Circuitry to account for continuous potentiometer "deadzone"

Hello!

I am trying to incorporate a RM Young 05103 mechanical anemometer into my project. I am only interested in logging the wind direction (the propeller will be replaced with another sensor array). From what I gather from the datasheet, the wind direction is measured by a 360 degree continuous potentiometer. However, the potentiometer has a deadzone mapped to 355-0/360 degrees. Additionally, the datasheet mentions that "signal conditioning devices must 'clamp' the signal to excitation or reference level when [the deadzone case] occurs... Avoid a short circuit between the wind direction signal line and either the excitation or reference lines... damage to the potentiometer may occur if a short circuit condition exists."

I want to try to capture this deadzone both to protect the potentiometer and reduce erroneous data logging. I have come across this Stack Exchange post that suggests the use of an OpAmp. Without replacing the 360 degree continuous potentiometer, I would like to be able to be able to log data from this sensor. I'm looking for any suggestions?

Many thanks!

Contrary to what you write, there are BOTH dead and short circuit conditions at the end of the resistance material. If you have one end connected to 5 volts and the other end to ground, when the slider hits one connection, you will get 5 volts. When it hits the gap, you will have an open circuit and when it hits the beginning of the other end of the material, you will have ground.

Been there and done that.

Paul

Hi Paul,

Thank you for the comment.

You describe three ‘main’ states for this potentiometer (GND, VIN, OPEN). I can tolerate each of those by simply mapping my outputs to the 10-bit ADC.

However, from what I understand in the manual, I believe that the three ‘main’ states are actually (GND, VIN, SHORT). Ultimately, this is the case that I want to avoid. A case that seems strange from a usability standpoint, so I may be misinterpreting that part of the manual.

It seems I will need to build a diode clamping circuit to deal with the 5 degrees of floating values. Otherwise, I may try to code in some hysteresis to filter out any floating values that result in large spikes.

Hopefully one of these approaches will work and I am very open to others.

Hello!

I am trying to incorporate a RM Young 05103 mechanical anemometer into my project. I am only interested in logging the wind direction (the propeller will be replaced with another sensor array). From what I gather from the datasheet, the wind direction is measured by a 360 degree continuous potentiometer. However, the potentiometer has a deadzone mapped to 355-0/360 degrees. Additionally, the datasheet mentions that "signal conditioning devices must 'clamp' the signal to excitation or reference level when [the deadzone case] occurs... Avoid a short circuit between the wind direction signal line and either the excitation or reference lines... damage to the potentiometer may occur if a short circuit condition exists."

I want to try to capture this deadzone both to protect the potentiometer and reduce erroneous data logging. I have come across this Stack Exchange post that suggests the use of an OpAmp.
I also understand that a diode clamping circuit may be able to mitigate the floating values. Finally, and less desirably, I can introduce hysteresis into the Arduino code to 'filter' out any rapidly changing data, but this could potentially result in loss of desired data.

Without replacing the 360 degree continuous potentiometer, I would like to be able to be able to log data from this sensor. I'm looking for any suggestions?

Many thanks!

Your potentiometer has a track that is connected to Vcc at one end and 0v at the other.
Between the two there is a small gap (dead band) during which the slider is open circuit.

While you COULD detect the open circuit condition it would not tell you the position of the slider within that band.

I'd suggest you connect a high value resistor to ground like this

pot.png

a reading of zero tells you the sensor is in the dead band.

The input capacitance (around 14pF) on the analog input will give a time constant of 14p * 1M = 14usec
and the voltage will change to within 99% of its final value in 5 time constants = 70usec. so should not affect readings.

The parallel resistance will change the readings by a VERY small amount - but you could even go to 10M if you wish.

you shuld use Vcc and 0V from the arduino, and "DEFAULT" (ie Vcc) for your reference.

You will almost certainly need some filtering - have a look at my post here

pot.png

@ntr4ef

TOPIC MERGED.

Could you take a few moments to Learn How To Use The Forum.
Other general help and troubleshooting advice can be found here.
It will help you get the best out of the forum.