Im in two minds about mounting on rubber rings and particularly if it allows the board to wobble even ever so slightly . Any movement at all of the board relative to the crafts frame I suspect will cause oscillation and probably jitters during quick maneuvers. . I'm more in favor of adding vibration dampening pads to the arms and frame rather than suspending the AMP board. Take for example if the motors are causing a horizontal circular vibration from the motors and this circular rotation is say clockwise, then this is without doubt is going to affect the yaw if the board is moving differently to the frames yaw. So if anything I feel the oring idea is in some cases more counter productive than what its trying to achieve. The APM board in my view must be rock solid fixed to the frame to be able to quickly and accurately read what the aircraft is doing. Any compromise even if only very minute will without doubt affect the stability of the aircraft. We have to realise that if the board can be so easily affected from such small amounts of vibration and it is so sensitive, then the idea of APM board suspension mounts possibly is not a good idea
I feel some of the people having trouble with the PID tuning could be because their boards are moving and are not firmly fixed the the frames.
One method I have used to dampen vibration is to wrap a metal rod in some soft rubber and then slid this into the arm tube just under the motor. The rod floating in the rubber wrap takes quite a bit of the high frequency vibration out of the arms.
Just my thought on this.
Comments
Ryan, A hard drive doesn't rely on its movement to help compute the hardware position so its not an issue to suspend it. The main issue is if the anti vibration system such as suspending the APM board in any way that allows it to move differently to the frame, then the on board gyros are getting incorrect input feed relative to where the position and attitude of the frame is. For example if you descend quickly and then stop, the APM might continue down slightly more than the frame causing the board to bounce , which then transmits this to the gyros and then the aircraft bounces around in response in an attempt to correct it, and this can increase and cause an ongoing oscillation even to the extent the aircraft would be virtually unflyable. I am looking more into what you suggested in using mechanisms to eliminate vibration in the frame it self rather than isolate the APM board.
Also I believe the more parts you 'insulated' with rubber, the more mediums the vibrations would have to pass, and each change in medium would continue to remove lower frequencies... but I forgot why I believe this, likely trying to apply thermodynamic principals to a mechanical situation. I am not an ME.
I'm a noob with quadcopters but I know with hardrives, vibrations can greatly affect performance. There are a variety of solutions, but often the best thing to prevent hard drives vibrations from affecting each other is to attempt to get the hard drive to float, in rubber mounts.
That being said, I believe that preventing direct metal on metal contact in different spots on the quadcopter, would greatly reduce vibrations. Take a foam pad, or small orings, and place it between each and every piece of metal plate, and screw, on the motors or vibration source(fans). Ideally, putting it in between the arm and the board mounts would also be helpful, but the ease of that would come down to your frame.
Just my thoughts, thats how I would approach the situation to remove vibrations, though I have little experience currently.
Robert this is good news.. so we are definitely onto something here and good to see others popping out of the wood works with this valuable information. When time permit I want to advance more on trying different systems to eliminate the vibration at the motor points or on the arms rather than at the board and try to keep the board as firm as possible to the center frame, which I feel might give the most optimum response efficiency in flight control. I think it is important the hardware needs to be optimized as much as possible before we even look at PID tweaking.
I've actually done some experimental work in this field with a somewhat scientific approach, using dampers of varying stiffness. My experiments basically proved my theory.
That is, that soft dampers can work, as long as the natural frequency of the APM is lower than the major vibrational frequency, but higher than the response frequency of the aircraft.
I've used mounts that were too soft, and absolutely caused control oscillation. Really badly. Like paint-shaker bad.
But a minor change to move to a stiffer damper, but one that is still relatively soft, and capable of absorbing a big amplitude, worked very well. Even with fairly severe rotational vibrations, the APM had no trouble keeping orientation, and had no problem with control response. The airframe basically shakes around the APM, and the APM determines the "average" orientation of the airframe.
One key to this however, is to make sure that your system does not allow "whipping" of APM in one direction, while still seeming to be stiff enough in the others. Basically what you need is a wide-set support platform.
Martin have you used 3 blade props on a quad? I have some here but they are badly balanced and Ive not got round to balancing them and try them yet.
I use double sided tap for the APM mount which in my view has enough cushioning but there is no movement of the board. I think what you are going to use look good as well. Be careful mounting motors on this though. There is a lot of torsional forces on the motors and any movement of them on the mounts will give erratic flight behavior .
Ill try get some posted later Scott. Basically its just a metal rod rolled up in some rubber sheet and this is inserted into the arm tube just under the motor. Though my feeling is the majority of problems people have with vibration is from either poorly mounted APM boards or other factors such as baro pressure interferences from prop wash etc.
Also three blade props spinning will have less vibration than a two blade prop while the frame is required a movement. When the blades are horizontal with 2 they have a harder time adjusting to a new axis vs a 3 blade propeller.
http://www.mcmaster.com/#catalog/118/3554/=h5zo10
I'm building my first hexacopter but it seemed like a good idea to use this between the arm and motor plate as well as the center plates.
-
1
-
2
-
3
of 3 Next