View Full Version : Need Help With Probes For DIY Controller

10/16/2007, 03:51 AM
I am getting ready to work on a DIY controller project for a senior design project. I will have the controller measure pH, temp, ORP and Conductivity. I understand these probes need a high impedence amplifier but the part that is really stumping me is what the set outputs are. I am trying to find like a data sheet for these probes that tell me how the output voltage varies with the conditions. Can anyone provide me with some information that can help me?

10/16/2007, 07:50 AM

10/17/2007, 04:32 AM
The link helps with probes but does not seem to show specs on how the voltage changes with the water. For the people who have built controllers, what did you do?

10/18/2007, 01:32 AM

Check the following page: http://www.emesystems.com//BS2index.htm#webRing

This page addresses how to interface various probes to the Basic Stamp controller specifically, but it is a good introduction to the theory behind how they work in general.

To summarize:

pH and ORP - Both of these will output a voltage proportional to the reading. I've personally not interfaced ORP, but from what I've read it outputs the actual reading. That is if the ORP is 400mV, then that is what the output from the probe will be. For the pH probe, it outputs 0mV @ pH 7 and about 59mV per pH unit above or below pH 7. The output is positive below pH 7 and negative above pH 7. So the output of the probe at pH 10 would be -177mV. For both of these, that is the ideal output and changes a little based on the individual probe and over the life of the probe, thus you need a calibration routine to adjust for the differences.

Temperature - On temperature, I've only worked with ones that output a voltage of 10mV per degree C. There is a similar model that outputs 10mV per degree F. There are also models that output a current of 1microAmp per degree Kelvin which can be converted to a voltage. The current version is nice for very long runs of wire or for electrically noisy environments since they are more immune to noise. There are also some temperature sensors that are little computer chips that communicate serially with the controller to report temperature and take care of all the conversion internally.

Conductivity - This one really depends on what type of sensor you are using. The one I built uses a conductivity probe from an old Octopus controller. That probe consists of a PVC body that has two graphite electrodes embedded in it. I used the interface on the page that I linked. Basically the probe is set up as the resistance element of 555 timer chip set up as a free running oscillator. The more conductive the solution, the faster the oscillation of the chip. To convert to a conductivity value you can either measure the length of each oscillation or count the number of oscillations in a given time interval. You do that at two known values (usually 0 {probe dry} and whatever your calibration solution is. From that, you derive values to plug into a formula (y=mx+b) that your controller uses to convert the oscillations to a conductivity measurement. I did a fairly in depth explanation of how to do this conversion in another thread. Try this link


If that doesn't work, search "diy TDS meter" in the DIY forum.

Hope this helps.