Arduino and TSP? Problems and inaccuracies?

Question

Arduino

Good afternoon.
I am looking for a solution to such a problem, there is an Arduino uno and a platinum resistance thermometer (pt100 with a resistance of 100 ohms at 0 * C, and 138.5 ohms at 100 * C), you need to get the temperature value from it.
My actions:
Since the Arduino cannot read the current at the analog input, but only understands the voltage, I had to assemble a voltage divider from a 100 ohm resistor and the RTD itself.
(It is impossible to connect the TSP directly with one wire to 5v, and the other to the analog input without using a voltage divider) - If there is such an opportunity, please explain how?
Next, I calculated what value corresponds to one division of the Arduino analog input: 5v / 1024 \u003d 0.0048v.
Using a calculatorfor the voltage divider, I calculated that the voltage for a temperature of 20 * C (107.79 ohms) should be 2.593v, which should correspond to the value (2.593v / 0.00488 \u003d) 531 from the Arduino analog input. What it approximately does not correspond to by 10-12 units.
Then I calculated the values ​​for 0*C(100 Ohm)=2.5v/0.00488=512 and for 100*C(138.5 Ohm)=2.9035v/0.00488=595 and set the translation of these values ​​from 0 to 100.

void setup()
{
Serial.begin(9600);
}

void loop()
{
Serial.print( map ( analogRead(A0) , 512 , 595 , 0 , 100 )  );
Serial.println("");
delay( 1000 );
}

And as an output value, I get an average temperature of 20 * C more than it should be.
What is wrong or what am I doing wrong?

Answer 1

here the accuracy will change when changing the shoulder resistor. They need to play. And it should be with a low percentage of deviation. In addition, the thermocouple measurement is non-linear.
I would advise you to choose something simpler for measurement: like ds18b20. And yes, it doesn't cost that much.

Answer 2

it is not a thermocouple that is used there, but a platinum resistance thermometer PT100 α = 0.00385, here GOST 6651-94 fully describes how the resistance changes and all the calculation formulas. If I were you, I would check the divider itself with an accurate instrument and measure the voltage drop across the TSP at the control points at 0 and 100 degrees, and everything became clear. What makes the error either the ADC itself or your divider. But you will have more error because both the divider and the ADC have their own error