|May 25, 2013, 04:45 AM|
Joined Mar 2013
Balancing multirotor for better flight?
I have been thinking about doing this with my Hexa in X config and have come up with this:
suspend multirotor from centeral point by single attachment point and balance acordingly covering all aspects X axis, Y axis and all inbetween? (use small bubble level to ensure accuracy).
I had thought of balancing on point below but landing gear would get in the way?
Once ballanced it would also be a good time to calibrate sensors as when copter is balanced it would be perfect time to do so!
What are your thoughts?
|May 25, 2013, 02:37 PM|
Joined Feb 2013
Wretched excess. Just make sure everything heavy (batteries) is roughly centered and if the basic build is centered, you'll be good to go. If the gains are correct, the EFC will cover any remaining small imbalances with no issues. If you want to be really cranked on this, check motor temps after flight with a non-contact IR thermometer gun. Any significant imbalance will show up as side-based temp imbalances as a result of the EFC leveling the copter, which in turn, results in increased motor drive to a specfic axis.
|May 25, 2013, 02:57 PM|
Assume for a minute that you’ve chosen to build symmetrically; that you have all components on one side approximately balanced out against the same components on the opposite side. For example the forward ESC and motor are symmetrically opposite the aft ESC and motor. Assume that you have no camera and transmitter forward with nothing aft to balance them. During PID tuning I’ve had a quad copter built that way where attention was paid to stay symmetrical. This quad sat on top of a hemispherical gas bearing which meant that any imbalance would tend to tilt the unpowered multirotor over. It became immediately obvious that no matter how much attention was paid to symmetry, the battery position (sloppy) by in large completely dominated the balance condition. Makes complete sense too with the battery being the largest mass whose position can shift.
Now to your question; let’s arbitrarily assign the X axis to run forward/aft, the Y axis to run left/right, and the Z axis to run vertically (parallel to the gravity vector). If you suspend the multirotor from a string and your CG happens to be located below along the Z axis, the pendulum effect will mask any CGx or CGy offset making it hard to detect such an X Y imbalance. The bubble level itself will upset the balance too if located to one side. Move the bubble level around from one side to another. If it does not change reading either your bubble level is light enough not to matter or your CGz is so low as to mask its effect. Which one is it? Perhaps a “leveled” camera can replace the bubble level and you can analyze level the condition via photographic means.
Many of these factors can serve to corrupt the balancing attempt perhaps more than attention to symmetry can preserve. It all comes down to a matter of degree I suppose. How imbalanced is the craft? How sensitive is the tuning? How close is your string’s attachment point to the Z axis CG location?
Really though, it is such an easy thing to try, why not simply try it and report back? Suggested alternatives: telemetry data can be useful here; is one motor habitually rotating faster than the others during extended hover. Do you have a cheap IR gun? Is one motor hotter after a long hover flight than the others?
Sorry to be redundant; Electro2 posted while I was typing....
|Category||Thread||Thread Starter||Forum||Replies||Last Post|
|Question||Multirotor Flight Controller for use as a gyro stabilization platform...?||macinr||Dynamic Soaring||13||Apr 25, 2013 10:52 PM|
|Discussion||Feels like each sport 3d flight gets better and better||wc_wickedclown||3D Electric Heli Flying||0||Mar 20, 2013 06:29 PM|
|Discussion||dynamic balancing for multirotor helicopters||18t||Multirotor Talk||7||Oct 11, 2012 08:58 AM|
|Question||A good prop balancer for multirotor-size props???||arch001||Multirotor Talk||11||Jun 10, 2011 03:37 PM|