0
\$\begingroup\$

I bought a thermistor together with my Arduino, because I thought it would be a good first project. I should have gone a different way, since it is obviously too complex for my level, but well, now I have got it, and I am trying to understand it, so I hope you can help.

enter image description here

It is an Epcos, PTC thermistor, from RS-Online (link). The datasheet (link) shows the following charts:

enter image description here

Why are there seven different graphs (NTT = 60, 70, 80, 90, 100, 110, 120°C)? The only possible explanation I can think of is that the thermistor comes in severen different versions with each their NTT, and I have one of them (without knowing which one). Is this correct, and if not, then what is this NTT?

NTT is short for "nominal temperature threshold", but little shows up when I Internet search for it.

The text on the thermistor says D801 A80. Perhaps the 80 means that it has NTT = 80°C?

\$\endgroup\$
1
  • \$\begingroup\$ Okay, I think that is done. \$\endgroup\$
    – gbulmer
    Sep 18 '14 at 18:53
1
\$\begingroup\$

Yes, the datasheet shows seven different order codes on page 3.

So there are seven different parts: enter image description here

B59801D0080A040 is the third one down in that list.

As you wrote, the 80A says its \$T_{NTT}\$ is 80 degrees C. That is the one in your posted photo.

It's resistance rises dramatically above 80C.

It doesn't look like it is intended for precise measurement. If it were, I would expect the datasheet to give the temperature coefficient, i.e. the change in resistance for a change in temperature, and it doesn't.

It is optimised for a 'trigger point', a dramatic change in resistance over quite a small range of temperature change. It's more like a thermostat, 'full-on' or 'full-off', than a sensor for actually measuring temperature.

The Spec says resistance changes from 570Ω at +/-5 degrees K to over 10,000Ω at +23 degrees K. A change of about 20x in resistance for a temperature change of about 6% (in Kelvin).

It's operating temperature range is stated as -40C to \$T_{NTT}+23C\$ providing the voltage used to measure it's resistance is \$V_{measure}\leq 7.5C\$.

However, the graph for \$T_{NTT}\$ goes up to 160C.

Apropos nothing:
That operating point is somewhat unfortunate. If it triggered below 100C, say 90C, it'd be able to detect water is close enough to boiling, without actually boiling, and so is a good temperature for making tea, Which would save quite a lot of electricity. (I'm British :-).

\$\endgroup\$
2
  • \$\begingroup\$ Interesring, hmmm... So this thermistor's characteristic is designed to trigger something around 80°, and not really designed to be used for actually measuring the temperature. And for measuring a temperature, I should have gotten one with a more linear characteristic? Sort of guessing here :/ :) \$\endgroup\$ Sep 18 '14 at 18:41
  • 1
    \$\begingroup\$ Yes, I agree. It triggers a huge resistance change above 80C. To measure temperature you'd use something with a more linear response. The datasheet for something like that will give a temperature coefficient which states the resistance change vs temperature change. Then it would be straightforward to measure, say with an Arduino ADC, and calculate temperature. \$\endgroup\$
    – gbulmer
    Sep 18 '14 at 19:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.