THERMOCOUPLE IC

Dear Aied - 

I think you also need a PGA in front of the ADC - here TI makes devices like the ADS1118 which you can evaluate

http://www.ti.com/lit/ds/symlink/ads1118.pdf

This part has PGA and ADC built in. 

Here is a reference design for K type http://www.ti.com/tool/TIPD109

There are around 17 parts with I2C to look at, so you can start here: 

http://www.ti.com/data-converters/adc-circuit/precision-adcs/products.html#p158=I2C&p776=Differential&p2192=PGA&o4=ACTIVE

and add more requirements on the left of the page to get to the part you need.

after that, you can use MSP430G2xx and above (for the MCU) 

Aied,

Along with what Josh mentioned, I wanted to also point you to some other application notes that are available for thermocouple measurement.

First there is a Basic Guide to Thermocouple Measurement. It will have information on making thermocouple measurements, starting with how they work, a survey of different types of thermocouples, and how you use ADCs to measure them. You can find it at the link below:

http://www.ti.com/lit/an/sbaa274/sbaa274.pdf

The reference design that Josh linked to in the last post is also the EVM for the ADS1118. You can find it here:

http://www.ti.com/tool/ADS1118EVM

It's also written up as a cookbook circuit from the Analog Engineer's Circuit series and as another application note showing some calculations. You can find them here:

http://www.ti.com/lit/an/sbaa354/sbaa354.pdf

http://www.ti.com/lit/an/sbaa189/sbaa189.pdf

I would note that the cold-junction compensation is important to the measurement. If you are using the ADS1118 with the internal temperature sensor, the device needs to be mounted so that it is at the same temperature as both of the thermocouple leads (at the connector). If you look at the EVM, the device is mounted between the connectors for the two leads. I would note that the ADS1220 is also suitable for this application and has better resolution, but the internal temperature sensor is not as accurate. You can find a cookbook circuit for that here:

http://www.ti.com/lit/an/sbaa355/sbaa355.pdf

Joseph Wu

Dear Aied - 

In looking at a typical rating chart - I think you could choose from Types J (-210 to 760C) , K (–270 to 1260C), T (-270 to 370C), N (-270 to 1,260C)  or E (-270 to 870C)

https://www.thermocoupleinfo.com/type-e-thermocouple.htm

Aied,

Any thermocouple measurement requires an accurate measurement of the thermocouple voltage, a cold-junction that keeps the thermocouple leads at exactly the same temperature, and an accurate measurement of the cold-junction temperature. The methods of measurement should be independent of the type of thermocouple. So much of the change between thermocouple types can be accounted for in software. I don't know of any special changes you need to make in the schematic for cryonic temperatures.

To measure the cold-junction temperature, you first must make sure the leads of the thermocouple are at the exact same temperature. If you look at the ADS1118EVM, the GUI uses the device's internal temperature sensor and the device is placed directly between the thermocouple connections.

The cold-junction measurement can be done in several ways. You could use an ADC with an accurate internal temp sensor like the ADS1118. As Josh mentioned, you can use an RTD, which can give very accurate temperature measurements. Similarly, you could use a calibrated thermistor or a TMP61 for the measurement. One other good solution would be to use a standalone temperature sensor like the LMT70. Regardless of what you use, you need to get an accurate measurement of the cold-junction and the sensor needs to be in or on the cold-junction.

Joseph Wu