Car has been stood over the (prolonged) winter. Got it started and idle was lumpy. Let it warm up, revved it a few times, fault seemed to clear. Stuck it in SPORT and gave it some beans on the way to MoT.

No faults found except CO emissions 3.5 percent on fast and slow idle tests. He also said they were okay initially, but started rising.

It was presented and tested with the PRINS LPG engaged and fuel level showing 3-4 lights.

The car has a conical air filter, which had disconnected from the inlet. Timescale unknown. It has passed successfully with that mod.

I've Googled a few probable causes (a couple with exact match to circumstances) but prefer to see if anyone here confirms them.

I have a Nanocom, but no idea what to look for in live readings.

Car is SORNed so I'm very limited re: test driving.

Use the Nanocom to check the O2 sensor outputs. Prins is a slave system so follows the fuelling for the petrol system. Any fault on the petrol system will also show on LPG.

Got it through.

On a hunch, I wondered if the conical air filter had been leaking for a while and thrown the long and short range trim out.

I reseated the adaptor ring, tightened the knock-off Jubilee for the conical air filter as much as I dare and clipped it back on.

Used my Nanocom to RESET ADAPTIVE VARIABLES.

Stuck it in SPORT, used D3 and D2 and ragged the nuts off it on the way to the test centre. It has a rev limiter!

Resubmitted for emissions on petrol just to eliminate the Prins as the problem.

PASSED.

Good news!

What were your CO and lambda figures when it passed?

The filter is before the MAF, a leak there really shouldnt affect anything? the MAF still measures all the airflow?

Keep an eye on your long term trims and see if anything creeps back in.

What you now need to figure out is wether its fixed, or wether the issue was somewhere else, like within the LPG system, or a bad sensor causing the fuel trims to drift.