High resolution threshold settings through POTs

[gary36]

I have a requirement for setting thresholds with the resolution of 1mV in the range of 0-10V. Is it possible using trim pots?

 

[KlausST]

Hi,

1:10000 dynamics is huge.

Maybe with a high quality geard pot.

0V really means 0mV? Mind: some pots can't go down to 0.00% nor to 100,00%.
Also there will be some "wiper impedance" which may cause additional errors.
And pots may drift...and drift by thermocouple effects..

There theoretically is an infinite resolution, but the useful resolution my be a mechanical problem, how much time you want to spend for adjusting it and how long you want it to keep the value within a given tolerance.

I guess you need to read datasheets, read application notes from pot manufacturers and/or contact pot manufacturers.

Klaus

 

[gary36]

Is it possible by using coarse/fine arrangement?

 

[KlausST]

Hi,

I did not say it´s impossible...

So yes, it will be possible with coarse/fine adjustment.
But remaning problems are: drift, wiper resistance(s), thermocouple effects.

I don´t know what it is for. And what true requirements there are....
Do you really care if there are 4999mV or 5001mV?

But why no digital control?

Klaus

 

[gary36]

Customer wanted a system with no micro controller or FPGA. Could you suggest method without considering drift specs

 

[KlausST]

Hi,

there are digital pots, maybe in carse/fine combination. Some are non volatile, some have an UP/DOWN interface.

Depending on requirement even a digital PWM (even without uC and without FPGA possible) and a analog SPDT switch followed by a 2nd order low pass filter could be useful.
The benefit is the very equal step size. (good DNL)

Or a rotary switch for the range and a pot.

Maybe a multiplying ADC DAC and a push button controlled UP/down counter..

As already said: It depends on the true requirements.

Klaus

 

[danadakk]

Linearity to +/- .005%, resolution 12 bits, not quite enough.....

ETA25PM 6x22 2405 F0.15B

Hall effect (wear-free) encoder. Field programmable up to 200 turns. Analog output. 25mm housing diameter. IP66. Order from P3 America today.

p3america.com

But if you need absolute accuracy probably need to start at 16 bits or better
before you analyze total error.

16 bit DAC with toggle switches to set binary value (UGH, thats how you
can avoid processor)......


Regards, Dana.

 

[gary36]

Threshold settings should be truly non volatile in nature. Almost all digital pots (non volatile) come up with SPI/I2C interface. It seems like it is impossible.

 

[danadakk]

Threshold settings should be truly non volatile in nature. Almost all digital pots (non volatile) come up with SPI/I2C interface. It seems like it is impossible.

16 bit DAC + binary coded rotary switches (4) to input hex value equivalent
to desired voltage....

Coded Rotary Switches Binary Code Hex 16 Position: Amazon.com: Tools & Home Improvement

Coded Rotary Switches Binary Code Hex 16 Position: Amazon.com: Tools & Home Improvement

www.amazon.com

Or use CPLD and Verilog to do 5 digit BCD to binary (using rotary BCD switches) -

Binary to BCD

my.eng.utah.edu

Or use (god forbid a micro) ATTINY85 to do BCD switch or keyboard scan and generate binary needed for DAC, switches and DAC I2C....

Regards, Dana.

 

[mtwieg]

"Analog" trimpots have infinite precision, but limited accuracy. You haven't given any requirement for accuracy though.

Customer wanted a system with no micro controller or FPGA. Could you suggest method without considering drift specs

How does the customer intend to interface to this thing then? Sounds like there's some info missing....

 

[crutschow]

Certainly you could use two trimpots to get a (theoretical) 1mV resolution.
Here's a article that discusses various configurations for that.

 

[KlausST]

"Analog" trimpots have infinite precision, but limited accuracy.

Hi,

It has infinite resolution, but not precision.
"Precision" means repeatability...and errors like noise, drift with time and drift with temperature. In case of a pot even mechanically caused errors.

Klaus

 

[mtwieg]

Hi,

It has infinite resolution, but not precision.
"Precision" means repeatability...and errors like noise, drift with time and drift with temperature. In case of a pot even mechanically caused errors.

Klaus

Resolution is probably the better term here, agreed. Precision of a trimpot would depend on how you measure it (is the wiper moving between measurements?).

In any case, the OP's requirements (accuracy, resolution, precision, etc) are ill-defined.