Problem with signal conditioning IC

[m.h.s]

 I am designing the signal conditioning circuit of thermocuple

There is a problem with the IC used in the signal conditioning  circuit. In the circuit there are three

stages 1) Buffer  2) Low pass filter 3) Amplifier. Even though the IC (OPA277P) is a high precision amplifier having very low input offset voltage, but it is adding noises internally. When a signal of 1.1 mV  is applied to the buffer from the thermocouple , the output of the buffer is not 1.1mV . This IC is adding internal noises and giving the 400mV output. I dont want to use any cold junction IC.

Any help appreciated.

[Henrik_Volkers]

Welcome in the analog world 😉

Dealing with TC EMF down to µV to build a TC amplifier is challenging if you are not used to it.

AD,National Semiconductor and Linear have some nice app notes on how to work properly with OP amps *. 400mV noise sounds like a very poor design.

However, you will always have a cold junction with TCs... how do you want to handle it?

 *) OP amps for everyone (nat?), or the linear circuit collections . Look for articels written by Jim Williams.

[Buechsenschuetz]

400mV is a lot of noise. You should check with an oscilloscope whether the noise occurs at a certain frequency, or whether certain frequency patterns show up. If so, you are probably fighting with an oscillating amplifier. Some opamps tend to oscillate when used as a buffer (i.e. with a gain of 1), not all opams are "unity-gain stable". Sometimes adding a pull-down resistor from the output to GND helps since the output impedance of opamps usually rises with frequency. Also, always take care of proper grounding, all GND points around the opamps should be tied together as close as possible to the opamp. On a pc board, using a ground plane usually is the best solution. Do not forget decoupling capacitors (properly grounded, of course) on the supply lines. The smallest capacitors should be the closest ones to the supply pins.