Degree of tilt - tilt switch which has a binary ability

[kyleg]

degree of tilt

Hello,

I am working on a project and need a component to tell the degree of tilt on the device. I have seen a simple tilt switch which has a binary tilted or not tilted ability, but I would like a little more detail to know how much tilt is occurring.

Can anyone point me in the right direction on this?

thanks

Kyle G

 

[KMoffett]

Degree of tilt

Google: inclinometer

ken

 

[pabloec]

Degree of tilt

use an accelerometer.

 

[Externet]

Re: Degree of tilt

To know the amount of tilt you need two locations: A fixed anchoring for a sensor, and a tilting item
Installing a plain linear potentiometer between them will provide a voltage proportional to the tilt.
If you want two dimensions, use two potentiometers, as a 'joystick' control.
The wiper terminal will sweep a portion of the resistance that can be calibrated to show the voltages. The voltage outputted can be read by an analog to digital converter, or a galvanometer, or other means.
Miguel

 

[arthur0]

Re: Degree of tilt

Hi kyleg,

I agree with pabloec: use an accelerometer.
There are many nowadays, in all sizes, price ranges and types possible so you’ll have no problem finding one that would do the job in your case.

I personally like these https://www.kionix.com/ (Freescale’s are nice too) but as I said, there are many others.

When selecting one you have to think about whether you need a digital or analog output, one or more axes, range, accuracy/noise level, whether temperature drift is an issue (and whether you need to calibrate and compensate for it) etc.
You might have more specific questions once you’ve decided which way to go.

Arthur

 

[kyleg]

Re: Degree of tilt

Thanks for all the help!, I'll have a look into these and see how it goes.

~ Kyle G

 

[doraemon]

Re: Degree of tilt

Hello!

The accelerometer may indeed be one solution if your device does not move.
If the device is in movement, the accelerometer will very likely give you
wrong information.

Dora.

 

[kyleg]

Re: Degree of tilt

The device will be moving (in a small plane), i will have software for compensation of any jitters and noise though, would an accelerometer still be ok?

 

[doraemon]

Re: Degree of tilt

Hello!

No, in this case, you cannot use an accelerometer for the tilt. Too bad.
You may consider buying a gyro. But it's more expensive, and you will
also have to think how to correct its drift.

Dora.

The device will be moving (in a small plane), i will have software for compensation of any jitters and noise though, would an accelerometer still be ok?

 

[arthur0]

Re: Degree of tilt

Yeah, Dora has a point!
Although, for the sake of completeness it must be added that movement does not interfere with the workings of an accelerometer as long as it’s not accelerated, which might not be the case of a small (model?) plane, no.
The thing is that even a plain tilt switch would be influenced by accelerated movement so if you could use that I assumed you could use an accelerometer as well.
It depends on your application really. If, for instance, you’re the one controlling the plane, you could think of an algorithm that would ignore the readings from the accelerometer while the plane is receiving a command that would change its direction and/or speed.
You could even infer the amount of acceleration that a certain command will cause and account for it.
Or you could allow for a substantial damping/filtering that will eliminate the errors introduced by brief accelerations other than caused by gravity.

So you see, everything is not in black and white. Look at all your options before throwing this idea and cosider using a gyro (with or without accelerometers), which will greatly complicate your design.

Arthur