I'm looking at buying a gyroscope and I was looking at either
they say they need to be used with their shield. Would they still work without their shield and would we get accurate values?
Also i understand that the main differnce between the board is the sensitivity. 1X or 4X. Can someone explain this?
Are they any other gyroscope worth exploring of buying. I was looking at http://playground.arduino.cc/Main/MPU-6050
but got somehow discouraged after reading the article as it does not seem very user friendly.
I don't know much about TinkerKit, but it seems that this is an analog sensor which output is directly available on the board pins . There is no reason you could not use this with Arduino (maybe you can not just plug and play with Arduino because of the pinout).
The MPU-6050 has both an accelerometer and a gyro, it could be interesting if you know how to use it Take a look in the data sheet, it probably has a I2C interface (Arduino Wire Library) which is not difficult too use, perhaps just a little bit more difficult than a analog sensor.
Which is your application, or your level of experience?
What I want to do is to embed the board and the gyroscope into a cube then determine the orientation of the cube thanks to the gyroscope.
So I guess you are right a 3 axis gyroscope might be what i am looking for. Any recommendations?
I need something easy to implement like the one from tinkerkit. Where I can get the analog value from the sensor but for the 3 axis. I don't think tikerkit does it. Does anyone else provide such a sensor?
kinrou:
they say they need to be used with their shield. Would they still work without their shield and would we get accurate values?
Tinkerkit is just a way of connecting things to Arduinos. The connector just has 5V, GND and pin connection in a single plug.
You can easily connect it without the shield - cut the plug off and connect the three wires to 5V, GND plus an Arduino pin. It will work exactly the same.
kinrou:
What I want to do is to embed the board and the gyroscope into a cube then determine the orientation of the cube thanks to the gyroscope.
On what timescale? Gyros only give an estimate of rotation speed, and the orientation
has to be computed by integrating the outputs (for 3D rotations you need 3-axis and
you need to use DCMs or quaternions to integrate orientation changes).
For timescales of a few seconds gyros will do quite well, but over a period of minutes
the drift will corrupt the orientation state inexorably, and 6 DoF sensor is required
(3-axis accelerometer, 3-axis gyro) - even so the yaw will still drift (3-axis magnetometer
is then added to make a 9DoF sensor.
If you don't need analog orientation information then a set of tilt switches could be used.
Yes I had thought about the set of tilt sensor. But it falls apart when you want to rotate the cube on itself.
It would be nice to have analogous information rather than just a on/off state.
Do you have any recommendations for a 6 DoF sensor which give analog output and work well with arduino and possibly come with code sample in how to use it.
You can actually do most of this with a 3 axis accelerometer, the only thing it cannot tell you is rotation about an axis drawn through the cube and extrapolated through the centre of the earth. Any other change in orientation can be sensed using a three axis accelerometer just working with the acceleration due to gravity.
To clarify, if the cube was sitting flat on the ground and it spun about the Y axis, you could not sense that, but any change in another plane would be sensed.
To sense the spinning about the Y axis, you'd need to add at least a 3 axis magnetometer, although a 9 DOF IMU might be a better option then.
The 6050 will do what you want to do, measurement wise. What are you going to do with the result of the measurments? That's is where it can get really tricky.