This might take a while but....!

Failed MOT on emissions CAT is clearly not working correctly BUT!

Customer has had new CAT (But no new lambda sensor) for each MOT for the last three years. (And a new one now would likly squeek it through) The CAT failure is the symptom not the cause I suspect.

Lambda switching was slow and when warm fixed at 0.45 volts at idle. At speed lambda does switch but not very well and only switches between 0.45 and 0.8v.

It only switches down to 0.2 at the point that the throttle is snapped open from idle, after that it resumes switching between 0.45 and 0.8

There is a "Live Data" line that says (something like) "O2 sensor lean signal 0.45v" (Unfortunatly I didnt write it down precisly)

The earths at the lambda sensor appear good (The switched earth for the heater bottoms out at about 0.5V but one would expect that for a semiconductor switch)

there is no apparant false signal on the ECU side of the lambda sensor with the sensor disconnected.

Thinking that, On a 200,000 mile car, the lamda sensor was still the one it came out of the factory with and that the misbehaviour might be consistamnt with a combination of age and possibly an internal voltage leakage I replaced the sensor.

Its still the same! (But then I did warn the owner that this was as much a "diagnostic" replacement as a fix one)

He is understandably reluctant to spend too much on the car, but at the same time he really likes it and does not really want to replace it. Appart from the MOT fail this car really does work really well!

Oh, another bit of data, the fuel trims are both essentially unmodified, both long term asnd short term.

Any suggestions?

Paul