Pull-up resistor on bread board

Thanks all for the very helpful information.  I now have the board reading in too sets of data.  To test the to see if it accurate I put a sensor in a glass of ice.  I get a constant 23.45 degrees Fahrenheit.  I assumed it would be at 32 degrees.  I have to order the more equipment for the rest of the project, but right now my concern is how to calibrate the sensors.

Thanks again all!!

According to the device datasheet it is calibrated in C.  

"The DS18B20 output temperature data is calibrated in degrees Celsius; for Fahrenheit applications, a lookup table or conversion routine must be used."

If you hare having an error with them, I would look at your code first.

Regards,

Dave M.

Hi Dave,

  The Celsius reading is -0.75 in icewater...from what I've read that is the way to calibrate for thermometer.  From what I read the DS18B20 is more accurate than most store purchase thermometers....and for this application it does not have to be bullsye accurate just close as possible.  So far so good in the project!

Thanks again

Hi Damian,

According to the datasheet (http://datasheets.maxim-ic.com/en/ds/DS18B20.pdf), the accuracy should be +/- 0.5 °C.

Melting ice has a temperature of 0 °C, so your reading should be between -0.5 and 0.5... Did both of your sensors measure -0.75?

Maybe your ice water was too icey...

Anyhow there is no way to recalibrate the sensor (as far as i know) but you could just add 0.75°C to your reading in your software to correct for the drift of the sensor.

Quinten

Hi Quinten,

I became suspicious about the accuracy of the sensor because I'm getting different readings from them and they are right next to each other:

Getting temperatures...

Sensor 1:  C: 27.00 F: 80.60

Sensor 2:  C: 28.50 F: 83.30

Getting temperatures...

Sensor 1:  C: 27.00 F: 80.60

Sensor 2:  C: 28.50 F: 83.30

Getting temperatures...

Sensor 1:  C: 27.00 F: 80.60

Sensor 2:  C: 28.25 F: 82.85

I could just bring up or down one programatically, but which?  and why are the reading differently? 

It is a problem , but it is fun.

Thanks!

Hi Quinten,

I became suspicious about the accuracy of the sensor because I'm getting different readings from them and they are right next to each other:

Getting temperatures...

Sensor 1:  C: 27.00 F: 80.60

Sensor 2:  C: 28.50 F: 83.30

Getting temperatures...

Sensor 1:  C: 27.00 F: 80.60

Sensor 2:  C: 28.50 F: 83.30

Getting temperatures...

Sensor 1:  C: 27.00 F: 80.60

Sensor 2:  C: 28.25 F: 82.85

I could just bring up or down one programatically, but which?  and why are the reading differently? 

It is a problem , but it is fun.

Thanks!

I think the best way is to put the sensor in ice water (which you know is 0°C) to "calibrate" it in software.

This ofcourse means that your code will be specific for the sensors you "calibrated", not the best solution but it will give accurate temperatures as long as you keep using the same sensors...

As for why they give different reading, every sensor is going to be a little different. Actually no two pieces of any electronics device are going to be exaclty the same...

Maxim probably exaggerated a little by stating the 0.5 degree accuracy. If that were true the reading shouldn't be more dan 1°C apart.