Help with Conductivity Cell Design Using Analog Devices Application Note

[digitalectron]

Hello everyone,

I am working on a project to measure the electrical conductivity of water using a 4-pole conductivity cell. I found a helpful application note from Analog Devices that explains the design process, but I am facing some challenges.

PWM Activation Timing: The application note mentions using PWM signals, but it doesn't clearly specify when and why these signals should be activated. Could someone clarify how to determine the correct timing and frequency of the PWM signals for driving a 4-pole conductivity cell?

Current Measurement: The design involves measuring the current through the cell. Why is current measurement critical for determining conductivity, especially in a 4-pole configuration?

Additional Considerations for 4-Pole Cells: Are there specific nuances or best practices for working with 4-pole conductivity cells that differ from 2-pole designs?

Any guidance, especially if you’ve worked on similar projects or with Analog Devices' application notes, would be greatly appreciated. I aim to design a reliable circuit based on the principles described in the note but need more clarity on these aspects.

Thank you in advance!
https://www.analog.com/media/en/reference-design-documentation/reference-designs/cn0359.pdf

[digitalectron]

I think the main reason for measuring the current of a conductivity cell is that conductance, and consequently conductivity, of a liquid is directly related to the current passing through the cell. Since the microcontroller cannot measure current directly, it must be converted to voltage for the system.

[sw_guy]

Typically a 4 electrode sensor is used over 2 electrode sensor due to polarization effect using certain electrode materials. Please note that there are more issues that impact to measurements, but this is often the first one on the white board to be solved.
I also used the Analog's circuit note as a starting point when I started a similar project, but in the end, the solution that went to the production unit became very much different.
You can calculate conductivity from conductance - and voltage+current measurements are for that purpose. The circuit note in question explains pretty well the functional part of the circuit.
For me the Analog's design looks more like a showcase of what kind of components Analog has rather than providing an optimal conductivity measurement solution. And... isn't it kind of the idea of such a circuit notes in the first place.
A hint: you have probably specified what is the range of conductivity measurement. There are several different sensor types that have different characteristics. 4 pole sensor is pretty common though.