You'll need 4 rubber rings ~1.5-2cm diameter, ( used on prop-savers ) and 4 M3x10mm Spacers.
Just squeeze it through the hole:
and lock it on the spacer:
you can use 4 extra spacers and mount the top plate:Good flying to you all :)
You need to be a member of diydrones to add comments!
Comments
Ian, yes, it's all really high frequency vibrations. 30hz for a big helicopter, to 200hz for a multi-rotor. And don't forget that we can't even begin to respond to any dynamics faster than 50hz anyway. Yet these ranges work out where helis, or quads with harmonics will create aliasing on the sensors. I have definitive proof that the vibes in my heli were totally messing up my sonar readings.
My next project, I'm going to make an alu plate on which to mount the APM, receiver, and sonar so that they can all be a single mass. Good for dropping the natural frequency, and also good for eliminating as many wires as I can from jumping the gap. Previously I made a super soft mount, but found that all the wires I had jumping the gap was pulling the APM out of alignment. I've even gone so far as to move the magnetometer directly soldered to the Oilpan instead of remote mounted. One more set of cables gone. I'm even thinking of going so far as to run all the servo leads to a remote power header, so that the only servo wires jumping the gap will be the signal wires. That'll leave me with only external power, Xbee and GPS wires jumping the gap.
I think there's a very good reason why the close-source guys have the IMU mounted remotely with a single wire coming out of it...
guys, how about this, three sets of linear actuators, with three accelermeters, with an arduino to do some active dampening, did anyone consider this for camera platforms :)
I still say that Jordi should add some inertial dampeners to the board. ;-)
By phasing the APM2.0 in and out of subspace on every vibration, we would virtually eliminate sensor lag. This is due to the space time compression that occurs while the APM is the subspace bubble.
@max we have seen some designs incorporating this feature e.g. flexicopter . Would have loved to have someone share some insights to the various mechanical vibration reduction systems ( and may be active vibration reduction :) - way too optimistic)
have to admit that it is cool :) as @chris pointed out, now joining the rank of commercial flight contraller with "vibration dampening" using advanced patented materials :)
Given the sample rate on the Gyros, vibration has always been a much bigger problem in this space than lag induced by a suspension mount.
@Max
All filters (transfer functions) introduces lag (phase delay), and in a closed loop, this lag can lead to instabilty of the closed loop system.
One thing is that the lag introduced don't lead to the instability, which as well depends on your system dynamics and your reference signal (as we have a non-linear system). On the other hand for a control system, your phase margin (for a linear system it measure its robustness) is compromise due to this lag.
It is always a compromise, the lag is never a good companion, but the collateral benefits (i.e. removing high-freq in a low pass filter) could be better.
Jonathan,
3-5 times the amplitude mmmm, maybe you are measuring aliasing right? if you are sampling at 50Hz. I am interested in your experiments as I want to try more isolation techniques. Do you mind to share them? :P
@Ian
APM2 board have very little weight, and it will not be effected by the quads dynamics,
Actually adding something that will make some little lag in to the frame reaction is a good thing !!!
(not in the code and electronics, but physical suspension)
Allowing arms to move (firmly) for just a little before effecting the APM2 sensors, will work like filter.
there is more to that, but I need to sleep :)
I should rephrase and say "3-5 times the amplitude". It's hard to sample very high frequency at 50hz :)
-
3
-
4
-
5
-
6
-
7
of 7 Next