Initially I started with this setup:
Now with top shelf added:
When I attached the top support sheet, removing the receiver from the top of the APM itself, and moving it to the top shelf, my vibrations increased quite a bit.
They went from this:
I'm thinking that with the receiver on a non-vibration shielded part of the quad now, the vibrations are being carried through the wires to the APM.
I'm going to try a different type of O-ring first, to see if that will have a positive effect on the vibrations. Switching from the red silicone o-ring size 017 to a smaller nitrile o-ring size 015.
Also, in the intial setup I had the APM attached to the base via a direct velcro strap. I've now added some silicon gel underneath the strap so the APM is only touching its supports through vibration dampening gel.
Now awaiting a battery charge to see how the new setup performs....
Alrighty. Got the new o-rings. Didn't realize what a difference twice the cross-sectional diameter would make.
Even though it looks monstrous, it fit nicely on my new acrylic base. (That's my original wooden test base in the photo, with the small original o-rings)
Had to get some more standoffs to fit the larger sized o-rings. Also had to double stack the gel pads so the APM wouldn't ride directly on the o-rings.
Overhead view of new base.
With the APM attached and strapped down. The black and white wirs coming off the right aref or my new external led arming light. It's hiding beneath the input wires.
Here's an oblique view. You can see the huge 10mm led under the input lines (once again, didn't realize it would be so big when I ordered it).
You can see the gel pads underneath from this side shot.
It's a bit rainy today so I'll have to wait a day or two to see how my changes have affected the vibrational characteristics.
I would prefer off/on to be honest...
I'm still working on the large o-ring setup, but i had an apm 2.5 question. I've wired up an led to AN4 for an external indicator of when I'm armed or not. Right now it blinks slow when disarmed, and blinks fast when armed. I was led to believe from my research that it would be on solid when armed. I'd like something a little more definitive than fast/slow blink to know when my flying lawnmower is ready to start chopping. I'm using 2.9.1b firmware.
Great Scott, sounds like your already in the very acceptable range.
Generally silicone does work best for O-rings because they actually have better vibration damping characteristics.
And the combo of gel pads and o-ring generally gives good results.
You are probably right about the acrylic intermediate platform too, a little extra mass in the isolated component actually helps.
One last thing you might consider, a slightly thicker O-ring on the not so tight side definitely works better than a thinner O-ring.
So if you go up 1/32 or even 1/16 in cross section it can improve damping.
I have a bag full of different compounds, sizes and thicknesses so I can tune my copters for the lowest vibrations.
And it is how I get below 1/10 G.
(F450 and F330).
Was also considering replacing the balsa plywood with an acrylic square, figuring it would have a bit more mass and perhaps perform better from inertia alone. But have to order slightly larger o-rings since the acrylic is about 3 times thicker than the balsa plywood, and i don't want them to be too tight like the nitrile ones were.
Went back to silicone o-rings, also placed the receiver on gel pads, with a gel pad on top so the velcro strap holding it on doesn't actually touch it. Back into the +/- 2 realm.
Next I guess is balancing each motor/prop combo individually. Previously each motor and prop was separately balanced, but I was holding off doing both together cuz I'm lazy. Ah well, here we go....
In my poor experience: The flame wheel is not solid enough to dumping vibrations from motors and other external energy source, such as wind. The o-rings tend to cut off the vibration, on the other hand could not dumping the vibration on the APM. I found your nice acrylic plate that is best place to attach APM with 5mm double side adhesive laminated sponge. Actually, you need to find the place for GPS. Other than the vibration isolation and dumping, the balance of propellers are essential.
I use o-rings doubled over through the corners of the APM 2 board combined [with a block of visco-elastic foam (like temperpedic) beneath the board under some compression. Works very well for me.
O-rings are tricky to use, need to carefully balance tension to acheive best results.
Too tense and they just tie the board to the frame, too loose and they let the board move around too easily, but when you get it right you can acheive 1/10G (+ and - 1 accel values) results.
It looks to me like your intermediate board might actuallybe hitting your standoffs at the corner though I cant be sure fro your photo and also, you need to put a little foam under your velcro strap if you are going to get any value from the foam you have the case sitting on (and not tighten it too tight.
In any case, the even newer new vibration section is now on the new Wiki here: http://copter.ardupilot.com/wiki/vibration-damping/