RTD PT100 solved almost

[internetuser2k11]

RTD PT100 and other measurements. The PT100 negative temperature measurement is giving a 4-7 degree C variance for the range (-200 - 0 degree C). I don't know why RA1/AN1 is not working. (you can check by shorting RA0 and RA1). RA1 is not giving any measurement values. You have to replace 2.5 V Battery by 5V Vin - 2.5V Vout type Precision voltage regulator/reference

 

[wp100]

Hi,

Just out of interest, why a PT100 sensor ? What are you measuring that needs such a vast temp range ?

Seems a lot of additional hardware and program code to give an accurate result.

Then you go the the LM35, again a linear device.

Did you ever consider the DS18B20 or SHT11 sensors, both digital and a temp ranges of over -40c to +120c.
The Ds18b20 having the advantage it will work on very long cables.

 

[internetuser2k11]

I did that project for someone else. He said that he was going to use 2 PT100's to measure temperatures of boiling water and oil.
I am not sure wheather I can use DS18B20 or SHT11 with boiling oil. I could have used DS18B20 instead of LM35 but he said he wanted LM35 only.

 

[FvM]

Your post is missing any information for detail help with your questions.

Regarding voltage references, you'll possibly remember my previous comment that Pt100 measurment is essentially ratiometric and doesn't need a stable voltage reference, if Pt100 circuit and ADC are using the same reference voltage. Apparently, you didn't yet think about the point.

I am not sure wheather I can use DS18B20 or SHT11 with boiling oil.

You can't.

 

[internetuser2k11]

No, I didn't get a 2.5V reference component in isis. So, I've used battery. I will try to use a potential divider to divide the +5V to 2 x 2.5V for op amp reference.

---------- Post added at 03:32 ---------- Previous post was at 03:24 ----------

I did the interpolation this way......... for negative temperatures.

Ref http://en.wikipedia.org/wiki/Linear_interpolation

[syntax = c]

//Negative Temperatures (-200 degree C to 0 degree C)
           if (pt100_r1 < 100.00) {
              /*
              x = r = 72.33
              xo = ro = 18.52
              x1 = r1 = 100
              y= t = ?
              y0 = t0 = -200
              y1 = t1 = 0
              t = t0 + ((r-r0)t1 - (r-ro)t0) / (r1-r0)
              */
              r = pt100_r1;
              r0 = 18.52;
              r1 = 98.44;
              t0 = -200.00;
              t1 = 0.0;
              t = t0 + ((r-r0)*t1 - (r-r0)*t0) / (r1-r0);
              pt100_t1 = t;
           }

[/syntax]

but it is giving 5-7 degree C variance.


and there was a typo in the code... It was not displaying AN1 channel. Now I've fixed it.

 

[FvM]

It's a linear interpolation, not accurate enough for the large temperature range. You should use a polynomial or a multi-point linear interpolation.

 

[internetuser2k11]

Do you mean polynomial or polynomial interpolation?

Which One?

http://en.wikipedia.org/wiki/Polynomial_interpolation

http://rjwagner49.com/Mathematics/Interpolation.pdf section 2.4

---------- Post added at 08:42 ---------- Previous post was at 08:40 ----------

I tried a polynomial. It didn't work.

---------- Post added at 09:05 ---------- Previous post was at 08:42 ----------

Or should I use Polynomial Interpolation in this article http://en.wikipedia.org/wiki/Interpolation

 

[FvM]

You have read the Analog Devices AN. They suggested a polynomial for the negative temperature range, supplementing the "direct methode" for positive temperatures. There can be no doubt about the principle correctness of this solution, it's just a matter of implementation in your code.

Table interpolation methods are probably better in terms of code execution time.

 

[internetuser2k11]

You are right FvM. There was a mistake in my calculation. The equation for negative pt100 temperature measurement works fine.

 

[sresam89]

The equation for negative pt100 temperature measurement works fine.

can you please share the program am also working on the same.

 

[pridepk]

Hi,
If we want to take values in F fahrenheit so what should be the changes
Thanks

 

[FvM]

I presume you know the conversion formula.

 

[jayanth.devarayanadurga]

Use Celcius to Fahrenheit conversion formula you learnt in Physics. If you don't remember then google for it.

Anyways. Here is the formula.

°C x 9/5 + 32 = °F

 

[Mithun_K_Das]

Calculation is not clear to me. Can anyone explain the calculation? Is Op-Amp need to use to amplify the voltage?

 

[jayanth.devarayanadurga]

Why don't you find it yourself? You can measure the current and voltage befor the amplifiers and see.

 

[Mithun_K_Das]

tried that. not working. showing wrong data.

void Display_Temp(void)
{
  ch = (tlong / 1000) % 10;
   //Lcd_Out(1,3,"Tmp");
  Lcd_Chr(row,position,48+ch);          // Write result in ASCII format
  ch = (tlong / 100) % 10;
  Lcd_Chr_Cp(48+ch);          // Write result in ASCII format
  ch = (tlong / 10) % 10;
  Lcd_Chr_Cp(48+ch);          // Write result in ASCII format
  Lcd_Chr_CP('.');
  ch = (tlong ) % 10;     // Extract hundreds of millivolts
  Lcd_Chr_CP(48+ch);           // Write result in ASCII format
  Lcd_Chr_CP('/');
  Delay_ms(100);
}
/*******************************************************************************/
/****************************  Get_Temp   **********************************/
/*******************************************************************************/
 void Get_Temp(void)
 {
  ADCON0 = 0x01;// select channel 0
  for(i=0;i<5;i++)
  {
   adc_rd0 = ADC_Read(0);
   sample[i] = adc_rd0;
   Delay_ms(50);
  }
  average = 0;
  for(i=0;i<5;i++)
  {
   average += sample[i];
  }
  average/=5;

  Temp = (float)average*1.00/1024.00;// using 1V ref.
  Temp = Temp/1.4; // found that PT100 changes voltage of 1.4mV/C with a voltage divider with 1K.
  Lcd_Out(1,1,"TEMP<");
  Lcd_Chr_Cp(223);
  Lcd_Out(1,7,"C>");
  tlong = Temp*1000;
  row = 1;
  position = 9;
  Display_Temp();
 }

 

[FvM]

Calculation is not clear to me. Can anyone explain the calculation? Is Op-Amp need to use to amplify the voltage?

Which calculation for which circuit?

 

[jayanth.devarayanadurga]

Where is you circuit? Zip and post your project files. It will save time to create a new project and test the code. Give as much details as you can regarding your adc and pt100 circuit. You are using a simple adc calculation which will not work. Where have you implemented the RTD PTC100 equation for temperature measurement? You have to use a constant current source and PT100's resistance vary depending upon temperature. You find resistance first and use it in the equation to calculate temperature.

 

[Mithun_K_Das]

visit: https://www.edaboard.com/threads/293673/#post1261073

 

[jayanth.devarayanadurga]

Use Thermocouple.