Close

Three Dollar EC - PPM Meter [Arduino]

A project log for Fly Wars : A Hackers Solution To World Hunger

Using Technology And A Hackers Mindset To Increase Food Security For The Coming Billions Last Updated [21/9/2015]

Michael RatcliffeMichael Ratcliffe 09/04/2015 at 13:56124 Comments

This Blog will Cover How to build a cheap EC meter for your aquaponics/Hydroponics or water quality related projects. We are not going to get into what the ideal value of PPM or EC is, Just cover how to measure and quantify a fluid.

We will be using this for the Urine based aquaponics unit, we need to be able to control the strength of the growing fluid in the system but for the person on a budget a EC meter is just to much money. the Solution a $3 EC Meter for any Arduino.

You can use this to measure drinking water quality to with a small change to the code and changing R1 [see below].

Parts:

-MCU of your choice with ADC

-DS18B20 waterproof temperature sensor

-500 ohm [or 1kohm resistor]

-Type A Two Prong american plug to Figure 8

-Female Socket for Figure 8 connector

So why are we using a plug:

-Cheap

-Available worldwide

-Standard size [makes calibration easy]

Use the solid prong one like below and not the one with holes:

images.jpg

Wiring it up:

Note: You want the Solid Prong type plug

Do not Plug the pronged plug into the mains

Pinout.png

Operating Principal

PPM is calculated from the EC of a fluid, EC is the inverse of the electrical resistance of the fluid. We are estimating the EC or PPM of a fluid by measuring the resistance between two probes [The plug pins] when the plug is submerged in the liquid of interest.

Ec measurement needs to be done using AC or the liquid of interest is polarised and will give bad readings. This has got to be a great example of asking why instead of just accepting a statement as fact, it turns out we can take a very fast DC reading without suffering polarisation. meaning we can make a really cheap EC sensor.

Want to use it and dont care how it works? Skip to the main EC code and using the wiring diagram it will work.

Temperature Compensation

Temperature has an effect on the conductivity of fluids so it is essential that we compensate for this.

It is common to use a liner approximation for small temperature changes[1] to convert them to their equivelant EC at 25*C:

EC25 = EC /( 1 + a (T - 25) )

EC25- Equivelant EC at 25'C

EC - Measured EC

T- Temperature [Decgrees C] of Measurment

a = 0.019 °C [Commonly used for nutrient solutions]

Deciding on Value of R1

//##################################################################################

//----------- Do not Replace R1 with a resistor lower than 300 ohms ------------

//##################################################################################

We can change the Value of R1 in the voltage divider to change the range of EC we want to measure. Below is the Equivalent Voltage divider circuit.

Voltage Divider.png

Ra

Ra the resistance of the digital pins is not stated in the data sheet instead we need to pull it out from a graph.

Going off the graph on page [387] of the atmel 2560 Data Sheet “Figure 32-25. I/O Pin Output Voltage vs. Source Current (VCC = 5V)”

V=IR

Ra= V/I [From Figure] V=0.4 I=1.5e-4 R=25 ohms estimated

Rc


Rc will change with EC [PPM] of the measured fluid. we will calculate the maximum and minimum values we expect to see for the range of fluids we wish to measure taking into account temperature changes and the cell constant K. [We will estimate K to be 3 for the plug probe, estimate from previous tests]

EC = EC25*( 1 + a (T - 25))

R=(1000/(EC*K)) +Ra

Min temp=0 [we arnt going to care about EC if the pond is frozen]

Max Temp = 40 *C [I doubt a pond should be above this]

Minimum EC 25=0.3 EC= 0.3*(1+0.019*(0-25) Min EC= 0.16 S/sm

Maximum EC 25= 3 EC= 0.3*(1+0.019*(40-25) Max EC = 3.9 S/cm

Min Resistance = 1000/(MaxEC*K)+25 = 1000(3.9*2.88) =114 ohms

Max Resitance = 1000/(MinEC*K)+25 = 1000/(0.16*2.88) = 2195 ohms

R1

Now we have enough information to calculate a good value for R1 to get the best resolution over our intended measuring range. We could sum it all up mathematically and differentiate to find the peak, but that hurts my head so I just did a quick excel spreadsheet for the Voltage divider for the EC I expect to see:

Exel.png


As we can see we get the largest difference using a value for R1 of 500 ohm, I only had 1Kohm to hand so I will have to live with a little less range.

So we chose a 500 ohm resistor

EC – Range /Voltage Range * (5/ADC steps)

(3.9-0.16)/3.14 * 5/1024 = 5.8e-3 resolution so that is a resolution of 0.0058

To put this is PPM [Tranchen [Australia] PPMconversion: 0.7] this is a resolution of 4ppm.

Much more than we need for aquaponics or hydroponics.

If you want to measure the quality of drinking water you will need to calculate the expected Ec values and increase R1 accordingly.


Calibration Code

If you want the best readings from your system it is advisable to calibrate your sensor with some known fluid. But If you dont need to if you use the plug probe shown above, it will still work well.

>Add your EC in S/cm into the definitions

>Plug your K value from the terminal window into the main EC code

you will need to use the modified one wire and Dallas library [download from www.michaelratcliffe.com] or add a pull up for the temperature probe data line [google it]

/*
  ElCheapo Arduino EC-PPM measurments Calibration
 
  This Script is used for calibration of the sensor and fine tuning of the Cell Constant K
  Submerge the sensor and temperature probe in the calibration solution and leave for a while so the temperature probe can settle
  Change the value of the calibration solution to suit the solutiton strength
  Stir the probe to make sure the solution is well mixed and upload the code to the arduino
  Open the terminal for an update of the estimated Cell Constant K [should be around 3] and use this new value in the main EC code.
 
 
  28/8/2015  Michael Ratcliffe  Mike@MichaelRatcliffe.com
 
 
          This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.
 
 
    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.
 
 
    You should have received a copy of the GNU General Public License
    along with this program.  If not, see .
 
    Parts:
    -Arduino - Uno/Mega
    -Standard American two prong plug
    -1 kohm resistor
    -DS18B20 Waterproof Temperature Sensor
 
 
    See www.MichaelRatcliffe.com/Projects for a Pinout and user guide or consult the Zip you got this code from
 
*/
 
 
//************************** Libraries Needed To Compile The Script [See Read me In Download] ***************//
// Both below Library are custom ones [ SEE READ ME In Downloaded Zip If You Dont Know how To install Use them or add a pull up resistor to the temp probe
 
 
#include <OneWire.h>
#include <DallasTemperature.h>
 
 
 
 
 
 
//************************* User Defined Variables ********************************************************//
 
 
float CalibrationEC=1.38; //EC value of Calibration solution is s/cm
 
 
 
 
//##################################################################################
//-----------  Do not Replace R1 with a resistor lower than 300 ohms    ------------
//##################################################################################
 
 
int R1= 1000;
int Ra=25; //Resistance of powering Pins
int ECPin= A0;
int ECGround=A1;
int ECPower =A4;
 
 
//*************Compensating for temperature ************************************//
//The value below will change depending on what chemical solution we are measuring
//0.019 is generaly considered the standard for plant nutrients [google "Temperature compensation EC" for more info
float TemperatureCoef = 0.019; //this changes depending on what chemical we are measuring
 
 
 
 
//************ Temp Probe Related *********************************************//
#define ONE_WIRE_BUS 10          // Data wire For Temp Probe is plugged into pin 10 on the Arduino
const int TempProbePossitive =8;  //Temp Probe power connected to pin 9
const int TempProbeNegative=9;    //Temp Probe Negative connected to pin 8
 
 
 
 
//***************************** END Of Recomended User Inputs *****************************************************************//
 
 
OneWire oneWire(ONE_WIRE_BUS);// Setup a oneWire instance to communicate with any OneWire devices
DallasTemperature sensors(&oneWire);// Pass our oneWire reference to Dallas Temperature.
 
 
float TemperatureFinish=0;
float TemperatureStart=0;
float EC=0;
int ppm =0;
 
 
float raw= 0;
float Vin= 5;
float Vdrop= 0;
float Rc= 0;
float K=0;
 
 
 
 
int i=0;
float buffer=0;
 
 
//*********************************Setup - runs Once and sets pins etc ******************************************************//
void setup()
{
  Serial.begin(9600);
  pinMode(TempProbeNegative , OUTPUT ); //seting ground pin as output for tmp probe
  digitalWrite(TempProbeNegative , LOW );//Seting it to ground so it can sink current
  pinMode(TempProbePossitive , OUTPUT );//ditto but for positive
  digitalWrite(TempProbePossitive , HIGH );
  pinMode(ECPin,INPUT);
  pinMode(ECPower,OUTPUT);//Setting pin for sourcing current
  pinMode(ECGround,OUTPUT);//setting pin for sinking current
  digitalWrite(ECGround,LOW);//We can leave the ground connected permanantly
 
  delay(100);// gives sensor time to settle
  sensors.begin();
  delay(100);
  //** Adding Digital Pin Resistance to [25 ohm] to the static Resistor *********//
  // Consule Read-Me for Why, or just accept it as true
  R1=(R1+Ra);
 
  Serial.println("ElCheapo Arduino EC-PPM measurments Calibration");
  Serial.println("By: Michael Ratcliffe  Mike@MichaelRatcliffe.com");
  Serial.println("Free software: you can redistribute it and/or modify it under GNU ");
  Serial.println("");
  Serial.println("Make sure Probe and Temp Sensor are in Solution and solution is well mixed");
  Serial.println("");
  Serial.println("Starting Calibration: Estimated Time 60 Seconds:");
 
 
 
};
//******************************************* End of Setup **********************************************************************//
 
 
 
 
//************************************* Main Loop - Runs Forever ***************************************************************//
//Moved Heavy Work To subroutines so you can call them from main loop without cluttering the main loop
void loop()
{
 
 
  i=1;
  buffer=0;
sensors.requestTemperatures();// Send the command to get temperatures
TemperatureStart=sensors.getTempCByIndex(0); //Stores Value in Variable
 
 
//************Estimates Resistance of Liquid ****************//
while(i<=10){
 
 
 
digitalWrite(ECPower,HIGH);
raw= analogRead(ECPin);
raw= analogRead(ECPin);// This is not a mistake, First reading will be low
digitalWrite(ECPower,LOW);
buffer=buffer+raw;
i++;
delay(5000);
};
raw=(buffer/10);
 
 
 
 
sensors.requestTemperatures();// Send the command to get temperatures
TemperatureFinish=sensors.getTempCByIndex(0); //Stores Value in Variable
 
 
//*************Compensating For Temperaure********************//
EC =CalibrationEC*(1+(TemperatureCoef*(TemperatureFinish-25.0))) ;
 
//***************** Calculates R relating to Calibration fluid **************************//
Vdrop= (((Vin)*(raw))/1024.0);
Rc=(Vdrop*R1)/(Vin-Vdrop);
Rc=Rc-Ra;
K= 1000/(Rc*EC);
 
 
 
 
Serial.print("Calibration Fluid EC: ");
Serial.print(CalibrationEC);
Serial.print(" S  ");  //add units here
Serial.print("Cell Constant K");
Serial.print(K);
 
 
if (TemperatureStart==TemperatureFinish){
  Serial.println("  Results are Trustworthy");
  Serial.println("  Safe To Use Above Cell Constant in Main EC code");
 
}
else{
  Serial.println("  Error -Wait For Temperature To settle");
 
}
 
 
}
//************************************** End Of Main Loop **********************************************************************//

EC PPM Measurement Code

>If you are using PPM and not EC make sure you note what conversion factor you are using [it isnt universal]

>Dont call the read function more than once every 5 seconds or you will get bad readings and a damaged probe

I tested this code in a solution for 48 hours reading at 5 second intervals without any polarisation or probe damage, the longer you leave between readings the longer your probe will last. 5 seconds is the minimum wait between readings not the maximum.

you will need to use the modified one wire and Dallas library [download from www.michaelratcliffe.com] or add a pull up for the temperature probe data line [google it]

/*
  ElCheapo Arduino EC-PPM measurments
 
  This scrip uses a common USA two prong plug and a 47Kohm Resistor to measure the EC/PPM of a Aquaponics/Hydroponics Sytem.
  You could modift this code to Measure other liquids if you change the resitor and values at the top of the code.
 
  This Program will give you a temperature based feed controller. See Read me in download file for more info.
 
  28/8/2015  Michael Ratcliffe  Mike@MichaelRatcliffe.com
 
 
          This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.
 
 
    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.
 
 
    You should have received a copy of the GNU General Public License
    along with this program.  If not, see .
 
    Parts:
    -Arduino - Uno/Mega
    -Standard American two prong plug
    -1 kohm resistor
    -DS18B20 Waterproof Temperature Sensor
 
    Limitations:
    -
    -
 
    See www.MichaelRatcliffe.com/Projects for a Pinout and user guide or consult the Zip you got this code from
 
*/
 
 
//************************** Libraries Needed To Compile The Script [See Read me In Download] ***************//
// Both below Library are custom ones [ SEE READ ME In Downloaded Zip If You Dont Know how To install] Use them or add a pull up resistor to the temp probe
 
 
#include <OneWire.h>
#include <DallasTemperature.h>
 
 
 
 
 
 
//************************* User Defined Variables ********************************************************//
 
 
//##################################################################################
//-----------  Do not Replace R1 with a resistor lower than 300 ohms    ------------
//##################################################################################
 
 
int R1= 1000;
int Ra=25; //Resistance of powering Pins
int ECPin= A0;
int ECGround=A1;
int ECPower =A4;
 
 
//*********** Converting to ppm [Learn to use EC it is much better**************//
// Hana      [USA]        PPMconverion:  0.5
// Eutech    [EU]          PPMconversion:  0.64
//Tranchen  [Australia]  PPMconversion:  0.7
// Why didnt anyone standardise this?
 
 
float PPMconversion=0.7;
 
 
//*************Compensating for temperature ************************************//
//The value below will change depending on what chemical solution we are measuring
//0.019 is generaly considered the standard for plant nutrients [google "Temperature compensation EC" for more info
float TemperatureCoef = 0.019; //this changes depending on what chemical we are measuring
 
 
 
 
//********************** Cell Constant For Ec Measurements *********************//
//Mine was around 2.9 with plugs being a standard size they should all be around the same
//But If you get bad readings you can use the calibration script and fluid to get a better estimate for K
float K=2.88;
 
 
 
 
//************ Temp Probe Related *********************************************//
#define ONE_WIRE_BUS 10          // Data wire For Temp Probe is plugged into pin 10 on the Arduino
const int TempProbePossitive =8;  //Temp Probe power connected to pin 9
const int TempProbeNegative=9;    //Temp Probe Negative connected to pin 8
 
 
 
 
//***************************** END Of Recomended User Inputs *****************************************************************//
 
 
OneWire oneWire(ONE_WIRE_BUS);// Setup a oneWire instance to communicate with any OneWire devices
DallasTemperature sensors(&oneWire);// Pass our oneWire reference to Dallas Temperature.
 
 
float Temperature=10;
float EC=0;
float EC25 =0;
int ppm =0;
 
 
float raw= 0;
float Vin= 5;
float Vdrop= 0;
float Rc= 0;
float buffer=0;
 
 
 
 
//*********************************Setup - runs Once and sets pins etc ******************************************************//
void setup()
{
  Serial.begin(9600);
  pinMode(TempProbeNegative , OUTPUT ); //seting ground pin as output for tmp probe
  digitalWrite(TempProbeNegative , LOW );//Seting it to ground so it can sink current
  pinMode(TempProbePossitive , OUTPUT );//ditto but for positive
  digitalWrite(TempProbePossitive , HIGH );
  pinMode(ECPin,INPUT);
  pinMode(ECPower,OUTPUT);//Setting pin for sourcing current
  pinMode(ECGround,OUTPUT);//setting pin for sinking current
  digitalWrite(ECGround,LOW);//We can leave the ground connected permanantly
 
  delay(100);// gives sensor time to settle
  sensors.begin();
  delay(100);
  //** Adding Digital Pin Resistance to [25 ohm] to the static Resistor *********//
  // Consule Read-Me for Why, or just accept it as true
  R1=(R1+Ra);// Taking into acount Powering Pin Resitance
 
  Serial.println("ElCheapo Arduino EC-PPM measurments");
  Serial.println("By: Michael Ratcliffe  Mike@MichaelRatcliffe.com");
  Serial.println("Free software: you can redistribute it and/or modify it under GNU ");
  Serial.println("");
  Serial.println("Make sure Probe and Temp Sensor are in Solution and solution is well mixed");
  Serial.println("");
  Serial.println("Measurments at 5's Second intervals [Dont read Ec morre than once every 5 seconds]:");
 
 
};
//******************************************* End of Setup **********************************************************************//
 
 
 
 
//************************************* Main Loop - Runs Forever ***************************************************************//
//Moved Heavy Work To subroutines so you can call them from main loop without cluttering the main loop
void loop()
{
 
 
 
 
GetEC();          //Calls Code to Go into GetEC() Loop [Below Main Loop] dont call this more that 1/5 hhz [once every five seconds] or you will polarise the water
PrintReadings();  // Cals Print routine [below main loop]
 
 
delay(5000);
 
 
}
//************************************** End Of Main Loop **********************************************************************//
 
 
 
 
//************ This Loop Is called From Main Loop************************//
void GetEC(){
 
 
//*********Reading Temperature Of Solution *******************//
sensors.requestTemperatures();// Send the command to get temperatures
Temperature=sensors.getTempCByIndex(0); //Stores Value in Variable
 
 
 
 
//************Estimates Resistance of Liquid ****************//
digitalWrite(ECPower,HIGH);
raw= analogRead(ECPin);
raw= analogRead(ECPin);// This is not a mistake, First reading will be low beause if charged a capacitor
digitalWrite(ECPower,LOW);
 
 
 
 
//***************** Converts to EC **************************//
Vdrop= (Vin*raw)/1024.0;
Rc=(Vdrop*R1)/(Vin-Vdrop);
Rc=Rc-Ra; //acounting for Digital Pin Resitance
EC = 1000/(Rc*K);
 
 
//*************Compensating For Temperaure********************//
EC25  =  EC/ (1+ TemperatureCoef*(Temperature-25.0));
ppm=(EC25)*(PPMconversion*1000);
 
 
;}
//************************** End OF EC Function ***************************//
 
 
 
 
//***This Loop Is called From Main Loop- Prints to serial usefull info ***//
void PrintReadings(){
Serial.print("Rc: ");
Serial.print(Rc);
Serial.print(" EC: ");
Serial.print(EC25);
Serial.print(" Simens  ");
Serial.print(ppm);
Serial.print(" ppm  ");
Serial.print(Temperature);
Serial.println(" *C ");
 
 
/*
//********** Usued for Debugging ************
Serial.print("Vdrop: ");
Serial.println(Vdrop);
Serial.print("Rc: ");
Serial.println(Rc);
Serial.print(EC);
Serial.println("Siemens");
//********** end of Debugging Prints *********
*/
};

Got any questions let me know.

The next tutorial will be on coding a self learning nutrient doser.

References:

[1]

John J. Barron & Colin Ashton "The Effect of Temperature on Conductivity Measurement" Technical Services Department, Reagecon Diagnostics Ltd

http://www.reagecon.com/pdf/technicalpapers/Effect_of_Temperature_TSP-07_Issue3.pdf

Discussions

stedes_dev wrote 7 days ago point

Hi, how is life Michael

What is the reading range in ppm of this design...

Thanks in advance...

Steph

  Are you sure? yes | no

rahul.kumar wrote 07/10/2019 at 10:40 point

EC must be divided by cell constant to get conductance. As resistance is reciprocal of conductance. But you used product of conductivity and cell constant. Can you elaborate this?

  Are you sure? yes | no

rahul.kumar wrote 07/10/2019 at 08:16 point

So i'm looking at the AMTEL to calculate the resistance internal to the pins. How did you determine your values from the chart? there is no 0.4 volts

  Are you sure? yes | no

askhasangms wrote 06/25/2019 at 09:02 point

Dear Michael

I researched about the typical EC values for hydroponics, and I find out the values are normally ranging from 2 mS/cm to 4 mS/cm. If we covert them to S/cm then the range would be 0.002 S/cm to 0.004 S/cm. I am growing starwberries through hydroponics and they are doing well with the EC range I have mentioned above. Can you please confirm whether the values you have chosen for EC standard solution for calibration is within this range or not?

  Are you sure? yes | no

Michael Ratcliffe wrote 06/25/2019 at 14:18 point

Hi, Depending on crop the range is more around 0.5-4mS/cm. The units for the probe and code are mS/cm, whilst this is not a common unit in the scientific world it is the defacto units used at a farmer level. 

All the best,

Mike

  Are you sure? yes | no

askhasangms wrote 06/05/2019 at 20:35 point

I am having trouble calculating the value of cell constant, K. I am using a calibration EC solution of 0.0045 S/cm. The K value turns out to be around 50 for most of the times. My aim is to get a value of K around 3 as you have mentioned. I am using a plug almost same as yours. What would you suggest me to correct the value of K?

  Are you sure? yes | no

Robert wrote 05/01/2019 at 20:35 point

So i'm looking at the AMTEL to calculate the resistance internal to the pins. How did you determine your values from the chart? there is no 0.4 volts and if you plug in values of V and I from the char you get like 900 ohms?

  Are you sure? yes | no

rahul.kumar wrote 07/10/2019 at 07:55 point

I am also facing this issue. Unable to find voltage and current mentioned above.

  Are you sure? yes | no

exploredevelopers wrote 02/13/2019 at 05:41 point

Hi Micheal !

I'm unable to find onewire & dallas temperature libraries, can you share the files here ?

  Are you sure? yes | no

vikash kumar wrote 01/29/2019 at 10:08 point

Hi Micheal, I'm not able to get the proper compilation of the program. plzz help me out.

  Are you sure? yes | no

Michael Ratcliffe wrote 01/29/2019 at 15:35 point

Copy and paste the code into the ide from the blog post, still have issues? 

  Are you sure? yes | no

vikash kumar wrote 01/29/2019 at 15:39 point

yes I have done this many a times. Its not working. I had added the libraries also. Still its not getting compiled 

  Are you sure? yes | no

Michael Ratcliffe wrote 01/29/2019 at 16:02 point

Cant help you with that information, I mean you haven't included the compile error etc.

  Are you sure? yes | no

vikash kumar wrote 01/31/2019 at 07:24 point

Arduino: 1.8.2 (Windows 10), Board: "Arduino/Genuino Uno"
C:\Users\vikash tiwari\Documents\Arduino\EC_METER_CODE\EC_METER_CODE.ino:1:21: fatal error: OneWire.h: No such file or directory
 #include <OneWire.h>
                     ^
compilation terminated.
exit status 1
Error compiling for board Arduino/Genuino Uno.
This report would have more information with
"Show verbose output during compilation"
option enabled in File -> Preferences.

this was the error message. I think I need the modified onewire and dallas library. Can you please provide me.

  Are you sure? yes | no

vikash kumar wrote 01/31/2019 at 07:29 point

Arduino: 1.8.2 (Windows 10), Board: "Arduino/Genuino Uno"
Build options changed, rebuilding all
Sketch uses 7056 bytes (21%) of program storage space. Maximum is 32256 bytes.
Global variables use 610 bytes (29%) of dynamic memory, leaving 1438 bytes for local variables. Maximum is 2048 bytes.
avrdude: ser_open(): can't open device "\\.\COM5": The system cannot find the file specified.

Problem uploading to board.  See http://www.arduino.cc/en/Guide/Troubleshooting#upload for suggestions.
This report would have more information with
"Show verbose output during compilation"
option enabled in File -> Preferences.

after addine Arduino inbuilt onewire library. I am getting this message

  Are you sure? yes | no

Michael Ratcliffe wrote 01/31/2019 at 11:07 point

First error message: You need to install the onewire library 

Second message: You have not selected the correct comport to upload /haven't plugged the arduino in. 

  Are you sure? yes | no

vikash kumar wrote 12/04/2018 at 10:10 point

I'm not able to get the code of the modified onewire dallas library. Kindly help me to get the modified onewire dallas library installed as soon as possible. 

  Are you sure? yes | no

charlie bart wrote 07/20/2018 at 12:54 point

Hi everybody.

I've been successfully reading EC with your code and my arduino UNO & Mega.

But the thing is, when i try with my ESP32 on an ADC port, i get incoherent voltages. Same thing with an arduino + ADS1115.

For example : 

Arduino+EC sensor+ 1.55mS calibration solution : Value read : 403 (so 1.96v)

Arduino+EC sensor+ 2.8mS calibration solution : Value read : 277 (so 1.35v)

Arduino without sensor plugged : Value read : 1023 (so 5v)

Arduino+ADS1115@15bits+EC sensor+ 1.55mS calibration solution : Value read : 24560 (so 4.6v)

Arduino+ADS1115@15bits+EC sensor+ 2.8mS calibration solution : Value read : 25520 (so 4.79v)

Arduino+ADS1115@15bits without sensor plugged : Value read : 27000 (so 5v under 15bits)

I'm always using a 1kohms resistor for those tests (between +5v and analog signal).

I don't have the numbers now for the ESP32+sensor@3.3v but it's the same than with the ADS (voltages read higher than on arduino).

Do someone have an idea about what is happening here?

I'm reaching the limit of my basic knowledge here :-(

  Are you sure? yes | no

466 wrote 04/10/2018 at 16:29 point

Nice project you got here!


Looking forward, creating an EC meter myself :)

Would you mind explaining:

Minimum EC 25=0.3 EC

and

Maximum EC 25= 3 EC 

?

Where do those values come from?

(Ok, 3 EC:  3 could be cell constant? but 0.3? did not find any info about it)

Thanks!

  Are you sure? yes | no

Cedric wrote 03/03/2018 at 12:39 point

Hi Michael, instead of a plug, I'm planning on using pin headers as the probe. However, I'm not sure what should the K constant would be. How did you get the K constant of the plug? 

  Are you sure? yes | no

Zaidan Alif Muttaqin wrote 02/02/2018 at 10:05 point

Hi Mike, can you solve my problem? the temperature is always -127 C.

  Are you sure? yes | no

Cody Lawson wrote 02/28/2018 at 03:36 point

Kind of late reply but -127 reading means bad probe or bad connection. If you have another probe try using that. If only one probe try resoldering the contacts or making a new plug depending on how it connects. That error likes to randomly pop up with these probes.

  Are you sure? yes | no

carlos.bruckner wrote 01/28/2018 at 14:48 point

Hi there. I've been playing with a similar project for a few days and fighting some distortion on my readings as the conductivity goes up. My guess is that it is due to polarization.
I'd like to add a few notes and questions:

a) When you say "You want the Solid Prong type plug" you meant the right end of the cable or to not use plugs with those little holes on the side? I changed one with the little holes to to another one (solid, round, little smaller area) and I think I got better readings.

b) The first reading gets discarded. Can I do that before setting the power pin, to high?, ie:

analogRead(ECPin);// first reading discarded, outside "HIGH zone"
digitalWrite(ECPower,HIGH); // or PORTA = B00000010; (for digital pin 23)
sensorValue = analogRead(ECPin);
digitalWrite(ECPower,LOW);

That way we spend less time with ECPower "on", so less polarization.

d) By my calculations, the resolution at a given water resisitance Rw is:

Res = R1 / (R1+Rw)^2 - which means that you should use R1 as close to Rw as possible.

However, while having the water at around 974uS (and my system reading it as 530~550ohm), to improve resolution, I used a 520ohm resistor, but the readings were more accurate when I used 1500ohm. My guess is that bigger resistance R1 means less voltage on the water, and thus, less polarization. If that makes sense, I think it is worth noticing.

Oh, and my K is currently 1.965.

Anyways, great project and great description, it was exactly what I was looking for.

  Are you sure? yes | no

Cody Lawson wrote 01/23/2018 at 22:36 point

Have you experimented with the ESP32 at all? My RO system already runs on one and it actually has multiple analog inputs compared to the ESP8266. I started to dive into it but the coding looked beyond my level. It appears a lot of the ADC stuff is still in development.

Some more helpful info. I called HM Digital and found out their inline probe K value is 0.67. Not sure if HM Digital is very big in the UK but in the US they seem to be the standard with RO units. Unfortunately their monitors do not have a data out but doesn't mean you can't recycle the probe! Also no data on the internal thermister (said they would have to call Korea) so an external temp probe is still needed.

If using their probes (or similar inline push fitting style) make sure you calibrate them INSIDE their T-fitting. The prongs are compressed together slightly while in the fitting and it appears the calibration point takes that into account. Results still weren't perfect using the 0.67 and 10,000K resistor but they were significantly better compared to measuring a naked probe. At ~100ppm I read 105, ~13-14 I read 15.46, and at ~3-4 I read around 6 ppm. My other probe from China I calculated K to be around 0.565. I would say a K range of 0.4-0.8 would be good for these smaller probes.

  Are you sure? yes | no

Cody Lawson wrote 01/22/2018 at 03:42 point

Great project! I ended up making a few interactive graphs with your equations for both the Arduino and ESP8266 boards. Both the K value and R1 have sliders so you can see how each will change the curve. Apparently if you double the R1 value but halve the K value you get roughly the same curve. That was one of a few different "OOOOOHHHHHHH, that's how that works!" moments. At least this felt better than plugging random numbers in and hoping for the best haha.

Arduino: https://www.desmos.com/calculator/qzbszykkbj

ESP8266: https://www.desmos.com/calculator/moptex4ufc

Feel free to reuse these if you want. Figured someone out there would appreciate the little extra help :)

Graph Instructions:

The X axis is the "Raw" variable, so 1 - 1024. The Y axis is PPM (Hana 0.5). The adjustable equation is the Red line. Sliders can be found at the bottom of the menu on the left. It starts off with what you chose in the tutorial. The next 4 lines can be used for comparison. They are set near my calibration point with Purple = R 10000, K 0.5; Orange = R 10000, K 1; Blue = 20000, K 0.5; Green = R 20000, K 1. Both Orange and Blue should basically be on top of each other  (what I referenced earlier).

If you want to convert back to EC just delete the *500 at the very end of the equation. To change to a different PPM just change 500 to 640 or 700. To change the range or step of K or R1 click on either the min or max number found on either side of the slide. Everything else I am going to assume you guys can figure it out.

  Are you sure? yes | no

frozenfritz wrote 12/12/2017 at 17:32 point

Hi Michael, I have a question for you: why do you use A8 as a power source? could I also youse a "normal" digital pin from an arduino? Really nice project btw!

  Are you sure? yes | no

Michael Ratcliffe wrote 12/14/2017 at 19:24 point

Yep, for a driving pin you can use a digital pin. 

  Are you sure? yes | no

cyberdevil001 wrote 12/10/2017 at 00:30 point

Hi Michael,

I'm using your code to measure and log the TDS that comes out of my RO system to replace the existing TDS meter. I only noticed that it seems to have issues with the low value measuring of the output water. This is currently 0 while before the DI resin this is between 1 and 7. Currently using the following probe: https://www.aliexpress.com/item/New-G1-4-0-8MPA-Water-quality-probe-TDS-conductivity-test-water-quality-testing-probe/32630094172.html?spm=2114.01010208.3.226.RCrLgm&ws_ab_test=searchweb0_0  . Calibrated this on 342ppm calibration water and when measuring 0 ppm water (demineralised water), it actually shows a few ppm instead:

Rc: 261350.00 EC: 0.00 Simens  2 ppm  21.69 *C 
Rc: 261350.00 EC: 0.00 Simens  2 ppm  21.75 *C 
Rc: 261350.00 EC: 0.00 Simens  2 ppm  21.75 *C 
Rc: 348816.65 EC: 0.00 Simens  1 ppm  21.75 *C 
Rc: 261350.00 EC: 0.00 Simens  2 ppm  21.75 *C 
Rc: 208870.00 EC: 0.01 Simens  3 ppm  21.75 *C 
Rc: 261350.00 EC: 0.00 Simens  2 ppm  21.75 *C 

Is there a way that I can tweak it so it will be able to measure these low values? Currently using the 1kOhm resistor as I'm still in the process to understand the R1 calculation you did in combination with my probe.

Thanks!

  Are you sure? yes | no

cyberdevil001 wrote 12/10/2017 at 02:28 point

I did try replacing the 1kOhm resistor with 2 in parallel (=500Ohm) and this seems to provide a bit more stable value, but still off, even with a new sample of 0ppm. (Higher value was spot on the calibration point) Also increased the resolution on the output of EC:

Rc: 268250.00 EC: 0.0050 Simens  2 ppm  16.56 *C 
Rc: 268250.00 EC: 0.0050 Simens  2 ppm  16.62 *C 
Rc: 178650.00 EC: 0.0075 Simens  3 ppm  16.62 *C 
Rc: 268250.00 EC: 0.0050 Simens  2 ppm  16.62 *C 
Rc: 268250.00 EC: 0.0050 Simens  2 ppm  16.69 *C 

So it is close, but for what I need to use it, the 2ppm off is still to much. Maybe 2 spot calibration could solve it?

  Are you sure? yes | no

cyberdevil001 wrote 12/11/2017 at 12:07 point

Trying to see if a 16-bit ADC will solve my issue. Will update in a few days.

  Are you sure? yes | no

Michael Ratcliffe wrote 12/14/2017 at 19:31 point

Hey, I made this with higher ppm in mind. 

Some advice to make it work better for low EC Start with the ones at the top:

>Increase the Value of R1 [ google voltage dividers and resistance of fluid vs ppm for your probe constant] I would imagine a 5-10K reistor would work better for you but I haven't done the maths.

> Potentially increase the delay after powering the probe up and taking a reading [LOW EC water shouldn't polarise easily]

Also you will need to look for a data sheet on temperature correction for low EC water, the one I use in the sketch is a estimate for hydroponics solutions]. Distilled water will absorb CO2 and this means distilled water does have a very small ppm.

  Are you sure? yes | no

yuyouliang wrote 11/09/2017 at 05:09 point

Dear Michael, thank you for sharing this project. It's wonderful. I have some questions. Could you explain 
 why EC = 1000/(Rc*K)? Why the number is 1000?  And the unit is S/cm? not ms/cm? Thanks.

  Are you sure? yes | no

denysafari wrote 10/29/2017 at 09:16 point

HI michael , i built this one, but i had a problem. the temperature result is negative (-127). could you help me please?

  Are you sure? yes | no

kutlutug wrote 11/13/2017 at 08:43 point

I have the same problem.  My temperature is -127 and my ec numbers dont change in different liquids even wiring the plugs polars with a wire. Is this problem about my hardware?

  Are you sure? yes | no