Hello, I am working on a project to measure the resistivity of salt water. As you probably know, just putting the leads of a multimeter into the water won't work because when the leads hit, they put a charge into the water. I was thinking that if I used a short DC pulse, something like 10ms, then I wouldn't affect the measurement as much. I'm trying to figure out the cheapest way to do this. The requirements are to:

1. generate a short DC pulse when a button is pressed
2. read, convert, and scale this to a resistance measurement
3. store the measurement for later viewing
4. reset the stored measurement to 0.

The measurement can be read by a multimeter, but I need it to be stored long enough to read it.

Can this be done for under $10 each assuming want to build 50 of these devices? I am not against buying a small microcontroller and programming it before it goes into the field.

 

As one who worked on these devices, I can tell you that the accepted method is to use a bipolar square wave between 2 carbon electrodes of known surface area and separation distance and determine the conductivity per cubic centimeter or cubic inch, or some such cubic area. Of course, you are free to set your own standards, depending on what your goals are for accuracy and standardization.

As for ten dollars? You can't even buy a 10 cent transistor without paying $10 in shipping charges! You have a lot to do before you place your order.

 

...I am working on a project to measure the resistivity of salt water.

Just wondering...why? I mean, you could measure specific gravity, refractive index, sodium content, freezing point depression, maybe a few other things, all of which would correlate extremely well with conductivity as long as the composition of the electrolyte is reasonably consistent.

 

Just guessing...conductivity can be measured almost instantly using nothing but electricity and electronic components.

 

Just guessing...conductivity can be measured almost instantly using nothing but electricity and electronic components.

Bingo! I want to design an electronics experiment for high school students. It seems like a $1.50 pre-programmed pic chip would be reasonable for this kit. I think it would do everything I want to do.

1. generate a short DC pulse when a button is pressed - easy enough, just use an NPN transistor to switch a 3V or 9V supply.

2. read, convert, and scale this to a resistance measurement - easy in software, what kinds of circuit would i need to build to read voltage and current so i can determine resistance?

3. store the measurement for later viewing - either use an LCD display, LED bar graph module, LED binary array readout, or send a PWM signal to a blinking light based on different resistance values

4. reset the stored measurement to 0. - simple pushbutton on digital input

 

You cannot use DC current to measure conductivity because of polarization.
As #12 says, you have to use AC or bipolar signal.

Also you will quickly get corrosion on the electrodes which will affect the reading.
You need electrodes made of carbon or platinum.

 

Here's a block diagram of a conductivity meter. I've seen this done with(2) 9V batteries.

 

You cannot use DC current to measure conductivity because of polarization.
As #12 says, you have to use AC or bipolar signal.

Also you will quickly get corrosion on the electrodes which will affect the reading.
You need electrodes made of carbon or platinum.

Assuming I generate a bipolar signal, I guess i'd just signal two transistors, one after the other, that are connected to battery with opposite polarities.

How would I read such a signal? I understand what it would look like when it is transmitted, but not the recieving circuit or the processing.

I have carbon electrodes I can use.

 

Here's a block diagram of a conductivity meter. I've seen this done with(2) 9V batteries.

Thanks! i'll try to draw up a circuit and have you guys look at it. Now that i see the receiving signal is rectified, this makes more sense.