Tolerance of Resistor when using Thermistors for Temperature Readings

[aubrey4485]

@JohnRob

Is your 2° F error from reading the NTC value from the ESP

This is what is happening… the analog value coming into the analog input for the apparent displayed temperature has changed significantly according to my calibration table. What set me to look into this is the kids complained it was much warmer than usual. I thought the setting must have been played with but NO, it was at the usual safe kid temperature setting and displaying the safe temp as well.

I wish I had a table for the NTC thermistor… it is buried deep within the sprayfoam of
the tub, i was rushing to get this done before the tub and pipes froze and never obtained a part number. when I reached out to the manufacturer they said, it was proprietary knowledge and they were not at liberty to discuss such matters with consumers . Hot tub dealer said the same. Job security I guess… LOL??

The setpoint temp varies from 95 in standby/not used to 98-99 for kids to 102-103 for adults. And has seemed fairly consistent and good results from last fall/winter until now

I have attached pictures of the thermowell that puts the thermistor in “contact” with the water.

[build_1971]

It is not the ntc that changed. It is also not a problem that esp adc is not linear. It is your reference that has changed.
To start with: get a better reference. Recalibrate your sensor. Two point calibration should do as your operating window is pretty small...
If still a problem: install a second ntc near your reference. Make a calibration fit for your pool ntc at different board temperatures (taking into account both ntc readings).
Have fun!

[aubrey4485]

So much advice… it can make you dizzy!! LOL. Thanks for this… i am still going to start with the new reference and new calibration and go from there…

4kids, sports and life will dictate how long it takes me to complete and get back and post results… however… I will.

[JohnRob]

Being an electrician, can you simply put you multimeter on the NTC and read the resistance? Then if things change you will have another data point.