I thought too, the foam isolates the vibrations but they'll just come right through the zip-tie.


All times are GMT -5. The time now is 07:17 PM.