Hi There all you guys! I'm a new member just fresh as milk but I have been following your forum for quite some months now.

I'm from Malta and have a 1.8 GT bravo 96'. Had it since new and i love the damn machine!! Will bullet point the history of the progress of my findings after I explain my problem.

Car runs perfectly when cold (automatic Choke).
Car goes from 500 to 1500 RPM when idling for some time at normalrunning temperature.
Cleaned MAF sensor and then replaced with a used one. Problem still there.
Cleaned idle stabilizer valve and then replaced with a used one. Problem still there.
Replaced plugs (wet with rich mixture) with new ones. Problem still there.
Cleaned lambda sensor and read further into the Zirconia Lambda sensor.

0v lean - 0.5v good mixture (14.7:1) - 0.9v rich : So I stuck the mutimeter onto the dash and a cable running to the sensor wires. Sensor heater fine too.

So the machine runs fine when warm: pressing the accellerator on gear 0.8v, releasing the accellerator on gear (engine brake) ecu cuts the fuel altogether and reads 0v.

Return soon after to idling 0.5v. (perfect) Till you leave it on idle for some seconds, say 5 to 10 seconds and voltage starts to increase slowly to 0.9v Richening the mixture like hell.

I can also make a cicuit for this with a bar graph of leds so I can observe better but I can handle the multimeter.

Sh#t I DO NEED HELP!

Love you all

when the lambda sensor's probe voltage rises after a few seconds of idle it means there is overfueling occuring at idle. the ecu is obviously not adjusting the fueling even though the ecu is receiving the lambda's signal telling it that overfueling is occuring. this immediately tells us that the problem is the ecu is unable to adjust the mixture at idle.

so next you need to consider what the ecu uses to control the mixture at idle. there must be a fault in one or more of these parts, preventing the ecu from controlling the mixture.

there are 2 ways to look at it, fueling and ignition.

ignition advance will depend on the coolant temp sensor working correctly, and the rpm sensor. these are possible faults. if ignition advance was not occuring correctly some fuel would not get burnt, causing overfueling symptoms even though the fueling is correct. you need to check the rpm sensor and coolant temp sensor.

fueling will depend on the lambda and MAF. we know these are working fine.

Thanks for the immediate invalueable feedback Jug, hoped YOU would have replied.

Removed the rpm sensor some weeks ago, had some oil on it due to a slight leak in the crank oil seal i'll soon replace. Cleaned it and regapped it and problem was still there. its not meant to have that much influence on the sensor?? but this sensor is used for ignition advance and ignition timing and would be a donkey on the road on gears! Am i correct?

Will check coolant temp sensor tomorrow. does this cause such a drift in ignition timing and ignition advnace if bust? is ignition timing temperature related? and what controls ignition advance independently of ignition timing which is fine a high rpm if we can say that this is happening?

Can't use a timing light as cois mount directly onto plugs without HT leads but will think of somethig to work this out.

Cheers#

J

Or else possible the coolant temp sensor could be reading cold even when warm and ecu runs on a rich state as the traditional "choke"?

jfaloi said:

Removed the rpm sensor some weeks ago, had some oil on it due to a slight leak in the crank oil seal i'll soon replace. Cleaned it and regapped it and problem was still there. its not meant to have that much influence on the sensor?? but this sensor is used for ignition advance and ignition timing and would be a donkey on the road on gears! Am i correct?

Click to expand...

yes, you would also expect there to be problems during acceleration (lack of ignition advance leading to overfueling).
do a test on the rpm sensor. it should have a resistance of 513-627 ohms.

jfaloi said:

Or else possible the coolant temp sensor could be reading cold even when warm and ecu runs on a rich state as the traditional "choke"?

Click to expand...

thats exactly what i'm thinking the problem is

Tested rpm sensor. it has in fact a resistance of 592 ohms.

The coolant temp sensor is reading very accurately from 20 to 100 degrees from 2000 ohms down to 300 ohms when boiling water. These are perfect figures found on other sites.

There is a ignition controlloer unit on the Right hand side of the cyl head are they prone to faults? Are ECUs prone to be faulty?

Thanks

the "ignition controller unit" you mention i think is the ignition power module. it prevents too much current going to the ignition coils. i've never heard of this part been faulty before. (careful you dont thread the bolts on this, they should not be very tight )

the ecu is rarely faulty, but it does happen. a diagnostic check at your local fiat dealer would be the best thing to try next. the fiat examiner smart machine can help a lot in situations like this.

will visit the local fiat agent and will definitely let you know about the problem!


Thanks

jfaloi said:

So the machine runs fine when warm: pressing the accellerator on gear 0.8v, releasing the accellerator on gear (engine brake) ecu cuts the fuel altogether and reads 0v.

Click to expand...

Thats normal for modern fuel injection systems. No need for fuel to engine when on run-down. Saves fuel (read, environment).

Sorry to state the obvious but are you sure your lambda is ok? If the lambda heater is not functioning properly then it will give the effects of ok when driving (because it's hot) then deteriorating performance at idle in cooler gas stream.

Do you have a scope? Can you scope up the lambda at idle and see if it's responding quick enough? It should cross the mid point (0.45v) around 7 to 8 times per 10 secs. If it's just slowly moving up to full rich then I'd be suspicious

Most has been said so far, but I'll put my 2 pennuth in.
Problem with Oxy sensor can be discounted because its sending back the "too rich" signal correctly ie stick at 0.9v. It should only allow this rich condition if the ECU thinks the engine is under acceleration, the ECU works this out from the manifold vacuum. So my way forward would be to check that the Oxy sensor voltage is correctly present at the ECU pins. Also just as important, check the vacuum at idle and compare it with the MAP sensor voltage. Also make sure the MAP voltage is correctly present at the ECU pins.

"Problem with Oxy sensor can be discounted because its sending back the "too rich" signal correctly ie stick at 0.9v".
How do you know that Reddy? You're assuming the O2 sensor is fine.

Say the mixture is fine (as it is stated to be O2 on 0.5v on tickover after a run which is good) but the O2 is incorrectly gradually saying "Rich, rich, lean off!" when sat on tickover. The ecu will continue to lean off the mixture (being a good boy and doing as it's told) until the tickover starts to struggle, causing the tickover to wander between 500rpm and 1500rpm while it struggles to keep revs. Meanwhile the O2 sensor continues until flatlining at 0.9v giving full rich signal. If the ECU is sending down lean and mean mixture but the O2 is saying "I'm drowning in the stuff" then the O2 sensor needs therapy or shooting

Deckchair5 said:

"Problem with Oxy sensor can be discounted because its sending back the "too rich" signal correctly ie stick at 0.9v".
How do you know that Reddy? You're assuming the O2 sensor is fine.

Say the mixture is fine (as it is stated to be O2 on 0.5v on tickover after a run which is good) but the O2 is incorrectly gradually saying "Rich, rich, lean off!" when sat on tickover. The ecu will continue to lean off the mixture (being a good boy and doing as it's told) until the tickover starts to struggle, causing the tickover to wander between 500rpm and 1500rpm while it struggles to keep revs. Meanwhile the O2 sensor continues until flatlining at 0.9v giving full rich signal. If the ECU is sending down lean and mean mixture but the O2 is saying "I'm drowning in the stuff" then the O2 sensor needs therapy or shooting

Click to expand...

hmm thats the opposite way of looking at it :chin:
the theory makes perfect sense, but what kind of lambda sensor fault could make it return a gradually increasing voltage at idle when the mixture is not actually getting richer?

What kind of sensor signals rich when it isn't? A duff one!
Easy test, pull a vaccuum line inducing an artificial weak mixture and see what the O2 sensor does. If it's still showing full rich then bin it

Deckchair5 said:

Easy test, pull a vaccuum line inducing an artificial weak mixture and see what the O2 sensor does. If it's still showing full rich then bin it

Click to expand...

good idea
(although i still dont think a lambda fault could do this, probe faults are either no signal or a loss of voltage due to increased resistance in the circuit, and neither could cause creeping voltage at idle)

The original post stated that the mixture was "richening like hell", one would expect a working sensor to send back a "too rich" signal under these conditions.

Hi there, Was abroad for a week!

Very interesting all of you but read the lambda using a simple multimeter not a scope and this on the dash while driving hills and straights. Brother drove too so i could simulate mixture.

Plugs are all wet with fuel and lambda drifts from 0.5 to 0.9 in 10 seconds after a good torquie run on gear and then down to idle. could be its not responsive enough though.

What should the readings of the voltages on the MAF sensor and the idle stabilizer valve be?

Anyone has a schematic of the Hitachi ECU or at leasty the pinouts.

I Have a computer booked for this tuesday. Will xhow better.

Cheers

for wiring diagrams, voltages, and ecu pin info see the fiat manuals at http://babarmalik.com/ledmods.htm

Thanks for the BRILLIANT link Jug!

Guess what was wrong? The fuel pressure regulator!!!!

Sounds like a rolex now.

Two thumbs up

jfaloi said:

Guess what was wrong? The fuel pressure regulator!!!!

Click to expand...

always a tricky fault to diagnose. i guess you stuck a pressure gauge on the fuel rail?