Technical "Low voltage" error and the ECU replacement...

[ser907]

Hi everyone,
I need a help with my Cinq 899 .
The error-check shows an error „Low Voltage“ and the voltages measured by ECU are significantly lower than the real voltages, measured by tester. So, when the engine is not working, ECU shows 10V (in reality 12.6). By working engine ECU measures 8.5V and the real is 13.5V. The earth is ok.
Sometimes the engine shows another error “Coil internal shorting” but the coil seems to be ok too. I think these two errors are connected together (ECU measures the current to the coil and if the voltage is too low, the current is too high -> looks like shorting)
I suspect that the ECU itself is nearly dead and (as a first cheap step) I want to replace it by one from the e-bay. My questions are the following:
1. Had anyone the same problem?
2. If would be great to have the diagram of IAW 16F.EO Ecu. Is it somewhere to find? What could be damaged in this ecu? Electrolytic capacitors look optically as ok.
3. My car have the “Fase 2” Immobilizer, I have two blue keys, the original card with “electrical” and “mechanical” codes but no red key. Can I replace the ECU only? If yes- how should I learn my new ecu to read old keys? There is no way to do this using IAW-SCAN2 or multiecuscan software…
4. If no- can I just replace the chips from both ecus? I suspect the electronics is not in order, but chip itself is ok. In other words, the immobilizer info seats in chip or somewhere else?
Many thanks in advance for your answers, comments and ideas.

 

[woj]

In reverse order - the immo code sits in the CPUs EEPROM, so replacing / moving the chip will not help. You have to either get a complete setup from another car: ECU + immo box + keys, and transplant it all, or get the ECU virginised, it will then learn the immo code from the box in the first run. This unfortunately is not a job that a diagnostic tool can do, the ECU needs to be treated by a knowledgeable person (you will surely find one in Germany, on this forum too, including myself).

But then - these ECUs are difficult to kill, so I'd suggest really looking hard first for a fault in electrics. You may also try to connect the ECU to another car - it won't start the engine because of the immo, but should otherwise work flawlessly and show no errors.

 

[g8rpi]

The fact tht the ECU shows a lower voltage with the engine running would tend to indicate a bad earth somewhere in the system. Trying the ECU in another car or a spare ECU (even without immobiser set to your keys) will give a good idea were the fault lies. If the ECU code is in an EEPROM it could be copied to the new ECU or just unsoldered and transferred to the new ECU.

Robert G8RPI.

 

[ser907]

Many thanks!

I think about bad earths too, but the tester shows that everything is ok, at least at the unit entrance... Maybe bad earths INSIDE the box, therefore the plan would be very helpful...

The ecu from another car will definitely confirm this idea. Of course, it is very difficult to replace the EEPROM as it is soldered. But what this funny procedure with the code from the fiat-card and gas pedal treatment will give? Second question - that will happen when the completely new EEPROM chip is mounted?

 

[woj]

Completely new will not work, or at least I think so, never tried (will some time).

If the tester shows ok voltages with the ECU unplugged but the ECU complains that does not mean much, remember that it closes some circuits that are otherwise opened.

The pedal start procedure is one time thing IIRC, and I am not even sure the 16F has it.

 

[ser907]

Ok, it's nearly clear what's to do... I will buy the ecu unit and then the fight will be continued....

I'm agree that testing the plug of disconnected unit will not give much. But, from the other side, it shows that the correct earths and +12V come to the unit. Maybe the next check will be to disconnect each particular circut/sensor measuring the voltage changes by tester and by IAW-scan.

The unit must have the pedal-procedure, at least, it is written in original papers. But do i understand the logic of this immobilizer correctly? The code sits in a key, the small box reads it and sends it to ecu. The ecu compares it to EEPROM value and allows the engine to start? By this logic the pedal-procedure will be not suscessfull because the codes from the card and saved in EEPROM are different....

 

[g8rpi]

Hi,
I agree with your understanding of the code storage. What is not clear is the contents of thee EEPROM in a new "virgin" ECU. There are 3 options I can think of
1\ Blank (all FF) This would be an easy fix, just fit new blank EEPROM, but is least secure.
2\ Some data eg a key code of 000000 which is rcognised by the software and allows a one time key programming.
3\ At least one key code already programmed. This is the case with some later models where you have to give the code when ordering ECU.

You could try to get a key with the new ECU. You might then program your keys to the ECU or transfer the transponder chip to your old key. At least you can wrap the old key body in metal foil to block transponder whileusing it to operate sterring lock and hold the new ECU's key next to it to operate the immobiliser for testing.

Robert G8RPI.

 

[ser907]

The plan is first to get a "new" ecu unit form e-bay, which must be cheap. If this ecu shows no errors, i can replace the EEPROM chip to the completely new one. (Of course, it is possible to change EEPROMs between two units, but this way is much more risky).
The really new ecu should be in a "virgin" state just because of technology. But it could be expensive if you get it from the official dealer....

 

[woj]

The EEPROM is inside the CPU, are you sure you really want to go this route? Geting a second hand virginised ECU shouldn't be a big problem.

And, I still claim it is not the ECU

 

[ser907]

i think 15 eur is a small price to check this version. If a "new" ECU will show no errors and real voltage readings then i will try to solder the new cpu or ask someone to do that or search for another virgin cpu. If the problem persists - that will be another story....

 

[ser907]

So, the problem was solved. Of course, the new ecu has shown the same error and, after detail check, i found the bad contact in the relais powering the ecu.

So, what should i do to erase the immo-code from the ecu? Just to solder the new processor there?

 

[woj]

Of course

Not sure if a blank CPU will do the trick, but if you have a suitable EEPROM reader you can surely find info through Google how to modify it. Or give it to a specialist.