Fiat TwinAir misfire

Fiat Panda TwinAir 0.9L turbo (2013)

I was called in to look at this vehicle, after the garage had replaced both ignition coils and spark plugs, in response to a cylinder number 1 misfire condition.

This being a 2 cylinder engine, meant that it was not so easy to detect when the misfire was present, as it has a little lumpy feel to it at the best of times, whilst idling.

However, sometimes it was more lumpy than other times.  It always repeated cylinder 1 misfire.

I carried out a serial data analysis, especially looking at fuel trims and any misfire counters.  There were no misfire counters available to me and fuel trims looked ok.

The engine was definitely running with a partial misfire, but not enough to make it clear as to which cylinder was weak.

I did carry out a basic ignition profile inspection and could see no anomalies with the performance of either cylinder.  This engine uses COP coils with no access to primary ignition, so I used my COP probe to capture the secondary image.  (Sorry no image).

I then carried out a relative compression test, followed by a cylinder pressure test.  The cylinder pressure test was just a compression test and not a running test at this point.  I didn’t open the throttle for the test.

Using ‘Reference Waveforms’, in the scope software, I have included both cylinder pressure waveforms on one screen.

I didn’t see any problems with the cylinder pressures, with the noted exception of slight peak pressure differences, (which is not unusual) and so went on to perform running cylinder pressure tests.  That was likely to be a problem, as there are only two cylinders and this would be asking the engine to run on one.  But these tests did show up an issue.   I noticed that both cylinder pressure profiles were different, however I needed to do some homework first.  The patterns were a little odd and I had never sampled one of these engines before.

As you can see from the comparison, with cylinder 1 ‘dead’, the engine seemed to start relatively easily, but with cylinder 2 ‘dead’, it seemed more difficult to get going.  Suggesting that cylinder 2 was doing the most work.

At first look, there appears to be no problem with the cylinder running pressure.  Then we look at the cylinder pressure profile from cylinder 2.  Remember, the ECM complains of cylinder 1 misfire only.

Cylinder 2 profile indicates a particular condition of the intake valve operation.  Bearing in mind the ECM complaint, I had to assume this profile represented a good cylinder.  This made me take some additional captures, focusing on intake manifold pressure and the intake valve operation control solenoid.  This way I could monitor the engine running naturally.

I used the WPS500X coupled into the MAP sensor port, for the best possible resolution sample, along with the low current clamp and voltage probe to the intake valve operation solenoid.

I took a number of captures, so that I might collect as much data as possible, to study at my leisure.

You can see, in the intake manifold pressure waveform, there is a partial misfire occurring.  It is illustrated by the uneven nature of the intake pressure peaks.  I did sample the same set-up on the No.2 timing solenoid and ignition, but the misfire was barely noticeable during that test.

I then continued with tests, but replacing the ignition probe for cylinder pressure.

As you can see in these images, the cylinder pressure profiles are different, along with the timing solenoid profile.  Also you should note the difference in manifold pressure, depending on which cylinder is disabled.

When looking at the relationship between the timing solenoid operation and the cylinder profile, I had to make the decision about whether it was trying to correct the intake valve operation or was it driving it the way it wanted, due to some other information I might be missing.  ‘Cause or Effect’?

The more I read up on the system, the more I had to think about ’cause or effect’, as this system has the ability to control the intake valve opening duration, advance/retard and lift, per cylinder per stroke! wow! What a challenge!

So I thought about it and decided to rule out any other component fault, that might influence one cylinder.  I had already sampled the ignition system and found no problem and both coils and spark plugs had been replaced with genuine.

This only really left the fuel injectors or back to the valve train.

So I carried out a fuel injection flow analysis, using the WPS500X.

As you can see in the image, there is no problem with the fuel flow comparison.

So with that being the case, back to the original anomaly and the complicated nature of the valve train system, it seemed logical to me, the ‘TwinAir’ module was the problem.

I have to say, I was a little nervous of this diagnosis, as I kept thinking of possible things I might be missing, due to the nature of the system and the fact that I had never worked on this system before.

However a new module was then fitted and the result was a complete fix.  Luckily I did manage to get an after-fix shot of both cylinders.

You can see how both cylinder pressure profiles are the same and so is the intake manifold pressure.  Also the ignition timing and the timing solenoids profiles are all similar.

This action totally transformed the running.

This vehicle only had about 38,000 miles and had a full service history, with no obvious signs of bad engine oil.