John:

Sorry if I did not give enough details on what I used.

The ohmmeter and saltwater is used just a quick test of operation. In
actual operation I said the electrolyte must be compatible with the wire
with a sensing current as low as possible. Even with AC excitation
contamination will occur if the current is too high (one reason AC
driven LCDs fade with time). The current is kept very low so that
Brownian movement gives a more constant reading with even DC. I used a
CMOS Op Amp with less than 10E-12A input currents and low duty cycle
pulsed mode of operation. The wire was 36 GA nichrome. The electrolyte
was ethylene glycol  with 2 mg NaCl and 50 mg Borax per 1000 cc.

- Wayne Foletta
BMI - Saratoga, CA

> ----------
> From:         johns@avenuetech.com[SMTP:johns@avenuetech.com] on
> behalf of John Shreffler[SMTP:johns@AVENUETECH.COM]
> Reply To:     pic microcontroller discussion list
> Sent:         Thursday, January 22, 1998 8:54 AM
> To:   'pic microcontroller discussion list'
> Subject:      RE: inclinometer
>
> Measuring resistance of an electrolyte such as salt water
> has been a thread some time back.  Any DC biased
> measurement system such as a ohm meter will build up
> a cloud of ions around it that will render the measurement
> useless in a few seconds.  Just put the leads in a glass
> of salt water and watch the reading.  Now stir and see
> what happens.
>
> Pardon the trailer junk.  I am a Microsoft victim
>
>       -----Original Message-----
>       From:   Wayne Foletta [SMTP:wayne@ELECTROTEK.COM]
>       Sent:   Thursday, January 22, 1998 11:40 AM
>       To:     PICLIST@MITVMA.MIT.EDU
>       Subject:        Re: inclinometer
>
>       Michael:
>
>       I needed a inclinometer sensor also. After research I found
> these
>       sensors are fairly expensive ($50 bucks and up - not a
> mainstream market
>       part).
>       You may want to build one yourself if you have the components in
> your
>       lab. The one I built is a variable resistance three terminal
> device.
>       Find or form a curved tube of glass or plastic about 1" - 2"
> long. The
>       curve radius depends on the angles you want to measure. Lay bare
> wires
>       straight in parallel along the tube length and attach to tube
> ends. Seal
>       one end of tube and fill 50% with an electrolyte (use tap water
> with few
>       grains of table salt). Seal tube and connect sensor wires to an
> ohmmeter
>       to test the sensor operation. For long term use the electrolyte
> must be
>       compatible with the sensor wires. Also the sensing current
> should be as
>       low as possible. Three or fours wires could be used for higher
> accuracy
>       in ratiometric mode.
>
>       See this site for a capacitive version:
>       http://www.euronet.nl/users/ragman/robot1.html
>
>       - Wayne Foletta
>               BMI - Saratoga, CA
>