I was tinkering with a thermocouple and a 741 Op-Amp (yes I am aware that this is not the way to amplify a TC, and that there are much better solutions)

Anyhow, I tried a basic non-inverting amplifier circuit and heating up the thermocouple as far as I could tell, was not showing any increased voltage on the output. However, when I rearranged the circuit a bit the output shows a very large jump in voltage when the thermocouple is heated.

The problem is, I have not been able to determine what it is that I have wired up here. It is not an Inverting or Non-Inverting amplifier circuit, and I do not believe it is a differential amplifier either. If I am not seeing temp rise as increasing voltage, what is it that I am seeing?

The schematic is below, Mystery circuit on the left, Non-Inverting on the right:

I am not an expert but, it looks like your non-inverting amplified is not correct. I guess you are trying to do a single supply design and trying to use a virtual ground. And that is the reason for the 100K pair as a voltage divider. I think you are missing a resistor for gain control in that configuration. In such a configuration only an AC voltage (voltage varying with time) can be amplified. Please see this. Page # 6. Also see Page #4 that explains about a required capacitor to that the voltage divider is stable.

One more thing is about the Thermocouple's polarity. You need to use may be a milli volt meter to ensure that you are connecting it correctly. If you connect it in the reverse then you may not get the desired results.

I am not sure what your circuit on the left is. I feel that the thermocouple's resistance will be few ohms and you are basically shorting out the inputs of the Op amp in that configuration and there by feeding Vcc/2 to both pins of the op amp. When you heat the thermocouple the resistance of the thermocouple changes and so there is some voltage difference (voltage provided by your virtual ground) and that messes up the operation of the Op amp. All this is my guess.

Let us know how things improve. We are interested to know.

-Raja

Ok I think I have it sorted. I added an input resistor(R3) and put the thermocouple - side at virtual ground rather than absolute ground. This is much like my drawing above on the left actually. My multimeter appears to be reading about 101 times the input at the output (relative to the virtual ground, not absolute ground). The only points of the circuit that are at absolute ground are the voltage divider, and VCC- of the op-amp.