Hi all,
I have a 1998 Punto 60S 1.2 8V, done 103000 miles.
I've owned the car 7 years,change the oil/ filters as recommended,plugs were changed about 3 years back.
I took it for its MOT last week and failed on emissions.
The garage did a code read and it said a fault in the lambda sensor circuit. They did some basic tests and told me there is a voltage coming out of the black wire which goes to pin 29 on the ECU which means the sensor is ok,apparently. They have said to try doing a continuity test from the black lambda wire into the ECU to look for dry joints/ lack of continuity.
The numbers on the ECU are IAW 16F.E3/6A37-27.
I have looked on the internet and found companies that will test the ECU for £50 or if it needs repair then it's around £200!
I have seen ECU + keys on e bay for between £30- £60 but I am not too keen on that route just yet!
Any help/advice please?

Before going the ECU way you should check really how it failed the emissions. The emission report data will be a good information to share here. My car same engine been passing emissions very marginally for the last couple of year with CO levels at 0.30%. If your emissions are well off the limit then yes a fault in the lamda circuitry is possible. If emissions failure is marginal then circuitry would be fine but the lamda or the CAT are becoming ineffective. Also when marginal the cold weather will not help as the cat is not heated to optimal temperature and fuel combustion is not optimal either. It may be worth replacing those spark plugs and do some spirited driving before you do a retest.

Puntonaka said:

Hi all,
I have a 1998 Punto 60S 1.2 8V, done 103000 miles.
I've owned the car 7 years,change the oil/ filters as recommended,plugs were changed about 3 years back.
I took it for its MOT last week and failed on emissions.
The garage did a code read and it said a fault in the lambda sensor circuit. They did some basic tests and told me there is a voltage coming out of the black wire which goes to pin 29 on the ECU which means the sensor is ok,apparently. They have said to try doing a continuity test from the black lambda wire into the ECU to look for dry joints/ lack of continuity.
The numbers on the ECU are IAW 16F.E3/6A37-27.
I have looked on the internet and found companies that will test the ECU for £50 or if it needs repair then it's around £200!
I have seen ECU + keys on e bay for between £30- £60 but I am not too keen on that route just yet!
Any help/advice please?

Click to expand...

A quick way, but not fully perfect way, to test the lambda is here:


The reason why I say is not perfect is because the lambda sensor produces a sort of square wave and the multimeter will sample the signal, giving just a rough idea. Still it worked for me to find out my lambda sensor was ok.

did they make sure the cat was hot enough to do an emission test?

:yeahthat:

Should let the cooling fan cycle a couple of times before doing emissions test. Fresh oil, filters and new plugs often get it through.

This isn't to rule out possible cat/lambda issues, but worth a go.

Hi
Thanks for the replies
Here is what the emissions sheet has on it:
CO Max limit 0.3% ACTUAL= 12.684%
HC Max limit 200 ACTUAL= 697
Lambda 1.030 ACTUAL= 0.646
Coolant pipes hot

Hope this helps?
I'm guessing because the CO is so far out, the ECU is suspect?
The garage did say because of the high CO it might have damaged the cat too.

Mk1 ECUs are ridiculously reliable. It'd be the very last thing I'd suspect.

Coolant pipes hot dont mean the cat is, did they try reving it high for about 5 minutes? that often works

The emission look way off so the problem looks like is related to the lamda or oxygen sensor. I reacon even disconnecting the lamda sensor the emissions should not be that bad so worth a try and redo the emssions. If the emission appear better then I would first swap the sensor and retest. Also check the cat if is blocked. Easy test will be to see if the car still has power while driving redline in third will mean the cat is not blocked. (non block does not warranty cat is effective though).

I have sent the ECU off to be tested and repaired if needed so will update when I get the unit back.

I have just spoken to the company that tested the ECU and they could find no fault, which is what you guys had suggested.
The people testing the unit said it could be the MAP sensor or temp sensor. The garage that failed the car did check the output from the Lambda and temp sensors.
Is it worth changing the MAP sensor?

Cheap and easy to swap. Yes.

Do you have a Fiat specialist near you?

If so it may save you a lot of expense as they are familiar with the common problems and symptoms of the car.

My first ports of call would be to check that the fuel mixture side of things is in good condition. Spark plugs, map sensor + piping, coolant temperature sender etc...

It is also not out of the equation that the CAT is knackered.

Puntonaka said:

Hi
Thanks for the replies
Here is what the emissions sheet has on it:
CO Max limit 0.3% ACTUAL= 12.684%
HC Max limit 200 ACTUAL= 697
Lambda 1.030 ACTUAL= 0.646
Coolant pipes hot

Click to expand...

Not sure how the garage checked lamda sensor as the figures above show Lambda is well off. Value should be around 1. I always am puzzled how they measure Lambda at emission time. Nevertheless also CO is well to much. A car not equipped with CAT will fail emission if the CO is above 3.5% so clearly there is something wrong with your car. Also there is some doubt how the garage measured the temperature sensor. So in short I will check those two sensor again and not assume they are OK.

When they check lambda at MOT they're not actually testing the sensor.

^^ that. they are just measuring the lambda value with the emissions tester at the garage - has nothing to do with the signal from the sensor on the car, it the actual value at the tailpipe.

How the garage checked it will just be they have checked that it is sending a signal to the ecu, i'm not aware of a way they can actually test its functioning properly. A narrowband as fitted to nearly all road cars will just oscillate from 0 - 5v, a meter can't tell you the thing is working or not, just that there is power and a signal. Although thinking on, if the ecu is reporting a circuit fault perhaps we're just looking at a dodgy earth??

fingers99 said:

When they check lambda at MOT they're not actually testing the sensor.

Click to expand...

Yes, this is why I am puzzled. However, I suppose what ever they measure must be related to the lambda sensor effectiveness? And if this trues the sensor is not working as it should according to the MOT figures he is got.

debatable, there is more than just the lambda that will effect the emissions, heck just a tiny hole in the exhaust in the wrong place can result in really bad figures..

But if the ecu reports a fault then there must be a reason - either bad earth or supply, a slight break in a wire or dodgy connector resulting in inconsistent signal or a sensor fault meaning it just sending the wrong signal all together. What the garage did is stated in the first post and that is checked there is voltage on the signal wire - so we know it sends a signal of sorts. Checking the continuity of all the wires would eliminate wiring but it would be quicker and easier to check the earths, clean them up and put them back. Next i would change the sensor itself as they are not expensive. Trying to find a slight defect in a wire could take ages as if its just kinked somewhere the fault might not show unless you giggle the wire in just the right place in the right way & i'd prefer eliminating the other possibilities first myself. I hate tracing wiring faults! can you tell?

Thought I would update you with the emissions saga!
I changed the plugs,plug leads, air filter, lambda sensor and temperature sensor.
Last night I was going to change the MAP sensor but noticed there was a split in the pipe from the MAP sensor to the inlet manifold, where it connects to the inlet manifold. Luckly I had the right size and length of pipe in my shed.
Today I took the car for its retest and it passed!!
Emissions readings are:
Fast idle test
CO 0.065
HC 43
Lambda 1.001

Natural idle:
CO 0.019

Got there eventually, thanks to the help given on here.:worship:

The emission figure look perfect now.
Puzzled again now as I don't seem to have a feel where my map sensor is on my mk1 1.2 8V engine. Perhaps next time I have the plastic cover off I have a better look to see if I can spot that map sensor. Certainly a split pipe will cause some issues if that map sensor relies on measuring vacuum in the inlet. As the pipe was split the sensor will see atmospheric pressure and then fuelling may be affected badly which may be well the reason you failed the emission initially.