# using sine or square wave to obtain conductivity of solution?

I see for designing a conductivity meter, Some Designer use square waves, as excitation to obtain conductivity and some others use sine waves for this work in their design, which wave do this job better for me?

• This question is posed as a question of opinion. Please make it specific to an application and ask a specific question. This is really dependent on the circuit in use – Voltage Spike Jul 16 '18 at 20:06
• What impedance do you expect? <1 or >1MOhm? what is conductance G [S] vs capacitive impedance if it is an insulator Z(f)[Ohms] for some unknown frequency with odd harmonics . Normally Constant current Sine is used. – Tony Stewart Sunnyskyguy EE75 Jul 16 '18 at 20:33
• If solution is water based then Dk = 80 so capacitance is lower Z than R – Tony Stewart Sunnyskyguy EE75 Jul 16 '18 at 20:47

I am in the process of designing an EC (Electrical conductivity) meter that works with the Raspberry PI and I am using square waves. It is much easier to generate a square wave than a sine wave. In my case, the RPi actually controls the timing of the square waves.

Typically, an an op amp is used to drive an AC signal to the EC probe. In this case, the edge rate is limited by slew rate of the op amp. Unless you are using a VERY fast op amp, you will not need to worry about issues from fast-edged square waves.

Simple EC meter designs rectify and low pass the AC signal output of the EC probe and measure the resulting voltage, but this is not extremely accurate. A better way is to use an A/D converter to measure the actual voltage output from the EC probe.

The design which I am currently debugging samples the EC sensor output after the rising edge of the square wave. It averages the actual voltage with a sample-and-hold stage driving an RC low pass filter to accurately measure the voltage with a low frequency A/D converter.

I have found that the main issue with accurately measuring the conductivity is that for high conductivity solutions, the ions in solution migrate very quickly, so you need to measure the effective solution resistance very close to the edge the square wave.

• Thank's for your answer my friend, but I am looking for an answer about which wave Do better effect in my conductivity measurement result, not about which wave can produce easier!!! and I am sorry for my bad English. – Rashid Jul 17 '18 at 7:13
• @Rashid Check out this app note from Analog Devices. It is for a high end EC meter. It uses square waves. They are both easier and better. analog.com/en/design-center/reference-designs/… – crj11 Jul 17 '18 at 11:54
• Yes, i checked rosemont emerson conductivity transmitter too, and also they used square wave , but i am looking for the reason – Rashid Jul 17 '18 at 11:57
• @Rashid My assumption is that since measuring high conductivity solutions requires measuring quickly after excitation, the slower rise time of a sine wave complicates the measurement. In the "olden days" before digital systems, I think sine waves were used because they were best for the associated rectification/low pass measurement techniques. – crj11 Jul 17 '18 at 12:01

I think that most use sine waves for simplicity. It depends on if a long cable between the electronics and the sensor will be involved. Some people use a square wave then. The cable capacitance distorts the square wave but the detector (rectifier) only looks at the second half of the return signal.

• Also, higher harmonics in a square wave could give false readings, appearing to be more conductive, due to capacitance between electrodes. – DrMoishe Pippik Jul 16 '18 at 19:37