[simpits-tech] Next question

[Marv De Beque]

Linear and rotary motion can be best detected with LVDT.  Essentially, it is
an 3-coil AC transformer that has a magnetically permeable slug.  The
downside is the electrical interface is a little complicated, but they are
extremely accurate and the longevity is nearly infinite.

Hall Effect generally looks for a reversal of the magnetic field and
measures the magnitude of the differential.  Interfacing is pretty simple
since the Hall Effect produces a linear voltage output.  Some care must be
taken with the mechanical design to insure a true linear response.

Capacitance works with small changes in distance, but I think it has the
potential to be finicky.

The problem with position sensing pots is the longevity stinks.  You can get
precision pots that have better lifespan.  They are simple to interface.

Marv


On 7/16/04 3:24 PM, "Fred Mahone" <fred.mahone at vt.edu> wrote:

> OK, I have been following the questions and comments about analog and
> digital joysticks and throttles.
> 
> Now the question, Are there any circuits that can take linear and rotary
> motion and convert it to a computer compatible signal,
> be better than potentiometers, and remember the position so the control
> does not have to be re-calibrated each time you start?
> 
> I have looked at various schemes, hall effect, optical sensors, variable
> capacitors and have not seen anything that is as good as a position sensing
> potentiometer like used to send position to an electronic ignition in a
> car. I would guess there are probably ones based on stress to some element
> but have not found any, as yet.
> 
> My Bell cyclic pivot will have two linear movements, the anti-torque will
> be either rotary or linear (I can go either way), collective will be
> rotary, and if I add the throttle, linear.
> 
> Any web site information would be appreciated.
> 
> Fred Mahone
> 
> _______________________________________________
> Simpits-tech mailing list
> Simpits-tech at simpits.org
> http://www.simpits.org/mailman/listinfo/simpits-tech
> To unsubscribe, please see the instructions at the bottom of the above page.
> Thanks!