LibrePilot Forum
General Category => General Discussion => Topic started by: quadfun on February 26, 2016, 09:10:37 am
-
Need some help, guys. I'm building a F450 clone. CC3D + 15.02.02 build. Went thru vehicle setup wizards and things seemed fine. I clicked on the scope tab, thing looks crazy to me. I've done the level calculation and I'm on a very level and stable ground. I think it's not the sensor actually gone bad, as it was fine (everything 0 and stable) before I did the firmware update (with erase all settings checked) in vehicle setup wizards. Really need some pointers on what went wrong. I dare not to fly it with this kind of crazy sensor readouts.
-
forgot to say, I did try librepilot as well as openpilot. same thing. see screenshot for librepilot.
i'm completely new to this, appreciate any help!
-
Hi,
Nothing wrong, look at scale on left side.
This "crazy sensor readouts" are just noise and scale change automatically.
Move your frame and look what appends.
Welcome here,
Laurent
-
The gyros are OK I understand. But the accel.z looks to be -10 m/s^2, as in close 1g free fall! Also the attitude.yaw seems to be drifting at a pretty fast rate! Could someone help check your sensor readout from scope and paste a screenshot? I recall before I updated everything was basically a flat line at 0.
-
The Accel.z is close to 10, should be 9.81 = gravity
There is drift in Yaw axis because there is only Gyro here (without mag)
-
Here is a "normal" scope
(https://forum.librepilot.org/index.php?action=dlattach;topic=930.0;attach=1723)
-
thanks, f5soh, so you are saying this is absolutely normal? ok, I understand the drift part with no mag, but why is accel.z -10? sure there is gravity, but there is an equal force of the floor pushing it up, such that the quad is still and isn't accelerating. so what would accel.z be when the quad is in a state of free fall?
-
thanks, f5soh, so you are saying this is absolutely normal? ok, I understand the drift part with no mag, but why is accel.z -10? sure there is gravity, but there is an equal force of the floor pushing it up, such that the quad is still and isn't accelerating. so what would accel.z be when the quad is in a state of free fall?
Hey, this is perfectly normal :) it's gravity ;) Link http://www.starlino.com/imu_guide.html explains how digital accelerometer works and citing
"In this case the box isn't moving but we still get a reading of -1g on the Z axis. The pressure that the ball has applied on the wall was caused by a gravitation force. "
-
thanks, f5soh, so you are saying this is absolutely normal?
Yes
so what would accel.z be when the quad is in a state of free fall?
I think the accel.z should be close to 0 in free fall, assuming there is no air drag and is falling since quite some time.
Gravity is also used to determine the Attitude state Roll/Pitch.
Without gravity (and his -9.81 value) how do you determine if level ?
-
thank you guys. i must misunderstood how accelerator works. it's not reading the acceleration of the quad, its reading the force needed to counter the acceleration to 0 (my guess).
-
Einstein said that there is absolutely no difference between the pull caused by gravity on the surface of the earth and the pull caused by a 1g acceleration in a rocket in outer space. The accels read the same in either case. :)