Vibration Sensor

[MrMikey83]

I am looking for a vibration sensor which will give a voltage output of 2.5 volts at rest.
I need this for a project where the vibrations from the motors are messing with the reading from an accelorometer. Would the Vibration sensor be able to cancel out the bad readings from the Accel?
Thanks
~Mike

 

[Klaus]

I am looking for a vibration sensor which will give a voltage output of 2.5 volts at rest.
I need this for a project where the vibrations from the motors are messing with the reading from an accelorometer. Would the Vibration sensor be able to cancel out the bad readings from the Accel?
Thanks
~Mike

Mike, do you mean the vibation sensor puts out 2.5v when there are NO vibrations? How much output would you expect then when vibrations do occur?

Actually, your accelerometer isn't really giving you "bad" readings, it gives you the acceleration, even if its caused by unwanted vibrations :wink:
What I would do is use another accelerometer as a vibration sensor and use its output inverted with the first accelerometer, hopefully you can then cancel the signal caused by the vibrations. A bit like the headsets that cancel noise with two microphones.
Klaus

 

[MrMikey83]

My other sensors, Accelorometer and Gyroscope both put 2.5V at rest. Then, they go up or down by about 2V depending on which way you move it.
Having two accelorometers seems like they would just cancel each other out. Because if one is supposed to cancel out the other...and they both sense the same movement, i wouldn't get any reading. Right?
Or...what if I put one Accelorometer turned 180 degrees so that they both read vibrations going the same way and can cancel each other, but the actual 'tilt' reading i'm looking for will be opposites and I will be able to distinguish the tilt from the vibrations!
This way sounds like it will work, please tell me if you dont think it would work.
~Mike

 

[MrMikey83]

No comments huh?

Now that I think of it, just turning around one of the sensors and using one to negate the other will also cancel out the tilt info along with the vibration info.
*sigh*
So...any ideas of how I can use one or two Accelorometers to accomplish canceling the vibrations??
~Mike

 

[LilGMan]

Don't know if you still need help on this issue... almost a year after your post. I think you can use a low pass filter on the output from your accelerometer so that only large scale signals (i.e. "real" motions") pass through. Vibrations that you're trying to ignore are typically higher frequency signals.

Also, if you know the motor's operating frequency, you can instead create a band-reject filter that filters out the noise from the motor Only.

Lastly, if you're capturing the data it's much easier to filter out the motor vibrations post data-collection (i.e. in a spreadsheet or other software). No new hardware needed and you don't run the risk of accidentally filtering out important information.