Buffered Voltage Bias using Op Amp

I have tried introducing an Op Amp, unsucessfully into my breadboard following from Learn | OpenEnergyMonitor, however instead of using an LMV321 I have used a TLV2462.

In the following design using a follower circuit with the midpoint of the resistors being ~2.5v.

My assumptions is that the output to the analog input should be ~512, I’m seeing ~300. Is there something obvious I am missing?

The data sheet is:
https://www.ti.com/lit/ds/symlink/tlv2462.pdf?ts=1669155051613&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FTLV2462

You are right, the ADC should read around 512, and provided that you’ve connected the op.amp. correctly, the output voltage should be around 2.5 V.

My first immediate concern would be that you haven’t indicated what you’re doing with the second op.amp in the package. With no connections, it could be doing anything, so I’d suggest you connect it as a follower (exactly the same as the half you’re using, with the non-inverting input connected to the centre voltage, but leave the output connected only to the inverting input (not connected in parallel with the other – that is additionally connect pins 6 - 7 and 3 - 5).

The second concern is your use of a prototype board. The op.amp has a quite high gain-bandwidth product, and the prototype board with have a lot of stray capacitance everywhere. It’s quite possible that you have an oscillator - definitely something that’s not wanted. I’d suggest trying something like Fig. 48 on the data sheet. You’ll need to experiment with the value for CLOAD - I’d think in the range 0.1 µF to 1 µF should be adequate.

Thanks, the stray capacitance was causing the issue and you are right that without it was oscillating. It takes up to about 6 seconds with a 0.1µF capicator to settle at ~506, the addition of a resistor in series (regadless of size) seems to have little to no additional value. I will need to experiment with load/no load and other combinations to make a conclusion.

I had not intended to connect the additional op amp, but I agree with your reasoning. As a side, I noted the capiciator on the divider no longer seems to be necessary as add or removing it makes no visible change.

I would retain it - it should remove any supply ripple coming down the divider chain.