Higher than expected voltage in RC low pass filter

I attempted to check the myth about fixing the P0420 error code (Insufficient efficiency of catalytic converter [$3000 part from Toyota]) by inserting a low pass RC filter between an ECU and a downstream oxygen sensor.

The sensor is connected with 4 wires: two for the internal heater, ground, and signal. The sensor generates a signal between 0.1 and 0.9 volts depending on the level of oxygen remaining in the exhaust gasses after the catalytic converter. If the ECU sees wild swings in the signal, it lights up a check engine light.

In some states which care too much about the ecology, it creates unnecessary waste in the form of a perfectly good car being junked, in the rest of the world, the "Check Engine" light always on masks other, potentially important error codes.

I inserted 1MΩ into the break in the signal wire, and 1uF electrolytic capacitor between the signal and the ground wire with the "-" side towards the ground. The capacitor is inserted on the side of the resistor closer to the ECU. Now both, the ECU and the voltmeter between the signal on the ECU side and the ground show a constant 1.28 volts. It is higher than the expected range of 0.1v – 0.9v therefore trows a "Voltage too high error".

In my understanding from some EET classes taken more than a decade ago, I should not see a voltage higher than 0.9volts in any way I connect passive elements and a voltmeter. What I am missing?

^{simulate this circuit – Schematic created using CircuitLab}