Hi all, I know I'm new here, but thought I might be able to help with your Ph predicament. I've been working on a computer for the last three weeks or so that I'm turning into an electronics workstation, currently I've got an oscilloscope (Dave Jones' DSO Mk3) a DMM (PCMCIA card Ines i218) a frequency counter, a function generator and, the reason I'm posting this here, a home built Ph meter.
The meter is actually designed to plug into the DMM and provide a voltage reading (0.01v = 1pH @ 24 deg C) with about a 0.1 +/- pH accuracy. The whole setup cost me about $70 US, but fifty of that was the probe itself. The other twenty was wrapped up in casing the main circuit, buying more etchant, buying the PCB materials, and actual component cost. All in all a cheap beastie considering that pH add-on probes for Fluke DMM's start at one hundred US and go up from there. But I digress.
I'm using a Cole-Parmer U-59001-65 General Purpose pH electrode. Mainly because it's no maintenance, has a top reading of 80 deg C, is designed for permanent/semi-permanent submersion, and looks pretty cool
I haven't bought a digital camera yet, so if you want to see the probe, go to
Cole-Parmer's Website
The circuit itself is actually a very simple design.
This was built with exacting measurements in mind, so the calibration is a bit long, but I'll put it here in case anyone is interested. Anyway. As far a design, it's basically an electronic representation of a mathematic equation. At 24 deg C, the equation for the slope is Y=-0.017X+7 so, to obtain the output in tenths of volts (just for a cleaner reading from the A/D DMM) the equation would be 10Y=-0.17X+70
The first chip (LM351) turns the probe output into a high impedance signal, the second chip (First half of LM353) controls the slope by taking the LM351 output and multiplies it by -0.17, and the third chip (second half of the LM353) while acting as a summing amplifier, controls the Y-intercept by adding 70mV to the signal.
Like I said before, the calibration is long, but makes sense as long as you understand the math. Start by shorting the inputs together, and adjust the offest on the LM351 to read 0mV at its output. Now, unshort the inputs, connect the probe, and place your probe in a pH standard, and measure the voltage at the output of the LM351. Multiply this voltage by 0.17, and now measure the voltage at the output of the first half of the LM353. Adjust the slope pot until you get the voltage (0.17 * X) INVERTED. i.e. if your reading at the first chip was Z adjust until you get -Z at the second chip. Finally, check the final output and adjust the Y-intercept pot until your voltage reading and the pH of the solution are matched. Remember, the way this circuit works (1pH = 0.01v) so if your standard is 10.1 pH your voltage reading will be 0.101v
Finally, check the meter by using it in other liquids that you know the pH of, just to verify proper calibration.
A few notes on the components used. The key on this circuit is accuracy. That's why the slope and y-intercept pots are both 15 turn. Yes, you can use a normal 3/4 or 1 turn, but I would recommend against it. Even with the x10 boost in voltage, the amounts we're dealing with are still so tiny, that a slight bump on the board might change the value of the pots enough to mess up your reading. On the chips, it's what I had on hand. If you can conceive of a better opamp to use, by all means do. The 351 and 353 are not application specific, so I'm sure there is a more 'refined' chip out there for this circuit.
As an afterthought, I have found digital pH meters around the net for about $60 US. The reasons I spent more than that are that one, I like building this stuff, and two, with a homebuilt design like this, you can build a second circuit to read the voltage, make it out of any kind of readout you want (LED, LCD, LPT input, etc) and get it that much more custom/closer to the way you like displays to be. Anyway. I know this was a bit long of a post, but I hope it helps.
The PÜNiSH3R