(T) Failed emissions

Hi all,

I have a 99 bravo 1.2sx (16v) it has failed its MOT emmisions, I dont have the readout But know that the h/c and co readings were a massive fail (although Lambda was ok). When I initially took it the first time I was advised to get a new lambda sensor which I did (£32) when it failed again I was advised to fit a new CAT (£68) These items made no significant difference! I then fitted a new set of plugs and ran an emmisions fuel additive to the petrol, and was 'adventurous' in my driving style for about 60 miles, again it failed (as badly as before) Anybody got any ideas? if not:

Silver Bravo 1.2sx 16v
92k - very nice quiet and smooth engine
one or two car park dings
new lamda, cat and front pads
goodmans CD head unit
end of my teather £200

07709 470612.

Hi ,
First of all i want to know if you put your car on the examiner of fiat ?
this examiner will help you so much anyway , i want you to tell me about your car temperature in general like ( never become 90 degree - its cold all the time - doubt in the reading of the temperature ...etc).

If that was ok then you need to check your manifold absolute pressure sensor it could be it .

Keep in touch with me may be i can help you.

Im glad im not the only one to go through this pain ! Having done just the same as you (plus injectors) I got technical. Firstly the fault does not show up on diagnostics, but it appears the ECU is running "open loop", and ignoring the inputs from the lambda probe. It has to do this for the first 20 second or so of running untill the probes get hot enough to give an output, but should start to really accurately control the mixture when this lambda signal becomes present. Ths is the bit thats missing, (ECU wise- not probe). The ECU continues to run the fixed part of the fuel map, that is close enough so you would not know there was a problem, untill MOT time. I mimicked the lambda probe voltages into the ECU and monitored the fueling when I changed it, and it did not respond. Problem is a new ECU is £350 +, and has to be coded to your car, so you cant simply borrow one to try. I have also stripped the ECU (Im an electronics engineer) and dont believe this type of circuit can be repaired as such, and Ive thrown enough money at it already, so Im not forking out for an ECU on a £1000 car. Ebay £250 next stop- sorry.

Mark

I forgot to add, it passes all the Fiat main dealer tests !

Have you had the cambelt done recently?
Does the car run rough?

If the answer to both of these questions is yes, i'd suspect incorrect cam timing as this happened to my car.

ive just been through the exact same thing with my brava same engine and everything fiat couldnt solve the problem even tho they had my car for 3 days they wanted to strip the head and de-coke the valves and stuff i took the car to a back street garage and they solved it in 2 mins! turns out the new cat i had put on (bought from andrew page £112) was faulty you may have the same problem for some reason non standard parts are not good enough i got a proper cat and it passed!! other thing you could do is redex down your spark plugs easy to do and it does clean out the chambers!
hope that helps.

The CAT can only work properly if the exhaust gasses supplied are in close proximiity to the stoic ratio, so first you need to get the stoic correct.
Measure the dc voltage o/p from the lambda sensor, if its running rich then the voltage should be 0.8v proving the lambda is ok.
If its ok then a sensor reading is out of spec causing ecu to demand open loop.
No MAP on this model, but a MAF, so this is a possibility check that all connections have been put back on.
My guess its cambelt timing.

The reading for lambda is:-
0.8v = too rich
<0.2v = too lean
dithering between the two is correct operation.

high hc and co with a pass on lambda is unusual. high hc and co suggests overfueling, but the lambda reading should be low if that is the case.

i would do a compression test. this problem is often caused by a slight blow between 2 cylinders on the 1.2 engine.

This is the critical bit, mine was dithering, but at about 3-4 second intervals, and not consistently. (Use a DVM across the grey and black wires). This I attributed to the basic mapping being pretty close to around lambda 1, so sometimes it was just switching the lambda probe. I think it should switch around 1-2 times a second if the ECU is doing its thing to keep the catalyst functioning correctly . Not that I could verify this as I did not have a good signal reference. On my car I had already had 2 catalysts, the cam timing checked, new lambda probe, new down tube, rec'on injectors, injector cleaner, a tank full of Optimax, plus the fiat diagnostic so this really was last resort. As a matter of interest disconnecting my Lambda probe did not bring on the ECU warning light, so either the ECU is very stupid, or this was part of the fault.

The 2 heat sensors in the throttle body are a possibility (MAF). Also voltage sensor from lambda (in ecu). Think thats M1.5 hybrid ecu, so cant do anything with it.
You could try pulling airhose off and check for immediate response from lambda and injector pulse width.
Do likewise with propane from a blowtorch (unlit) squirted into air intake.
You could get the bits ecu/immo/key from a scrappy and try that.
Think I have a hybrid here which I could make into a kit if all else fails.