# How does a thermistor work (with a transistor?)

#### Cretin

Joined Dec 13, 2012
69
Hey all I, I've been fiddling with this the other day and I am trying to use a thermistor with another resistor to make a 2 transistor switch that triggers at the output....take a look at the schematic, and here are my values

VCC = 5V
Thermistor Temperature at room temperature: 5400 ohms
Thermistor temperature while holding it: 4400 ohms
Q1 and Q2 are NPN transistors with ~200 Beta

Basically, I want it so that once i place my fingers on the thermistor and dropped its resistance to 4400 ohms, then it triggers the base of the first transistor and sends voltage to the output of the second transistor (at collector/VCC junction on second transistor)

I calculated my voltage divider as such:

5V * RB/(Resistance of thermistor at room temperature+RB) = .5 V (<.7, so voltage is not high enough to trigger transistor)

Rearranging we have:

5RB = (.5*Rthermistor) + (.5*RB)

5RB = = .5*5400 + .5RB
4.5RB = 2700
RB = 600

So I place a 600 ohm resistor into RB, expecting the voltage to to increase to .7 at the base to trigger it..but nothing happens...what am i doing wrong

#### #12

Joined Nov 30, 2010
18,210
Your second transistor is always on until the first transistor grounds its drive current, but to do that the first transistor has to get its Vce really low, saturated. You are figuring the current to just start turning on the first transistor, but you need it slammed on.

#### Cretin

Joined Dec 13, 2012
69
Hey #12, any calculation suggestions?? What am i missing here?

#### kubeek

Joined Sep 20, 2005
5,718
Hey #12, any calculation suggestions?? What am i missing here?
So I place a 600 ohm resistor into RB, expecting the voltage to to increase to .7 at the base to trigger it..but nothing happens...what am i doing wrong
Didn´t go through the equations, but this stands out: Changing Rb from 610 to 600 ohms will definitely make the voltage on the base go lower, not higher.

now that I look at it, you are calculating it for 0.5Vb so no wonder it is not 0.7V.

#### Cretin

Joined Dec 13, 2012
69
Didn´t go through the equations, but this stands out: Changing Rb from 610 to 600 ohms will definitely make the voltage on the base go lower, not higher.

now that I look at it, you are calculating it for 0.5Vb so no wonder it is not 0.7V.
Don't i have to calculate the voltage for RB first, by doing a voltage divider using the thermistors room temperature value first, and THEN substitute that new value back into the equation??

#### kubeek

Joined Sep 20, 2005
5,718
I think that would apply when you are trying to see what resistance the thermistor needs to be to trigger the transistor.

#### #12

Joined Nov 30, 2010
18,210
You are much better off with a 1k pot in the bottom position so you can see (with a meter) why it doesn't work.

This is why differential amplifiers were invented. The Vbe changes with temperature, it is different from one part number to the next, when the applied voltage slowly moves up into the operating range, the transistor s l o w l y starts to conduct. This is NOT a snap-action switch.

#### ScottWang

Joined Aug 23, 2012
6,882
In the slow changing circuit, you can in seires a 1~5K resistor with Q2_b, it will more easy to make Q1_Vce to cut off the current of Q2_b.

#### Cretin

Joined Dec 13, 2012
69
Gusy I have tried everything and nothing is working.... Please help me out, with referencing the image in the first post...

If my Thermistor is 5000 ohms at room temperature (non triggered state), and then drops to 3500 ohms when heat is applied (triggered state) then I want the transistor to send current to an output at its collector.

What values do i need, how do i do this..I have tried everything guys and it just won't work...

#### kubeek

Joined Sep 20, 2005
5,718
The problem with this single transistor configuration is that the setpoint heavily depends on the supply voltage and on temperature of the first transistor, so as these change the setpoint will change too. Also you don´t really define what should happen between the 5000 and 3500 ohms.

In my simulation using 2n2222 and 520 ohm resistor instead of the 600 one I get it to trigger at about 3800 ohm in the thermistor. Vbe is at around 0.63V at the trigger point. Changing the temperature of the first transistor to 50°C then changes the setpoint to about 3300 ohms.

#### Cretin

Joined Dec 13, 2012
69
If it is really difficult for a single transistor, then perhaps i could use a two transistor? I'm just really lost kubeek...I need help understanding how to design a transistor trigger point with voltage divider

#### kubeek

Joined Sep 20, 2005
5,718
If you want this as anything else but getting experience with transistors, then simply get a comparator and do it the "proper" way. For example LM393 is a pretty standard part with wide power supply range and probably lots of examples on the net.

With a circuit like that you will definitely need a trimpot to set the actual trigger point.

#### Cretin

Joined Dec 13, 2012
69
I have to use a transistor unfortunately. Could you point me to a schematic i could look at, or could you pen one yourself or of your simulation of how this circuit works?

#### kubeek

Joined Sep 20, 2005
5,718
I simulated the same circuit you posted. You simply need to change the 600 ohm resistor to a 390ohm and a 500ohm trimpot in series to adjust it to your transistors and thermistor.

#### Cretin

Joined Dec 13, 2012
69
so the 390 ohm resistor and trim pot are in series from base of the first transistor (the one on the farthest left) going to ground?

#### kubeek

Joined Sep 20, 2005
5,718
yes........

#### Cretin

Joined Dec 13, 2012
69
Hey Kubeek so it is still not working...here is a picture of my circuit setup.

Things you should know

VCC = 6V
Potentiometer = 5k
Blue resistors are 5 colour band, but the values all matchup to what found

Everything should be connected correctly according to the original schematic, but please let me know if you see any errors.

Questions I have:

1: when hooking up a potentiometer in this circuit, do i connect the wiper to the circuit, and then one of the outer pins to ground? Should i be ground and powering the potentiometer?

2: the resistor coming out of Transistor 2, is it hooked up correctly??

#### kubeek

Joined Sep 20, 2005
5,718
First, I don´t see the bottom leg of the first transistor connected to anything.
Second, what transistors do you use? Any chance you have the footprint wrong?

#### #12

Joined Nov 30, 2010
18,210
You have this set up as 2 transistors fighting. Here is a circuit with 2 transistors cooperating.

#### Attachments

• 8.5 KB Views: 48