DIY DIY EC Meter

[donkeykong]

Here I'll blog my efforts to design and build an EC meter. For my next grow project, water culture, Im going all science fair and planning something special, part of that will be continuous monitoring of vital statistics. The first thing I decided to tackle was EC.
I need something that can output and log readings rather than just on a screen.
Also I want to be able to monitor several channels continuously. Getting some good quality equipment three or 4 times over is going to cost an arm and a leg, I see decent EC and pH meters cost a lot.

The aim for me is to arrive at a good multi-channel system for a low cost. There are less than a handful of designs on the net on how to do it, and that's where I started.

As usual I will no doubt end up spending more on the exercise than it would have cost to buy a meter, but its not the point. If I can arrive at something others can use or replicate what I've done that would be good. Plus it's my kinda fun doing things like this.

For a start I want to test the theory we can put something together that gives results on par with expensive kit but at a modest price. I did cost up the sensors and systems I was going to need ready built btw, it comes to thousands by the end of it.

A rough outline of what I have planned... multi channel EC, pH, DO, CO2, air and rootzone temp, control of cooling, heating, venting, pumping, lighting etc etc etc.

Not quite sure where this will go in the end but if folks are interested then some info will be here and if enough were interested I might even see if it works to crowdsource the thing and make it happen. For the moment anyway Im putting some background hours into solving the smaller pieces of the puzzle until I have a prospect for the whole system.

 

[donkeykong]

First thing to be done was understand exactly what is involved in a conductivity meter and how its typically measured. No need to reinvent the wheel if not necc, so I also scanned for reference circuits.

For this a few links that I trawled helped a lot ;

EC/TDS/PPM Meter On Limited Budget : good explanation of what it is we are measuring and how it needs to be measured. There is also a circuit diagram provided for the kit Isabella has setup and used successfully. There is a nice model there on something cool called molecular workbench that lets you actually see the simulation of what happens applying AC or DC to the water.

Conductivity Theory and Practice, by Radiometer Analytical : Great doc explaining the basics as well as the different types of probes or conductivity cells available, their advantages and disadvantages, temperature corrections and more.

EC-Sensor circuit on CircuitLab

EC Shield for Arduino : Nice and low component count with this method using a simple 555 timer. I stopped to consider whether a bunch of these would not suit my purpose. I simulated this design and also have put it together to test against the other methods for accuracy and stability.


The thing to learn here was that the probe is quite an important part. Its a fair part of the cost of the eqpt. It needs to fare well under water for long periods with out needing calibration. On the market there are various types from platinum to graphite to carbon electrodes. Platinum probes are very expensive, >£100 a pop. The others can give good results and we only need it for a limited useful range , 0-5000uS lets say.

Im playing with some DIY jobbies but also I have found after a lot of searching some reasonably low cost industrial options.

the first diy probe was not so well thought out, but has served just fine to test the circuits on the breadboard until something better arrives.



Right now there are a couple options Im going to check out .

1) See if I can hack this soil moisture probe into a conductivity probe. Its immersion gold, so should fare better than the regular stuff for this application. This is the v.cheap option.
2) An industrial probe I found on AliExpress , about $28. This one is a carbon probe.
3) I have a probe on order from RS components, £30. It belongs to a £175 piece of kit, and £30 is good going from what I've seen. this one is a carbon probe. At that price I don't know if its worth messing about with option 2, but its already on order so will be interesting to compare.

The Chinese one I don't expect to arrive for some weeks yet. The RS carbon probe will be here by tomorrow , and the DFRobot moisture sensor I have in front of me.
Bunch of stuff turned up today beakers, measuring cylinders and sealed containers etc. calibrations solutions should be here soon also and then Im getting in good shape for the task at hand.
Probably I will have a fun weekend

So far there is an Arduino micro board controlling the breadboard on the left where the test EC circuits are hosted, and driving the LCD display. Oh I forgot the touchscreen should be here soon too 9 bucks Ebay SainSmart 3.2" with touch controller.
using my crappy probe (not anywhere near enough surface area for a start) I was able to calibrate against my £15 stick EC meter and read a third sample within 15uS of the pen. not bad, but that will only work for now for these simple tests and not be repeatable.

 

[donkeykong]

I've started modifying the soil moisture probe from DFRobot , to something which might work as a conductivity probe.


There is a very simple circuit on board and a connector, both of which are no use to us. So first thing is I severed the 2 small traces that lead from the conductor legs to the circuit , isolating them. This took a scalpel to do. Next I removed the connector as we wont need it and it might get in the way later. With that done I prepared a small area on the top edges of the conductors by rubbing with a pcb buffing block (don't do ay more than needed!) and soldered two cores of a shielded cable to it ;
Next I need to get this lot into a round thing like a probe, and get it watertight , past where I have soldered.

Plugged it into the circuit and it seems to be working OK as such. Reading are all over the place but that is to be expected, the huge conductor area compared to the first attempt, is very different. I think I will need to reduce the area actually.
I plan to pot this inside a tube and leave exposed a smaller part of both legs, and waterproofing the rest up to the cable.

 

[lowbudge]

rep and following!GoooAuto

 

[donkeykong]

Since the last post I did a bit more work on the probe, I needed to mount this within something, so that it has some fixed "walls" around it. Without that 2-pole probe is going to be affected by its location in the measuring vessel. Also I needed to coat everything making it waterproof underwater, exposing only a small section of the end of the probe legs.

To do it I found some waste water pipe in the garage that was perfect width, the probe is a little on the wide side so its not going to be dainty like my first effort. First I cut a small section and used some thermoplastic to mould the wire through the middle and push it into the circumference of the pipe ;
once that had cooled I was able to pop it out and I've made me a "stopper" of sorts :

After this I prepared a longer piece of the pipe , and cut two slots in the side. This will allow bubbles to escape, which would spoil readings, and also solution to easily enter and leave. Here is it coming together ;


Now it was tricky but I pushed this lot into the pipe and then glued the stopper in place. I added a bit more thermoplastic to the stopper to hold the probe level, with in the pipe. Having let that set last night , earlier today I poured in some potting compound, its ip68 rated and very good against water. it will stop any ingress and so the area of the conductors exposed in the water will be constant.
Here is after I poured in the potting compound , it will take 24-36 hours to cure.


The carbon probe didn't arrive, buggers delivered it to someone else somewhere else so now I wont have it till Monday
Still, there is other stuff I can get on with for now.

Be interesting to see how this probe fares, hopefully Ive given it a fighting chance!

 

[donkeykong]

Other task of the day was get some proper solutions together to be able to test and evaluate better. So far I just used my EC pen to match readings with and calibrate with, not so good.
I used the method mentioned here to make up a range of solutions , taking care to wash everything out with distilled water before use and keep it as accurate as I can.
for this the bits and pieces I got from Rapid came in very handy. some widemouth sealing flasks, measuring cylinder, funnels... rinsed out and filled up a sprayer to make it easier to clean things...

so now I think im good on this count, plus I have two reference solutions to work with also so I'll know im not just on my own planet somewhere

 

[donkeykong]

 

[Nam80232]

Good start. I'm interested in seeing how it ends up!

 

[realcarlos]

Smart man!

 

[gradolabs]

dude I'll have to catch up on this thread. looks sick man.