[~~h]

I have a Sept '09 Punto 1.2 with 30K+ on the ODO. Few days earlier the the OBD indicator light (image below) came up in yellow. The car does start and stay in idle, throttle has no response at all.

The car is at the dealer for 3 days, all they did so far as diagnostics is erase the error codes and drive for a while; pretty obvious they are clueless. What they assume now it's an accelerator wiring or ECM issue and checking by replacing them with parts from another car.

Any idea what could be the possible issue here?

[DerAlte]

Didn't they tell you what the error codes were before erasing them? Analysing the error codes will help in zeroing in on the problem. Neither the accelerator wiring nor the ECU would be a problem. If the ECU has a serious problem, it wouldn't have started.

How did they 'drive for a while' if throttle has no response? It is easy to check with a multimeter at the ECU connector end to find out if there is a problem with the accelerator wiring.

[sandeepmdas]

Apparently, your car's EOBD has detected some error and it promptly put the engine into "limp mode" in which you can limp back home. This is the way the EOBD of most modern cars are programmed to do.

Fault indication during start up:

the failure of the warning light to go out when the engine has been started indicates that there is an error memorized in the control unit.

Fault indication during operation:

- the warning light comes on in flashing mode to indicate possible catalytic converter damage due to misfiring.

- the warning light comes on in constant mode to indicate the presence of engine management or EOBD diagnostic errors.

[~~h]

Originally Posted by DerAlte

Didn't they tell you what the error codes were before erasing them? Analysing the error codes will help in zeroing in on the problem. Neither the accelerator wiring nor the ECU would be a problem. If the ECU has a serious problem, it wouldn't have started.

That's classic Fiat service. They don't seem to be analyzing them, and the replies given to me vary from day to day almost all of the them circling around accelerator wiring.
Even I think ECU is fine, thinking of taking a day off and go sit on the dealer's before he mess up something else.

Originally Posted by DerAlte

How did they 'drive for a while' if throttle has no response? It is easy to check with a multimeter at the ECU connector end to find out if there is a problem with the accelerator wiring.

After erasing the error codes, the the throttle works normal, drive for a few mins and the light comes back. Seems he did the multimeter check, I was told there is variation in voltage.

This car has gotten me into more of such situations in 2 years than all other cars in 20 years combined, so much for Punto love.

[DerAlte]

Originally Posted by ~~h

... After erasing the error codes, the the throttle works normal, drive for a few mins and the light comes back. Seems he did the multimeter check, I was told there is variation in voltage. ...

And still they didn't wonder what the error codes were? Time to escalate to TASS! One can understand lack of knowledge, but one cannot reconcile to lack of sense to access help from elsewhere.

There is no voltage that comes from the Acc pedal sensor - the ECU reads it by injecting voltage. If you take out the ECU-side connector (that's what they should do to check wiring defects) they should read resistance with a multimeter. That resistance should vary according to pedal position (low resistance without pressing pedal, increasing to higher resistance when fully pressed).

If the ECU is switching to limp-home mode, there is something seriously amiss with one or the other component: throttle body servo, injector system, rail pressure sensor ... something that makes the ECU think that it is not safe to drive at speed. Yes, the best would be that you spend time with their diag equipment and drum some sense into their heads.

[sandeepmdas]

The accelerator pedal is equipped with two built-in potentiometers:

- one main one;

- one safety one.

The injection control unit implements the following recovery strategies in the following conditions:

- if one of the two potentiometers fails, this allows the throttle to be opened to a maximum of 40° over a long time period;

- if both potentiometers fail completely, it prevents throttle opening.

If there is a fault with the two potentiometers or no power supply, depending on the position of the accelerator pedal, the control unit applies a recovery strategy with the consequent inferior operation noticeable to the driver and the EOBD fault diagnosis is disabled.

[DerAlte]

Originally Posted by sandeepmdas

... and the EOBD fault diagnosis is disabled.

Err... why? Why should all the diagnostic routines be disabled? That would prevent other component failures from being recorded! EOBD is only output over the diag port, and is never disabled, no?

[anilisanil]

That is the warning light which means "check engine" it means anything could be amiss with the engine. Most modern ECUs put the engine in limp mode (where there would be no throttle response) so that you could safely drive your car to a more favorable place.

I think the service guys do not have a clue on what they are doing and advice you to ask them to stop working on the car until you give further instructions. Meanwhile find out if they are getting any technical assistance from Fiat for the same and try from your side if they haven't.

OBD codes are the most important link between a broke down engine and its diagnosis, I am surprised they deleted all of them! Sigh!

[avira_tk]

The EOBD light is something I don't trust, our Fiesta light used to blink whenever it started raining - apparently it is sensitive to moisture. The throttle response malfunction when the rpm does not cross 1000 and the dashboard info getting blanked out was rectified by replacing something called VSS(vehicle speed sensor), EOBD did not blink at this point .Fiat must have something similar, I doubt Service folks will be able to fix this, electrical/electronics is something way beyond TASS regular capability, it took two days for an experienced guy at Ford. Most likely they will try to replace whatever they can if its a paid service.

[sandeepmdas]

Originally Posted by DerAlte

Err... why?

Same here...

Please find the attachments, it's there at the THROTTLE body SPECS.

Also please note that this document speaks about the Marelli system; I am not sure whether it is the same one used in the GP India-spec version.

[DerAlte]

Atrocious translation (Italian to English)!!! (Yes, the engine electronics stuff is all Magnetti Marelli)

The general principle is: If it is sensed that there is a 'catastrophic' irretrievable failure with a particular sensor, that sensor is marked 'disabled' and no other software unit uses that information. This prevents unnecessary periodic / repeated logging of the same fault from the same sensor. This is on a per-sensor basis, not *entire* diagnostic mechanism.

If the information from the sensor was critical to operation, the central watchdog module decides to put the vehicle in limp-home mode - just so that the vehicle does not process implausible data and put the driver AND the surroundings to risk. Hence the limited speed (@anilisanil - not no throttle response, but very limited throttle response in that case).

The other phunny information in the above document is 'potentiometer' controlling the throttle. It is actually a servo motor that does that. There *is* a potentiometer (2 in fact, for redundancy, like the acc pedal potis) - that is used by ECU to read back actual throttle position. Like, "I didn't send a throttle position output, so why is the position showing as not zero?" is a plausibility check on that poti.

In @~~h's case, they just have to zero in on which sensor / actuator has gone kaput bad enough for the ECU to go to limp-home mode. Acc pedal sensor is the most obvious (and the most obvious wrong diag in an almost new car) one, but there are a couple of dozen other possibilities.

[~~h]

The update: I went to the dealer yesterday to see what they are doing; tells me they are getting an error code which is not a standard one. I ask them to run diagnostics again and finds them selecting 1.3 MJD in the screen corrected them and bingo there's a valid error code.

PD 1602 AD Convertor Fault, Signal not feasible.

Surprisingly they have no good answer to what did the do since 27th Nov.

Now the second part of troubleshooting; the error was found at 12 Noon yesterday and I spend till 6 PM at the dealers and they could not fix it. They were to call back in the morning; now they are not picking up the phones.

I have few friends in Tata motors, i guess it's time to trouble them; in the mean time if you have any numbers/ email for Fiat please forward them.

Originally Posted by anilisanil

I think the service guys do not have a clue on what they are doing and advice you to ask them to stop working on the car until you give further instructions. Meanwhile find out if they are getting any technical assistance from Fiat for the same and try from your side if they haven't.

the reason for not getting a good code was because they were selecting a wrong engine; even now after zeroing on the problem there's no progress

Originally Posted by sandeepmdas

The accelerator pedal is equipped with two built-in potentiometers:
- one main one;
- one safety one.

.

Thanks Sandeep for your inputs; apparently this car is in my native (Trivandrum), so you know which dealer I'm talking of (don't want to name them yet on a public forum). Please PM me if you need details.

Originally Posted by DerAlte

Atrocious translation (Italian to English)!!! (Yes, the engine electronics stuff is all Magnetti Marelli)

.....
In @~~h's case, they just have to zero in on which sensor / actuator has gone kaput bad enough for the ECU to go to limp-home mode. Acc pedal sensor is the most obvious (and the most obvious wrong diag in an almost new car) one, but there are a couple of dozen other possibilities.

Thanks I was wondering what's this Magnetti Marelli.
The car seem to have no issues except it's not able to communicate to the Cat Con sensor ; sad they still can't fix it and is still telling me it's ECM issue and ECM earthing is not good.

[DerAlte]

1. Ask them to replace the ECU and check again. Will isolate if it is an ECU internal problem, or if it has something to do with the rest of the system

2. In the FIAT OBD world, PD 1602 could be any 1 of 4 defects: A/D converter, Battery voltage, Voltage circuit or Voltage regulator. Unfortunately, FIAT guys are not very eloquent in their error descriptions. Under the circumstances, one can only presume A/D converter, voltage circuit or voltage regulator would be ECU internal defects, and battery voltage an external problem (does not mean battery is at fault; just that the voltage at the ECU terminal would be less - bad contact or high resistance somewhere)

ECU earthing *may be* an issue, but it is amusing to see how the gentleman is taking shots in the dark. Instead, it is better to build a binary decision tree by systematically eliminating possible fault areas.

[~~h]

Originally Posted by DerAlte

1. Ask them to replace the ECU and check again. Will isolate if it is an ECU internal problem, or if it has something to do with the rest of the system

That's what they are trying to do now; but Kerala is a diesel nation and I haven't seen another petrol Punto other than mine, so finding another ECU is not easy.

Originally Posted by DerAlte

2. In the FIAT OBD world, PD 1602 could be any 1 of 4 defects: A/D converter, Battery voltage, Voltage circuit or Voltage regulator. Unfortunately, FIAT guys are not very eloquent in their error descriptions. Under the circumstances, one can only presume A/D converter, voltage circuit or voltage regulator would be ECU internal defects, and battery voltage an external problem (does not mean battery is at fault; just that the voltage at the ECU terminal would be less - bad contact or high resistance somewhere)

Thanks DerAlte for your valued inputs; I'll pass on the suggestions.

Originally Posted by DerAlte

ECU earthing *may be* an issue, but it is amusing to see how the gentleman is taking shots in the dark. Instead, it is better to build a binary decision tree by systematically eliminating possible fault areas.

The lack of knowledge of these guys are obvious.. sad.

[~~h]

Update : I mailed Mangesh (thanks to this )and got a response saying he'll look into the matter ( that was on 3rd Dec; no action since then.
My family happen to hold Tata motors shares and I'm trying to reach out to investor relations as my last resort.