There seem to have been a few threads on this topic over the years but this has been dogging our Twinair Panda ever since its first MOT three years ago.

The lambda figure in the MOT has always been borderline on the high side of the limits and sometimes the tester has had to work really hard to acheive a pass. I've had the same guy's signature on my MOT test certificates now for well over 20 years so I know he's doing a trustworthy job. A few months ago I had cause to replace the pre cat Lambda sensor due to a misfire when cold so I hoped that the problem at MOT would now be solved.

..unless there's a specual nuance with the Twinair engine?

This year it's failed on the same reading! with as usual the HC's and CO indicating that actual emissions wise, the engine is running perfectly clean ...so no need to try the garage's £20 fuel treatment then thankyou! ..Or taking it out for a thrash wouldn't really improve matters would it?

Here's the readout..


Now, the MOT's Lambda figure is derived internally in the testing machine from measuring the ratio of oxygen in the exhaust gases. It's nothing to do with a figure from the Lambda sensors.

The high figure indicates an excess of Oxygen. This could possibly suggest a misfire of some sort but that would be noticeable and associated with a high HC figure which there is not. (spark plugs have also been replaced with Iridium types in the last 3 months too)

A very highly possible cause reported on other threads and boards is that there is a leak in the exhaust pipe downstream of the catalytic converter that's letting in air and hence oxygen. I can find no such leak. However, there are two drain holes, one in each end of the silencer that could be the only possible sources.

How could I best temporarily block up these holes for the re-test on Monday and is it indeed worth doing? Is there anything else I could look for? Multiecuscan shows no faults.

If the exhaust isn't leaking I would be tempted to get the emissions tested elsewhere.

My TA tested ok, but the exhaust on the 4x4 if different than the standard car, I wonder if this might be contributing to the issue?

Hi.
High O2 is essentially a weak mixture. The fuel pressure could be low enough for the ECU not to compensate, without the ECM illuminating. I'd do a pressure test on the fuel system initially then look for air leaks or vise versa.
I've been reading up on the TA recently and its quite possible the uniair unit may be faulty too.

I'm foxed by the precat O2 sensor showing up as faulty on a cold engine as it will be running open loop till the engine is at a predetermined temperature, so in other words is not used till its warm.
TBH its all pointing to low fuel pressure.

I think the drain holes are a red herring; they're there by design, and located in a part of the system where you'd always expect there to be some back pressure. I doubt that the test is being compromised by air being sucked in through those holes. Besides, they're fitted to all the cars and this isn't an issue that's commonly reported here.

I'm thinking this will be (a) somthing that's specific to your car (since "they don't all do this, sir") and (b) something that's always been there to some degree (from previous MOT issues and you've said the fault has persisted through a lambda sensor change).

So I'm wondering if this could be due to a combustion imbalance between the cylinders, perhaps an incorrectly coded or out-of-specification injector? I doubt something like this would be easy to diagnose.

There are limits to the performance of closed loop fuel management systems in multicylinder engines with only a single upstream lambda sensor and it is possible that, in the future, we will see designs with multiple manifold sensors to permit the fine tuning of the fuel trims between cylinders. There's a paper on the subject here; some of it is rather technical, but the conclusions are worth a read.

So far my 2012 punto has had nice low figures.
Similar mileage..
but quite possibly a very different look to the exhaust.. mine looks very 'royal enfield..'

My MOT is on Weds so time will tell

Not had ANY running issues from day 1..

Just oil and plugs at frequent intervals.

Charlie . Cardiff - Oxford

My Panda Twin Air Lounge MOT Lambda results for 2015, 2016 and 2017 were 1.000, 1.007 and 1.004, so I think you can rule out the exhaust, or that they are all close to the limit.

Hi.
On my last comment I failed to mention that I had the same issue on a 1997 Skoda Felicia on its first MOT. The MOT garage was the Skoda dealer and got stuck with the issue, the car was with them for days and eventually they more or less gave up. On the Skoda there was a simple fuel rail pressure valve using a disc and spring type of regulator. I deduced that it had to be pressure as the garage had replaced the throttle assembly, inlet manifold gasket and other logical tests. They didn't though ckeck fuel pressure. It had a Schrader connection and using a tyre pressure gauge it was 1 bar down, didn't know how accurate it was but I then opened up the regulator to find a mangled spring. A new one borrowed from a stock vehicle cured the issue.
No idea how the fuel pressure is regulated on the TA but I'd definitely try to get it tested.

Thanks all for your suggestions and experience. I have found one more exhaust joint to check (behind the torsion beam) that I hadn't spotted before and will have a good look at all parameters on Multiecuscan.

A note that the car runs really well, always has done, no flat spots or hesitation and no signs of overheat on the catalyst (a sign of lean mixture apparently). The Stop-Start stopped working about 4 years ago but I put that down to the battery. It does , however, do predominately short journeys of under 5 miles but then it is a 'Town Car' isn't it!?

If the retest comes up similarly poor on the Lambda I think a visit to a Bosch service centre is in order. Staying away from Fiat dealers being a priority! :yuck:

Ok, Retest time...



So, scraped through, skin of its teeth. Only thing I did was take it for some long runs with some fuel system cleaner in the tank. (managed 63MPG) and a longer drive before the test to warm it up maybe a bit more thoroughly (not easy in this weather) which clearly made no difference for the first run.

Something clearly still isn't 'right' but at least I've now got another year to investigate further. I'm wondering if moving the MOT by taking it in for one in the summer might help

Here's some sensor figures from fiatEcuscan.


Problem with this is is the Battery voltage and Lambda 1 Voltage readings are bugs in that poorly supported piece of software. (Give me VCDS on the Audi any day!)

Correct me if I'm wrong but my interpretation of this is:-

• The Lambda sensor 1 Integrator hovers around 1 (roughly between 0.98 and 1.03 showing it is correctly reacting to changes made by the ECU.
• The Lambda Sensor 2 Integrator stays at 1.00 indicating that the catalyst is functioning correctly.

Perhaps I need to look at some temperature sensors maybe?

Firstly, I'm glad it's passed. That keeps you mobile and kicks the can down the road for another year.

But something isn't quite right; we're not seeing a flood of reports of emissions failures at MOT time for this engine.

I'm still thinking this is consistent with some sort of combustion imbalance between the cylinders.

You could be right jrkitching ..and I reckon it's likely an issue that's been on this vehicle from new!:bang:

I've read elsewhere that the injectors come with a calibration figure printed on them that has to be programmed into the ECU? Maybe something to check that this is correct?

..But i'll wait till it warms up a bit!

Vic20 said:

You could be right jrkitching ..and I reckon it's likely an issue that's been on this vehicle from new!:bang:

I've read elsewhere that the injectors come with a calibration figure printed on them that has to be programmed into the ECU? Maybe something to check that this is correct?

Click to expand...

:yeahthat:

Both those points were behind my thinking in post #4 . Of course, even if you manage to find the calibration data for each of the injectors, I can't see any easy way of verifying that it's actually correct.

Have a read of that paper I linked to in my previous post; it just may give you some food for thought during the cold winter evenings.

My twinair failed an mot on emissions.... My mate then retested it and it passed. He mentioned at the time that they are harder to pass as they have to test them in a different way to a standard car.... I'll ask for more info. My mot is due soon and I'm not confident lol

quotethepigeon said:

My twinair failed an mot on emissions.... My mate then retested it and it passed. He mentioned at the time that they are harder to pass as they have to test them in a different way to a standard car.... I'll ask for more info. My mot is due soon and I'm not confident lol

Click to expand...

Please keep us informed - this could be interesting if symptomatic of a wider issue.

A lot of TA's are going to be coming up for their first MOT about now .

Off the top of my head I remember he said the car has to be at the exact right temperature...

quotethepigeon said:

Off the top of my head I remember he said the car has to be at the exact right temperature...

Click to expand...

Not too far fetched..
Considering the service parametrr presets also include temperature info.

My TA is not quite as good now a non.fiat specialist is doing the work...

Makes you wonder ..

quotethepigeon said:

Off the top of my head I remember he said the car has to be at the exact right temperature...

Click to expand...

It depends on the year of manufacture; also the 85HP and 105HP TA's are different.

All the details are in this document.

Early cars seem to be less critical with regard to oil temperature and rpm limits; note the much tighter parameters for the later cars in the detailed tables.

This leads me to wonder if some cars are being unfairly failed because they are not being tested properly.

I wonder if this might be due to the way the valves operate?

The Multiair system alters the valve timing at low engine speeds.
Basically, the cam is "hot" for high loads, but the fluid links back it off for lower power/speeds.

It also times the valves to negate pumping losses, so it must at times cause air to pass through the combustion chambers and into the exhaust without being fired and used up.

Now if the fast idle test isn't quite fast enough (or the engine load not high enough) for the Multiair unit to stop this action, that could be causing excess oxygen to be detected in the exhaust gas, it's naturally passing the stuff through the chambers!

Hi.
I've been watching this thread after making a couple of comments.
Most likely the differing technology of this engine is where I perhaps tripped up a bit on my thoughts. Goudrons suggestion makes sense, I believe the TA engine is very low friction and at fast idle would only require very little fuel, could the mapping of the ECU be on the lean side with this type of loading? Would turning on the headlights, HRW, Air con etc be enough to put some load on the engine and perhaps bring the Lambda volts down?
Watching the thread with interest.

murphyv310 said:

Would turning on the headlights, HRW, Air con etc be enough to put some load on the engine and perhaps bring the Lambda volts down?
Watching the thread with interest.

Click to expand...

I think you may have confused Lambda is this case.

Lambda is a Greek letter and is being used to indicate a calculation.
In the emissions test this calculation is made by the exhaust analyser.

It measures the exhaust gases for oxygen, calculates it against Stoichiometric. (perfect fueling of 14.7 parts air to 1 part fuel).

So say the engine is running lean at 16.1.
Divide 16 by 14.7 = lambda reading of 1.08

Running rich at 13.1.
Divide 13 by 14.7 = 0.88

Confusingly, the car it's self has Lambda (calculating) sensors, we also know them as O2 or Hego sensors.
They self generate a small signal voltage based on heat of the exhaust gases, the hotter the gas, the leaner the engine, the higher the voltage.

These sensors aren't what's actually tested by the analyser, more like their results.