Ecilop dynamic calibrations

[SuperSharpShooter]

I've just done the dynamic calibration on my ecilop framed quad, and I'm wondering if it's acceptable.

The thing with the ecilop frame, is that when rotating it to +/- 90 degrees, the floating gimbal arm will sometimes swing back the other way, or oscillate a little due to it being controlled by independent gyros which the servos are connected through. During some of the movements and transitions, there were some gyro/servo induced wobbles, but would these affect the readings enough to cause problems? Or is the only way to find out to run the calculations and see if the performance is good?

Here's my log anyway.

[LPR]

Super

You wont be able to have your AQ board centered on a ecilop frame and because the AQ uses the accelerometers a lot, it will make a difference in how it flies your multi will fly.

Larry

[SuperSharpShooter]

Yeah the board is mounted off to the side. Oh well I'll find out how it flies tomorrow with a bit of luck!

[LPR]

Super

If the board is mounted off to the side your AQcopter will likely rise if rolled toward the side the board is mounted or lower if rolled the other way. If you have the board in the rear your AQ C will rise as you stop and lower as you accelerate. I mounted my board on the rear once. Other autopilots do not have this situation because they don't rely on the accelerometers as much as the AQ. The newest firmware for Arducopter is now using some accelerometer sensing also.

Larry

[SuperSharpShooter]

Ok, yeah I'll see how it goes and if it's usable. I just had an APM2 on it, and it flew fairly well, but I'm after the precision and stability of the AQ board.

One thing I noticed today during a hand test, was that the control inputs were a little sluggish, and it didn't seem to respond to my yaw input. I haven't had a chance to review the logs yet.

[kinderkram]

Hi Shooter,
team member teramax already tested the AQ6 on an Ecilop - with not so satisfying results:



Some pics of his setup:
https://picasaweb.google.com/1109859514 ... 354/Ecilop

[SuperSharpShooter]

That doesn't look too good

I've put the APM back on the Ecilop for now.

I think I'll build a rusty UAP2 hex or y6 for the autoquad, and use it with an AlexMos gimbal setup.

[LPR]

Shooter

The new firmware for APM is using the accelerometers and might have some of the same problems as AQ.

Larry

[SuperSharpShooter]

I spent some time flying yesterday with the APM in manual, alt hold, and loiter, and didn't see any such behaviour.

I wonder if it's within the realms of possibility to provide an offset to the AQ board which will allow it to be mounted off centre and perform correctly?

[bn999]

Quite possible if one wanted to spend the time to figure out the math. You would need to translate the raw IMU data to the center of rotation of the craft before it is published to the rest of the system. The GYO rate measurements I don't think are affected as the rotation looks the same from the perspective of any location on the craft. The ACC's will be affected due to centripetal force. I'd have to think more about the MAGs, but I'm inclined to say that they will not be affected either.