To follow up the guy who owns and uses a PicoScope a member of this forum is called Terry and I would guess based on Steve's location live in the Worcester area. With luck he'll post his view on this issue.
Today's post by Steve is as follows:
OK, this is not going to be easy to explain clearly.
I have already etsablished that when viewing real time 02 data
(voltage and STFT) the STFT does NOT represent an output from the
02's as you would assume. It is an aspration, a requirement or
expectation of the ECU, not an 02 reading at all. So, STFT info has
to be ruled out as evidence that the 02's are working.
I've run the engine up from 13C coolant temp this morninng and
watched the 02 info. They are both flatlined at 0.000V and +24.2%
STFT.
When the 123 failure occurs around 80C coolant, an engine restart
will either see the engine run on 456 only, or sometimes on all 6,
it's a bit of a lottery. When running on all 6, I may see 02
voltages as I'd expect, 3.2V is one I rememeber seeing, and
voltages are shown for both B1 and B2 02's for a time.....along
with STFT's you'd expect.
Until cyls 456 drop out that is.
When 123 fail once more, both 02's flatline at 0.000V and the STFT
shoots up to 24.2%.
I'm beginning to think the 5V supply to both 02's is being cut
somehow. I doubt the 02 heater circuit is at fault as I'm revving
the engine to high exhaust temps, so the 02's are not going cold
due to the lack of their heaters.
In both cases, I'm going to have to check for the presence of both
the 5V 02 sensor supply, and the heater circuits, although the
latter I'm sure I did months ago.
Note;
Having seen proper, varying voltage outputs from both 02 sensors on
occasions, this probably provides proof that both 02's are good,
viable. It would be easy to think they are both useless when they
are aparently flatlining, but that is effect, not cause. Something
is causing both of the 02's to drop to 0.000V when they should be
bouncing up and down the voltage range. That might be the source of
all my trouble with this car.
Here is a theory going on a bomb that was dropped early in this discussion. The vehicle had been gas converted and now reversed? This really does move the goal posts. From a pure mechanical view, how many miles have been covered using LPG? My experience with LPG results in rapid valve seat wear and closed valve clearances. Vehicle will run fine from cold then lumpy as the heat builds, then misfire. Misfire detection will then shutdown cylinders, fuel and ignition. Excessive air through the catalyst will then rapidly increase its temp. If these cam followers are not hydraulic, can we check the clearances? Could we also establish the order of events? Do we lose spark, fuel injection, then gain a high fuel trim or vice versa? Sorry if this has already been asked. LPG switches over from petrol to gas at a predetermined temperature normally during the warm up. Does the vehicle still have the LPG installed even though it has been reversed? The vehicle fuel injection process has to be simulated when running on gas to allow the PCM to continue managing the engine, and so the LPG controller interrupts the fuel injector switched signal to halt injection, but still simulates this process as seen by the PCM Can you confirm if this is still installed on your vehicle? Sorry if this is very long winded. Take care. Steve Smith
If you've got 'known good' cars to compare with, why not take an injector waveform from each bank, on each car. If your problem car has a much longer injector dwell it would tend to back up the STFT data and low lambda's.
Is 40 seconds long enough for a lambda sensor to warm up? Check it against a known good engine and see how long it takes to get a reliable reading from lambda sensors.