Making a simple Thermocouple Amplifier using LM 358P

[cloudsky]

Hello experts,
I'm a student and I'm trying to study how an op-amp works.
I want to make an amplifier circuit using a thermocouple (TC) as the input (K-type, range about 0 - 50 mV). The output I expect is about several V. In the case of this LM358P that I'm using, I expect the output to be from 0 - 2 V (for a connection with Arduino to calculate the temperature later).
Here is the circuit diagram I found on the internet. I made the circuit as shown below.

I think the circuit works correctly because what I received in the output is ~ 6V and reducing (due to the feedback of the R2=100k to the 2- pin) when I used a 12V supplier to the Vcc+.
Here is the simulation that I also tried to understand the circuit.
The question is, how can I change or improve anything to receive the output (~ 0 ->1V) that also changes with the input (from TC ~0 - 50 mV)?

Thank you so much for your time and support.
If it's possible, I would like to send you a cup of coffee as my gratitude.

--- Updated Dec 23, 2023 ---

the simulation I forgot

 

[FvM]

You know that thermocouple voltage is DC, why are you using AC voltage in simulation?

--- Updated Dec 23, 2023 ---

1 V output for 50 mV input is gain of 20, no inverting amplifier gain is 1 + R4/R1, R4 has to be reduced respectively.

Do you actually use full K temperature range of 1250 °C?

LM358 has relative large offset voltage and offset drift, typical offset of 2 mV is corresponding to +/- 50 K temperature error.

 

[KlausST]

Hi,

I´m a designer for accurate measurement devices. Thus the use of LM358 for thermocouple is out of question.

As FvM says the offset and drift is just too high. But in detail it depends on your needs. We don´t know them.

Klaus

 

[betwixt]

In addition to the advice given, think about what actually makes the voltage across the thermocouple junction. It is produced where two different types of metal are bonded together. In your breadboard construction there are may places where different metals touch each other, including where the thermocouple wires are plugged in. Each of them also produces a voltage which will dramatically alter the voltage you want to see. I don't see a capacitor across the supply pins of the amp either, you are asking for trouble if you don't fit at least one.

Brian.

 

[KlausST]

Hi,

and where is your cold junction? don´t you use cold junction compensation?
There are ready to buy ICs. Or modules. For a good reason.

Klaus

 

[danadakk]

Some ref material :

https://www.ti.com/lit/an/sbaa274a/sbaa274a.pdf?ts=1703333145122

This is an ARM solution but basics apply to most micros :

https://www.infineon.com/dgdl/Infineon-AN75511_PSoC_3_PSoC_5LP_Temperature_Measurement_with_a_Thermocouple-ApplicationNotes-v06_00-EN.pdf?fileId=8ac78c8c7cdc391c017d0730fc5054bb

https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/14/Correlated-Double-Sampling-to-reduce-low-f-noise.pdf

Regards, Dana.

 

[cloudsky]

You know that thermocouple voltage is DC, why are you using AC voltage in simulation?

--- Updated Dec 23, 2023 ---

1 V output for 50 mV input is gain of 20, no inverting amplifier gain is 1 + R4/R1, R4 has to be reduced respectively.

Do you actually use full K temperature range of 1250 °C?

LM358 has relative large offset voltage and offset drift, typical offset of 2 mV is corresponding to +/- 50 K temperature error.

Thank you for your reply.

Because I want to simulate the changing of voltage due to the changing of temperature. That is the reason why I use an AC input.

 

[cloudsky]

Some ref material :

https://www.ti.com/lit/an/sbaa274a/sbaa274a.pdf?ts=1703333145122

This is an ARM solution but basics apply to most micros :

https://www.infineon.com/dgdl/Infineon-AN75511_PSoC_3_PSoC_5LP_Temperature_Measurement_with_a_Thermocouple-ApplicationNotes-v06_00-EN.pdf?fileId=8ac78c8c7cdc391c017d0730fc5054bb

https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/14/Correlated-Double-Sampling-to-reduce-low-f-noise.pdf

Regards, Dana.

Wow, thank you so much for your documents. I will read them carefully.

 

[cloudsky]

In addition to the advice given, think about what actually makes the voltage across the thermocouple junction. It is produced where two different types of metal are bonded together. In your breadboard construction there are may places where different metals touch each other, including where the thermocouple wires are plugged in. Each of them also produces a voltage which will dramatically alter the voltage you want to see. I don't see a capacitor across the supply pins of the amp either, you are asking for trouble if you don't fit at least one.

Brian.

Thank you for your advice.
I took it seriously to improve my circuit.

 

[cloudsky]

Hi,

and where is your cold junction? don´t you use cold junction compensation?
There are ready to buy ICs. Or modules. For a good reason.

Klaus

Yeah, I also thought about the cold junction compensation. But in the first trial, I just want to see the difference in output between room temperature and about 400 500 degrees Celcius (made by the lighter). If the result is reasonable, then I will try to improve the circuit for a more accurate result.