Hello,

I come up with problem, and i'm looking up for solution.
I got basic idea, what i want to do, but problem is with calculations.

I have heating element, i would like to turn it on / of on certain temperature.
I would like have range between 40 - 60 Celsius, made by few presets.

For example
- 40C
- 45C
- 50C
- 55C
- 60C

I will get to that later, because idea may change during project.


What's my idea ?

I would like use thermistor NTC 100Ω model B57164K0101J parameters
100Ω25*C +- 5% R=5mm
K(B25/100)=3200 ±5%


Idea is:

Use 3.3V voltage source.
Use that voltage source on voltage divider made by 100Ohm termistor AND 100 ohm resistor.
Then use second voltage divider made by 200Ohm variable resistor.

That 2 input, connect to LM324 OpAmp

As fare as i remember, NTC lose resistance when temperature RISE.
When resistance will goes down, voltage divider will go up.
Than OpAmp Comparator will send at output signal.
And i can plug there mosfeet or something else, like relay.

I know i can get pre set temperatures by measuring resistance of variable resistor, measuring temp, and replace them by resistor.

But... i'm looking for something else!

How can i count it...

I gues it will be better to count that on paper.

On thermistor say

100Ω25*C +- 5% R=5mm
K(B25/100)=3200 ±5%

100Ohm in 25*C

How calculate resistance in for example 40*C ?

I know it's non linear.
There are special calculator to do this ?

Is my thinking is correct ?

 

The Steinhart-Hart equation is typically used for calculating the temperature of a thermistor given a resistance input.
http://en.wikipedia.org/wiki/Steinhart–Hart_equation

A good reference to calculate the steinhart coefficients for your beta value (K) is here:
http://www.ilxlightwave.com/technotes/thermistor_constant_conversions_beta_steinhart-hart.pdf

Also:
http://www.ilxlightwave.com/appnotes/AN 4 REV03 Thermistor Calibration and Steinhart Hart.pdf

 

Have you looked at the LM35 IC. For your temperature range I would recommend this instead of using a thermistor.

 

The Steinhart-Hart equation is typically used for calculating the temperature of a thermistor given a resistance input.
http://en.wikipedia.org/wiki/Steinhart–Hart_equation

A good reference to calculate the steinhart coefficients for your beta value (K) is here:
http://www.ilxlightwave.com/technotes/thermistor_constant_conversions_beta_steinhart-hart.pdf

Also:
http://www.ilxlightwave.com/appnotes/AN 4 REV03 Thermistor Calibration and Steinhart Hart.pdf

Thank You, that's look userfull.

From myself i would like to share this page
http://www.giangrandi.ch/electronics/ntc/ntc.shtml

At the bottom is calculator. You can Put there values from datasheet, and it will calculate resistance at any temperatures. I'm not sure is this reading are correct, but i hope so!.

Have you looked at the LM35 IC. For your temperature range I would recommend this instead of using a thermistor.

That's IC LM35 looks brilliant!.

Could i conntect output directly to input of LM324 ?

If i understand correctly, i can hook it up with +4V.

It should add 10mV / 1C to output.

So at 40*C i should get 400mV ?

Is this correct ?

What if i would like measure in 2 points ?

Actually not measure, but detect "overheat".
Could i connect them together in parallel and put in 2 different points ?

 

I've done a few of the lm324 fan controllers,i always just used trim pots instead of fixed resistance.

10k pot,10k thermistor and a handy dandy IR thermometer to set the temps to my liking.

Word of warning w/ this setup, zero hysteresis it will kick off and on constantly as temps fluctuate.

Another warning op amps don't like when you accidentally hook the power up reversed polarity.

They go pop w/ no warning

-luvv-

 

I've done a few of the lm324 fan controllers,i always just used trim pots instead of fixed resistance.

10k pot,10k thermistor and a handy dandy IR thermometer to set the temps to my liking.

Word of warning w/ this setup, zero hysteresis it will kick off and on constantly as temps fluctuate.

Another warning op amps don't like when you accidentally hook the power up reversed polarity.

They go pop w/ no warning

-luvv-

Hello,
That's great answer!

I will do this way.

Word of warning w/ this setup, zero hysteresis it will kick off and on constantly as temps fluctuate.

I love this!

Right now i use bimetal thermostat and that's cause problem. When it turn off, i have wait to much time until it cool down.
That's main reason why i change to more electronic design than mechanical...

Another warning op amps don't like when you accidentally hook the power up reversed polarity.

They go pop w/ no warning

Few years ago, i make capacitor explode ( i take from wrong box, and don't look at voltage).
Now i always wear glasses when i'm playing with electronic, soldering iron etc. and i always recommend this. Safety first!

Regards

 

Thank You, that's look userfull.
If i understand correctly, i can hook it up with +4V.
It should add 10mV / 1C to output.
So at 40*C i should get 400mV ?
Is this correct ?
What if i would like measure in 2 points ?
Actually not measure, but detect "overheat".
Could i connect them together in parallel and put in 2 different points ?

40*10e-3 is 0.4 so you are indead correct. And yes you can hook it up to 4 (up to 30 volt should also be OK) Here is the applicayion note for LM35 http://www.ti.com/lit/an/snoa748b/snoa748b.pdf Your application will easy to do with a micro controller with AD converter. But it is no problem doing it in an analog way. For your two point measurement use perhaps two analog comparators.
For your application what will be your supply voltage?