Thermistor - Voltage Circuit.

Thread Starter

NMC

Joined Nov 16, 2018
13
Hi All,

I am looking to design a circuit using an Omega4404 thermistor which at 20 degree's outputs 0v.
I want to draw a graph of corresponding voltages for 20-30 degrees.

What type of circuit do I need to design?

Thanks
N
 

Alec_t

Joined Sep 17, 2013
14,280
Jeepers! That 44004 is expensive! There are MUCH cheaper thermistors available and which may do the job, unless you are constrained to have that particular resistance (2252Ω @ 25C).
What tolerance is acceptable for your purpose?
What is the purpose?
 
Last edited:

MrChips

Joined Oct 2, 2009
30,720
Thermistors, in general, are non-linear resistors with negative temperature coefficients (NTC), i.e. resistance goes down as temperature goes up. The simplest way to get a linear response is to get the data into a computer and do the data to temperature conversion mathematically.

If you want 0V output at any specific temperature, you can use a difference amplifier to subtract any DC offset voltage or use a bridge circuit.

Personally, I prefer to convert the thermistor resistance to a frequency and get a microcontroller (MCU) to read the frequency. Without knowing your specific application it is difficult to suggest any such options.
 

Thread Starter

NMC

Joined Nov 16, 2018
13
Hi all,

Thanks for the replies.
Apologies it is an Omega 44004.

I want to design a circuit that at 20 degrees Celsius the output voltage is 0V.
I then need to determine the voltage for temperatures 21-30 degrees Celsius which have given resistances as below:

20 degrees C = 2814 Ohms
21 = 2690
22 = 2572
23 = 2460
24 = 2354
25 = 2252
26 = 2156
27 = 2064
28 = 1977
29 = 1894
30 = 1815

Below is a circuit which I tried to simulate -

upload_2019-3-13_16-40-8.png
 

MrChips

Joined Oct 2, 2009
30,720
This is a bridge circuit.

upload_2019-3-13_12-42-42.png

You can make three resistors all 2814 ohms.
Or you can make one resistor 2814 ohms (eg. Ra is thermistor, Rb = 2814 ohms).
R1 = R2 = 1000 ohms, for example.

or R1 = 2814 ohms,
Rb = R2 = 1000 ohms, for example.
 

crutschow

Joined Mar 14, 2008
34,285
Here's the bridge circuit with an op amp differential to single-ended output (LTspice simulation below).

Q1 and Q2 generate a current load on the output to help pull the output close to zero, which single-supply op amps don't generally do.
If you have a plus and minus supply for the amp, then those aren't needed.

The circuit gain is 2 as compared to the voltage across the same bridge connected to ground.

Pot U2 is for offset adjustment.

As you can see, the output voltage goes from ≈0V to ≈1V for a temp change of 20 to 30°C.

upload_2019-3-13_12-17-13.png
 
Last edited:

Thread Starter

NMC

Joined Nov 16, 2018
13
Hi All,

Thanks for the replies again.
I'm still struggling to design a circuit for my thermistor-voltage circuit.
Do I use a simple bridge circuit or do I need to use some sort of op-amp?

N.
 

joeyd999

Joined Jun 6, 2011
5,237
Because an MCU is better at counting pulses than measuring analog voltages.
Seriously: I would love to see your calibrationless T-to-F solution that gives milli°C resolution and centi°C accuracy with only one external component besides the MCU and thermistor.

Edit: I've discussed my solution.
 

MrChips

Joined Oct 2, 2009
30,720
Seriously, no one is talking about calibrationless temperature measurement. The thermistor is non-linear and some form of transformation is required. Thermistors are interchangeable once the system is calibrated.

My solution consists of thermistor, capacitor, and MCU.
 

MrChips

Joined Oct 2, 2009
30,720
How does that accurately convert the thermistor resistance to a frequency? :confused:
I charge the capacitor then allow the capacitor to discharge through the resistor. I do this many times and then count how many times I can do this in a certain time period.
 

crutschow

Joined Mar 14, 2008
34,285
I charge the capacitor then allow the capacitor to discharge through the resistor. I do this many times and then count how many times I can do this in a certain time period.
And you really think that's more accurate than using the processors A/D converter to measure an analog voltage, such as from the bridge circuit?
 
Top