Finding the time constant of a thermometer

[strigner]

The time constant of a thermocouple is to be found by exposing it to a step change.

Then the thermocouple and a thermometer are exposed to a ramp function and measurements from each are taken simultaneously.

The time constant and its uncertainty for the thermometer are to be found.


I think the time to get 63% of the way to the final temperature can determine the time constant for the thermocouple. But then I don't know how that can be used to find the time constant of the thermometer when they both are giving a ramp input.

 

[vk6kro]

Sounds like two different experiments.

You plot the output with time after plunging the thermometer and the thermocouple into boiling water.

The problem will be that this will happen very quickly and you will have trouble timing it.

If you wait long enough, (maybe a minute) the reading from both devices should be stable and you can compare this final reading with the ones you obtained at the beginning.

Because all this happens so quickly, there is a better way.

You could do an initial run to get the final temperature for each device.

Then calculate the 63% point for each device.

Then do the experiment again and use a stopwatch to find the time to get to this temperature readout.

For example with the thermometer, if room temperature is 20°C and the final temperature is 100°C then the difference is 80 degrees.
63% of this is 50.4 degrees so add this to the 20°C to get 70.4°C. So, you just have to time it until the thermometer reads 70.4°C.

It won't be easy and you may have to do it a few times to get an average. Let the thermometer get back to room temperature in between readings.

Then do the same for the thermocouple.
You now know the time constants.

After that, you can slowly heat some water and compare the readings on the thermocouple with that on the thermometer so that you can calibrate the thermocouple for use as a thermometer.