Injector light cause on 16VT

Hi,

Last few times I used my car I’ve seen the injector light twice, and I’m wondering :
- what are the problems covered by this light ?
- first time I saw it, few weeks ago, the car was running very poorly, it was smelling unburnt fuel, I presume because of the cheap fuel I put in it and the fact that it was almost empty. No more problem after a few kilometers, and now it’s running with good fuel again.
- this morning, the light stayed on after turning the ignition on, stayed on after starting the engine, then turned off after a few meters ; I’m wondering what could trigger the light without the engine running ? A sensor value ? Because there was no light when I cut the engine the day before.

Thanks

PS : I won’t use Startrek because I’ve had a really bad experience with those cheap KKL cables

You can't know for sure, without using Startrek I'm afraid. But, common faults would be the Lambda sensor, the throttle position sensor, or hall-effect sensor on the camshaft. Since you said that you smelt unburnt fuel, maybe it's the lambda sensor? Also, can you rev over 3500-4000 rpm?

Otherwise, I have an intermittent TPS sensor failure, which triggers the injector light like in your case. The light comes on when I start the car and no throttle is applied, but when I apply throttle it goes away. It rarely comes on during driving.

Is your temperature gauge reading normally?

When the car was running poorly, I think I was able to rev normally. But it only happened once, and as I said the fuel wasn’t a quality one and it was on low level. Don’t think it was the lambda, it didn’t feel like overfuelling.
Maybe I have an intermittent problem like, that could explain the light before starting the car.

Thank you for the info about the sensors, I didn’t remember they were covered by the light. Also, I will change the lambda sensor, it won’t hurt, it is getting old.

Didn’t notice when I the light was on yesterday. But today the gauge was working, as usual.

You can still have other faulty sensors, like intermittent crankshaft (I don't think in your case), ECU engine temperature sensor, or VAE. So, best that you connect it to StarTrek with a good cable. If it comes up again, that is.

Sure.

I will start to worry if the light stays on or the car stops working.

Please... Coupe Fiats don't 'stop working'. It's just that occasionally they may fail to proceed.

Well, it works but (I have noticed since I'm using it as my daily car that) it's getting worse .
I previously had misfires because of the bad valve clearance. It has been fixed, but still isn't running perfect :
- a slight misfire is still present at idle
- when starting from cold, it needs more throttle to start moving
- accelerating hard doesn't feel smooth
- now the light sometimes turns on
- I just took the car and it started to shake and smell unburnt fuel at a stop. Few meters further it was OK.

Sparks, coils, leads ere changed 2 or 3 years ago and don't have many miles.
I'm suspecting an injection (dirty injectors) or fuel related problem (fuel pump on its way ?) but I'm not sure if it's compatible with the last problem ?

I also know the catalyst is damaged, it was rattling before but I didn't hear it for a long time, could it cause these symptoms ?

The lambda sensor is the main issue that can turn the injector light on, and more significantly, off again on the 16. I don't know if a knackered catalyst might cause confusion in its measurement, but I'd suspect not; the purpose of the lambda is to feedback control the engine to keep its fuel air mix in the range that the cat can handle.

So if the light is caused by the lambda sensor it tends to confirm there is a problem in the fuel feed.

Not necessarily: most likely that the sensor needs replacing.

Ordered a new one, let's see if there is any improvement.

Please don't shout at me if there isn't!

I took the car for a drive today, tada, injector light stays on before starting the engine and then turns off...

As I have just received the new lambda, I took a look under the bonnet and noticed that the wiring had already been modified. And now I remember seeing a bill from few years ago with a lambda sensor replacement, so it’s probably not 23 years old...
I tried disconnecting the lambda, and it doesn’t even trigger the light with the engine off, only when running.

And now I remember this :

There are 2 sensors on the thermostat, one for the gauge, and another for the ECU... The gauge can be right while the ECU receives wrong info, and vice-versa.
So I tried disconnecting the ECU sensor and it triggers the injector light, but the light extinguish for a few seconds first, while it really stayed ON when it was present before starting the engine. The sensor on the intake also triggers the light.

I have to try disconnecting the sensor with a cold engine to see if the light stays ON without extinguishing for a few seconds.

The only thing that I don’t understand is the shaking and unburnt fuel problem ; if the coils, plugs and leads are OK, what could cause this behaviour ?

It can be bad TPS reading or bad MAP sensor reading, so it would over-fuel the engine. I still don't get why are you doing this blind? There are ways to check the sensors manually using multimeter, but it's way easier if you connect to the ECU with StarTrek.

The few times I tried to read the ECU, it was a fail, and the last time I tried to read it with cheap chinese cables, I fried it. Maybe I could ask Fiat to do it but I’m not even sure they will be able to read it or get some useful information from it, and it will probably cost me the equivalent of some sensors so...

And this last problem doesn’t even seem to trigger the light. In fact it looks like there are 3 different problems :
- unsmooth driving and slight misfires at idle / low rpm, problem present for a long time. I’m waiting for the garage to give me a price for injector cleaning, leak inspection and fuel pressure test.
- injector light sometimes on startup, or sometimes for a few seconds when driving. It started recently.
- the shaking and unburnt fuel problem, it happened twice but I don’t remember seeing the light when it happened. This also started recently.

I will measure the TPS sensor values.

Good idea on cleaning the injectors. Btw, is your battery good and alternator charging properly? I think voltage drop sometimes can be an issue, especially as the rpms drop on hot idle.

You can also test the map and temp sensors.

Battery and alternatif should be OK, I started several times with the battery discharged (and the help of another car or battery), and after driving few kilometers it starts straight away. Now that I'm using the car daily it starts instantly even after a week.

I ordered new temp sensors (air and water), and a new MAP. For the throttle sensor, should I check a voltage with the car running or a resistance ? You said you had an intermittent problem with yours, did you replace it and what reference did you buy, I'm not sure if the ones listed on autodoc or eBay are correct.

The throttle position output is 0.25v to 5v, with the ignition on. Doesn't need the engine on; you're looking for a continuous change in voltage without any breaks or steps as the throttle position changes.

Yes, as barnacle said, you need to measure the wire in the middle with ignition on. On closed throttle, you should have ~0.5v, at least that's what the manual says. As you open it, you should see the voltage rise.

I have an intermittent problem, I think my TPS looses continuity as I vary the throttle. But I haven't measured it yet, and I will be able to get back to my Coupe in a few months.

I found this :
http://www.lancisti.net/forum/showt...r&s=c8154f6dc36fb7568f3d62a99d9ded5d

https://youtu.be/5SeS-_KOBGA
https://www.integralingham.com/tps

I removed, opened and cleaned my TPS then adjusted it as stated above. It is in good state and I highly doubt it may fail.
However, by adjusting it this way, I have a resistance of approximately 630 ohms closed and 176 ohms opened (didn’t try by pushing the accelerator but by moving the cable by hand). Someone in the thread linked had 630 and 130. When measuring the resistance with the TPS opened, extreme values were 630 and 130 too. Anyone with values to compare ?

It looks like that way of adjusting the TPS isn't really working. I came jack home and the light was ON during almost all the ride.
I adjusted it via the voltage (and I already did it once, because I found the small hole previously made in the wiring) and set it to 0,3v (I was at 0,15).
Let's see tomorrow how it runs, maybe I will even try 0,5v to see if it really goes mad.

Don't get hung up on resistor values; the ECU is measuring the *voltage* from the wiper, not the resistance. There is, as might be expected, lots of guesswork and misinformation on that forum you linked, and of course a comment scorning the only guy who knows what's happening as 'theoretical'...

(note: this is specific to 16v models. 20v models do it differently but in the same general way.)

Idle is detected when the voltage value is below a certain value; the manual says 0.5v or lower in the idle position (which will be compared against an internal reference probably slightly higher). The function of the TPS idle position is to tell the ECU that it should be in idle mode; that is, controlling the idle speed via the idle control valve (ICV). In that mode, the throttle valve is assumed to be closed (apart from a bleed valve) and the majority of the air getting into the inlet plenum is via the ICV - which has a bypass route around the throttle butterfly valve. The ECU measures the manifold pressure, air and water temperature, and calculates how much fuel to inject to match the air to give a stoichiometric mixture.

In this respect, it's the amount of air in the system that controls the speed. If the ECU thinks the idle speed is low, it opens the ICV a little to allow more air in; it then notices that there's more air than fuel and increases the speed (and the same in reverse, of course). The bleed valve I mentioned is to allow enough air through so that the engine will run at about 750rpm with the ICV and throttle both fully closed.

If the value read from the throttle valve is significantly less than the desired 0.5v, then the throttle can be opened to some amount *without* informing the ECU that it's left the idle mode. With the ICV set correctly, as the throttle opens, the ICV will close to compensate for the throttle opening, but if it's too low, the ICV may be completely closed as the throttle moves and there will be no control of idle available. As the throttle is opened further, there is normal operation once it moves out of the idle mode and into load control mode.

If the value of the TPS is too high at idle, again, idle is not being controlled; the ECU is never in idle mode *but* the only air getting into the manifold is via the bleed valve.

That's rather a long winded way to say if the TPS voltage is (much) too high, it will idle slowly; if it is (much) too low it will probably idle fast and be poorly controlled (er, rather like mine ) but after all that the thing to remember is that the voltage on the TPS at idle *only* affects idle, and possibly a small transition glitch as you start to pull away from idle.

While the engine is running in normal driving mode, at part throttle, the ECU is watching the voltage mostly to see how much it changes over short periods, basically to decide whether you're making a sudden change to it (i.e. throw more fuel in to improve throttle response, or kill the injectors to stop things farting and banging on overrun, and save a little fuel).

However, it does note when the voltage gets to maximum, i.e. wide open throttle. At that point it ignores any feedback information about the mixture and goes into mapped mode, throwing calculated values of fuel into the mix.

Which brings me back to the TPS: as long as it's within 20% of the stated resistance values, and it has a smooth change in resistance between the 0v side and the middle pin with no steps or jumps or open circuits, and the DC measured at idle is between 0,25 and 0.5v, then you almost certainly don't have an issue with the TPS.

I didn't expect such an explanation Thanks.

The only part I don't understand is this one : don't you mean "If the TPS set correctly" instead ? And in bold, for me it doesn't coincide with the previous sentences of your explanation :

Or you mean that the ECU will stay in idle mode because of the TPS but will close the ICV because of the measured air ?

----

Back to the adjustment, yes the TPS is probably fully working, and outside the random injector light problem that I am looking after, I also think it (the "idle switch voltage") might be (even partially) related to the low throttle response when pulling in 1st gear or the jerky driving in town when releasing the throttle in low gears (I can literally shake the car if I floor the throttle and release it quickly). What I understand from your explanation and the "theorical" one on the forum I linked is that the TPS should be adjusted to have the highest idle voltage possible without getting mad ; that way, the throttle would react faster when moving the car from stop and maybe be less prone to shake the car if released ?

Will see tomorrow how high it can go.

Thanks barnacle on the explanation, I'm sure many of us will 16vt-ers find it quite useful.

Yann, are you sure the shaking from pressing on/off on the gas pedal is not attributed to bad engine mounts?

No, I meant the ICV. When everything is working, the engine will idle (slowly) on just the air through the bypass valve - the brass screw on the throttle body - with the ICV disconnected, i.e. closed. Ideally the bypass is set so that at idle the ICV is in the middle of its range so it can control the idle speed in both directions.

If the TPS voltage is too low, you can get a period - probably quite brief - when the throttle starts to open but the ECU is still in idle mode. In that case, the ECU will try to slow the engine down by closing the ICV, which might cause a quick hiccup when you continue opening the throttle as things change mode.


Yes

Ok I see. I tried with 0.5v (engine ON), between gears the RPM tend to be quite high, it's not good. At 0.34v (only ignition ON) it seems to be better. The idle speed doesn't seem to be affected by the TPS, but something must be wrong ; with the engine at 80-90°C the idle stays quite high, I closed the bleed on the body but it doesn't seem to help much, there must be a leak somewhere.

This morning, light was ON on the way to work, but only outside town. On my way back home I only drove in town and no light.

May be worth setting the bleed screw, if you haven't. With the engine at full operating temperature:

Remove (counting the turns) and clean the bleed screw, and squirt some cleaner down the hole.
Screw the bleed screw back in the same number of turns.
Start the engine; might run rough until the cleaner is cleared.
Remove the ICV connector. Engine speed should drop; anticlockwise to slow it, clockwise to speed up - you're aiming for about 700-750rpm. If it stalls then unscrew a turn and restart.
With the engine still running, replace the ICV connector; the engine should speed up to about 850rpm. If it doesn't, the ICV is broken or jammed so it may be worth cleaning it to make sure everything is moving freely and that the coil is neither open nor short circuit.

If this doesn't help your basic engine setting, then I'm stumped. Personally, I'd get startrek onto it, but then I would say that, wouldn't I?

I know how to adjust it, I removed the throttle body and cleaned it inside / outside like 2 years ago, and I think I adjusted the idle at the same time.

But it looks like it still idle around 1000 RPM when hot even with the bypass closed (or almost closed) (and with the ICV disconnected).

Then you may have another leak somewhere in the system. Split hoses (I found a leaky pipe from the ICV some years back) or perhaps dirt in the butterfly stopping it closing properly, or worn pivots for the butterfly (speculating here, but this was a common problem with old carburettors), or a bleed in the inlet manifold? Or the MAP sensor pipe is known to split...

Or possibly you have a faulty MAP sensor; not returning the correct voltage for a given pressure. That will require specialist equipment to test and fix, I fear, but it should give a voltage ranging from 0-5v as the pressure rises from zero bar to 1.25 bar (NA) or 2.5 bar (turbo) but I *think* there may be an offset involved as well. That would give a voltage of close to 4v for normal air pressure, engine not running, for the NA and 2v for the turbo - but as I said, I'm speculating guessing on the pressure sensor chip probably fitted; these are more modern but give the general idea: https://docs-emea.rs-online.com/webdocs/14f2/0900766b814f2293.pdf

I just received 2 new sensors, the blue one for the water and the MAP sensor you're talking about. I should receive a new air sensor tomorrow, and will try to fit all of them on wednesday. I will try to compare their value at the same time.

There is probably a leak somewhere, maybe a worn seal. It will be checked when the injectors will be cleaned.
I cleaned the throttle body and didn't use the car much from then on, I doubt it is the culprit. I remember checking the pipe from the MAP too, it was fine.

No more light since I replaced the MAP sensor, and both water and air temperature sensors.

Injectors will be cleaned next week, and now I have to focus on the MOT issues...

Did you solve problems ? If you did please explain me how.I am having almost same problems after rebuild.Car doesn't accelerate smooth.Exactly when reach 3000 rpm.Engine shake poor idle but sure not misfire.During rebuild cleaned injectors and next will be MAP sensor and TPS. Worst part there is not any error or issue on IAW Scan 2 and Star trek.Don't know wherever else to check problem.Thanks in advance!

What does your temperature gauge read hsnlpy?

And does it take long for the heaters to work?

I think it reads normal.Car start at one time on cold.But when is hot if stop the engine sometimes dosen't start easy.I will give check with details next week I am out of town now.I can add one screenshot from diagnostic maybe it can give some idea.I have doubt about throttle opening angle.

https://ibb.co/txywH4H

Try the bonnet-open test....start the engine from cold and put your hand on the top hose from the radiator. It should be cold for a few minutes until the thermostat opens and it suddenly goes warm. If it gets hot immediately it's likely your thermostat has stuck in the open position which can affect your performance, including the injector light coming on and excess fuel being burnt.

Thanks i will give a check.Did you see something not Ok on screenshot ? Because there was no error in IAW Scan2 ..

Sorry, not my area!

The light have not appeared since I replaced the air, water and MAP sensors.
I also replaced the lambda few weeks later.
The car ran very poorly for a few seconds twice, it doesn't occur anymore.

I still have a slight misfire, and still have to check for air leaks, I'm waiting for a "tool" to arrive.

Have you checked your crankshaft sensor, it may be related to the starting problem.