IR Temperature Sensor (ZTP-135S)

Hi all,

I am new to this forum and have a question relating to the output voltage of the **broken link removed** IR temperature sensor. I'm currently working on a small project which uses to sensor to read the temperature of an object which is placed in close proximity to the sensor (same distance every time). The circuit contains an Op-Amp and RC circuit to provide an amplified stable voltage output from the thermopile device and an independent reading of the thermistor for compensation. The problem I am having is converting the amplified thermopile voltage to the correct temperature. Can anyone advise me to the relationship between these two. Is the relationship purely linear? i.e Say the thermistor temp reads 25 deg/C and the Vout for the therm is 1.3V, will a change in Vout relate to a change in T linearly or is there more to it? I have trawled google for quite some time in search for an answer but am no further ahead at this point. Any help would be greatly appreciated.

Regards,

P-J-V

Different models of thermistors/thermopiles behave differently. You're pretty much left to experiment with a given unit, to find how to adapt it to your particular need.

Response is not necessarily linear.
Or it may be linear for the thermistor, but not linear for the thermopile.
Or even if both have linear response, they may not have the same coefficient, or the same zero-offset coordinate.

Are you using the thermopile to show room temperature, or as a reference to compare the thermistor output to? (Or vice versa? It isn't clear.)

Whatever sense circuit you use, and whatever volt reading you get, you must also multiply by a certain coefficient, and add a certain constant voltage.

You must experiment to determine the correct gain, and the correct add-on voltage.

You must run enough tests to get a consistent response for a given room temperature. And as you run each test, you must wait a minute or more for the device to match surrounding temperature.

The whole calibration process become time-consuming. And that's just for your reference thermo-device.

Then you must do the same with your measuring device.

It becomes unwieldy, doesn't it?

This is why we don't find a schematic that applies to all thermistors. Even with a packaged unit (example, Radio Shack) that give a table of readings, it's up to you to determine whether it is reading correctly over the entire range of temperatures you wish.

Hi Brad,
Thank you for taking to time to reply to me. I am using the thermistor as a point of reference as what I have gathered so far is that the thermopile is used to calculate the difference (not the actual) between the cold and hot junctions and provides this value as a voltage at the output.

The thermistor itself is not linear and needs a variation of Stein-Harts equation to calculate the correct value given the current resistence (see here) which is completed in software. The thermopile will be used to read the temperatures of various objects placed in its FOV.

For the conversion from voltage to temp, I have found an article with explains the basics for calculating the voltage for a T- Type thermocouple which I'm not sure is correct but it gets me close to the value I want.

I've been playing about with the gain experimentally over the last few days and honed it into a value between 1000-2000. Haven't decided on a crisp number yet but i've tabled the values of resistance for this and will be choosing a whole number for ease of calculations later. And yes, very unwieldly! By addon voltage I am assumming you mean the voltage supplied (via 10k/1k divider network) to the negative input of the Op-Amp to pose as a virtual ground and thereby compensate for some of the noise of the system. This currently sits at 0.447 volts given a Vs of slightly less than 5V.

When measuring a sample temperature:
-I first measure the Vout of the Thermopile when no object is present in its FOV
-Place an object infront of the thermopile and give it a minute
-Subtract current Vout from previous Vout to provide the difference
-Subtract Vref (0.447V)
-Divide by the gain and scale to micro Volts
-I then use this value in the calculation provided by the earlier link having already pre-calculated the other variables needed.

Hope this makes sense. Going to be looking closer at the algorithm today and try not to pull my hair out!
Regards,
P-J-V

Hi,

This should help you. It consists of the Temperature resistance curves.

Temperature resistance curves: **broken link removed**

Hope this helps.
Tahmid.

Hi Tahmid,

Yes, the values listed on the sheet are for the Thermistor as I understand it (Please correct me if I am wrong). These values are currently being used to calculate the ambient temperature of the Thermopile housing and surrounding area. Its more the thermopile output that I am interested in.

Regards,
P-J-V

P-J-V said:

Hi all,

I am new to this forum and have a question relating to the output voltage of the **broken link removed** IR temperature sensor. I'm currently working on a small project which uses to sensor to read the temperature of an object which is placed in close proximity to the sensor (same distance every time). The circuit contains an Op-Amp and RC circuit to provide an amplified stable voltage output from the thermopile device and an independent reading of the thermistor for compensation. The problem I am having is converting the amplified thermopile voltage to the correct temperature. Can anyone advise me to the relationship between these two. Is the relationship purely linear? i.e Say the thermistor temp reads 25 deg/C and the Vout for the therm is 1.3V, will a change in Vout relate to a change in T linearly or is there more to it? I have trawled google for quite some time in search for an answer but am no further ahead at this point. Any help would be greatly appreciated.

Regards,

P-J-V

Click to expand...

IrThermo click 5V
https://www.mikroe.com/click/irthermo-5v/

IrThermo click 3.3V
https://www.mikroe.com/click/irthermo-3.3v/

It is calibrated in wide temperature ranges: -40 to 85°C for the ambient temperature and -70 to +380°C for the object temperature (-20 to 120°C in PWM mode).

You have circuit, sensor description, and examples in source code.

Watch video for demonstration.