Hi. This Kia i have been told failed its mot this time last year on the emissions The mil had been on for a few weeks before the test. The garage replaced the front sensor cleared the codes and the car then passed. However the light came back on
after two days. They then replaced it again twice the first two replacements were aftermarket but then both front and rear were replaced with gen kia.
but they still kept getting the light two days after each replacement.
They then told the customer they could do no more.
the customer now had an mot so ignored the light all
year until the next mot.

It has now of course failed again at a different garage. and has now been given to me to check.
I don't know what the codes were last year
but likely the same. P0133 b1s1 slow response.
The fail is on co at fast idle at 0.5
Lambda is a pass. although there is a slight leak in the back box that can give a lambda fail on my gas bench but rag up the pipe with the probe proves the leak is only at the back box.

The fail on co at only 0.5 and a pass on lambda
would to me point to a failed cat. But for the lambda sensor code.
this car has a manifold cat and an under floor cat
1500 quid for genuine ones.

Now what i have done so far.

The car drives very well. It is in closed loop
and the fuel trim is short term and long term
always plus or minus 2-3. this is the case at idle
and cruse. also checked on the motorway.
front sensor switching on graphing scan tool.
but rear sensor switching also not as much as front.
In EOBD mode 5 long before the mil comes back on
the weak/rich rise and fall times are a pass
but the time it lingers rich or weak is a fail,
limit is .6ms. The sensor is taking .9ms.

On the pico, front and rear sensors, rear is switching at almost same rate as front.
The front is hanging high/rich three times longer than low/weak. The sensors are both basic 4 wire.
I have confirmed the sensors are not pluged in back to front. The front sensor heater is getting a feed and switched neg and has the same resistance as the rear. I have been testing for
leaks in intake and exhaust. None found
also tried unplugging maf and it did not make any difference.

My main question is if the fuel trim is spot on
why would the sensor hang rich longer than weak.
i understand it being in closed loop if there are enough cross counts but if its at 0.9v why would the ecu take longer to say go weak than it does
to say go rich. Also i understand as the sensor heats up it should speed up.
and at higher rpm it should increase in frequency
however this appears to not speed up and may even be slower.

Also the freeze frame when the code sets shows
closed loop, fuel trim short and long good and engine speed 2800 rpm, 80deg c

Sorry for the long post.
Any advice would be great thanks
Alan