NefMoto

Hi.

I've taken one extra ecu to play with so I was wondering if it is possible to do the following:
1. Read my ECU flash
2. Read my ECU eeprom (immo)
3. Write all this on the new ecu.

Basically I want to do a clone of my car ecu.

Thanks!

There are different hardware versiones of the ME7.3.1. The most obvious is that the 4-cyl ECUs don't have outputs for a 6-cyl engine, but there might be other differences as well. I did try to put the FLASH and EEPROM from a GTA on a 4-cyl ME7.3.1. Car did start, but naturally only ran on four cylinders.

A straight clone might work. You can also try to make the ECU virgin. I've just written a small paper on how to "virginize" the ME7.3.1. Seems to work alright when I test it on the bench, and did work with the GTA/4-cyl test as mentioned, but it would be interesting if more people did test it.

I attach the PDF with the guide. Please report back your results, and also report if anything is wrong in the guide.
 

The spare ECU came from 2.0ts engine. My car is 1.6ts. So everything should be the same except for one additional driver that 2.0 ecu should have to control the variable length intake.

I'll go trough the guide and try to make a virgin ecu. Will let you know the results.

Tnx!

I've had a look at the guide. It is pretty clear, however I have one question. How do you know the memory type of the eeprom?
When I was downloading the eeprom I selected 95040 and got a successful read. It worked on both ECUs.
But when I compare the hex with the one in the manual it looks different.

I've attached one bin so you can have a look.

Open the ecu. Read the numbers on the eeprom.

I used trial and error. Read as 95080 and you will find that there's more information in the EEPROM past address 0200h. Reading and writing as 95080 seems to do the job fine.

ME7.3.1 is a hybrid ECU. I tried identifying the bare dye chips on the ECU hybrid under a microscope, but I couldn't find any readable markings. The protective silicon coating blurs the picture, and the microscope I have access to is limited in magnification. If anyone has better equipment and can identify the bare dye chips, any help is most welcome!

I tried reading all formats. I agree, there are more data after 0200h. When you read as 95P08, all the data after 0200h are filled with FF. When you read as 95080 you get some data after 0200h. When you read as 95160 you get exactly double of 95080.

So I'd say looks like 95080 :)

I've compared it to the file in the manual. The virgin line in this immo looks like only for address 0068h. And the checksum calculation works for this line. The line with 0058h doesn't compute. I've attached the bin so you can have a look.

That's strange! All the EEPROMs I have read from ME7.3.1 have had the two duplicated identical lines. Was it the same result on both your ECUs?

I didn't check the second ECU yet. I've just read the spare one. I can check the original one today when I get home. I'll let you know.

Same with the other ECU...

Really strange.  ???
I'll put together all the ME7.3.1 eeprom dumps I have as a reference and post them. Don't have the time until after the weekend though.

Here's all of the eeprom dumps I have at the moment. They all have the duplicated lines, and two of the dumps wasn't read me, but from two different Alfisti.

Neno-ri, did you use the exactly same command line as in my guide?

I've used exactly the same command line. I have no idea we have this difference.

It might be a difference in the ECUs . Your two ECUs have different P/N from the ones I have, so we can't compare them.

It will be interesting to here if you succeed with the cloning.

Yesterday I finalized the cloning. Today or tomorrow I will try the ECU on the car.

I tried the clone today. Works like a charm :)

No immo errors, nothing. Re-calibrated the throttle and started the car without and problems or errors.

Great news!
Did you just clone the eeprom, or did you try the virgin mod?

ECU/SW numbers was: 0261206707-1277356512 and 0261206715-1277356511, right?

I just cloned them. Virgin mod will be next :).

Correct. Those are the numbers.

Hi redtop, I have ME7.3.1 too, and in my dump aren't duplicated lines too. It looks like Neno-ri's dumps. I attach it for you.

If I can have one question: I tryed download ROM image from my unit via Galletto in bootmode. It was downloaded successfully, but it looks totally different when I compare it with dumps from other types ECU with this kind of processor. Missing the interrupt vector table on start of file for example. But in some places I can see readable texts, so it can't look as garbage. It is anything wrong on my side, or it is normal in this unit?

A bit off topic!  ;)

I take it it's from a Lancia Thesis 3.2 (0261208118-1037368765) ?

I don't know anything about the structure of the software, but there are some good info about the Alfa 147 3.2 here:
http://www.ecuconnections.com/forum/viewtopic.php?f=25&t=11594
Maybe it's useful for you?

I attach a couple of other dumps from Alfa 3.2 with the ME7.3.1 so you can compare.


(BTW are you Alfachip in other forums?)

Yes, you are right, its as you say. Lancia Thesis 3.2.

I am sorry, if it's offtopic. It's the same ECU, so I sended it for info, because it haven't duplicated lines and have the same structure as Neno-ri dumps.

Thanks for yours dumps. Looks like my dump. (The AR166 zip file is broken).

I am not Alfachip. All my nicks on every forums ended with "ertik" (czertik, e-certik, ecertik...).

And for info, the reinit of virgin ECU work fine. I tried it on Thesis, I changed the ECU for another (i have two spare cars) and car normally started.

OK, someone was asking for an immo-off file for this ECU in other forums. I thought it might be you.

You just cloned the eeprom, and that did the trick?

(The 166 file works fine here, I upload it again just for reference.)

Thank you. I can download the file, but I can't unrar it. Maybee I have old version of rar plugin. I use totalcommander.

I didn't clone eeprom, because i haven't original one. But clonning working fine, if you clone booth FLASH and EEPROM.
My solution do not disable imo. Imo is still active. It (only :-) ) cause le-learn the ECU from bodycomputer (BC) (because i didn't have other source. My original ECU was broken). After this the ECU working like original or cloned one. It have original IDs (I mean IDs printed on label), but fully cooperate with original IMO and BC. But without clone. :-)
I don't know, if only on Thesis, it can work in any car with BC by my opinion.


 

Hello friends. I am Alfachips.
I did a clone ecu. I have taken ecu from 147 1.6L with 105 hp and I put on 156 1.8L and the car started normally.
Cetrik, can you attach the original flash file from Lancia Thesis 3200 ?

Redtop: I used winrar and i successfully extracted the AR166 file. ;-)

Alfachips: It look like not complete dump, but the same structure as some other I downloaded.
Here it is:

Thank you Certik.

so

if i clone eeprom from one ecu (me 7.3.1) to another (flash on both ecu's is different) car would start or not?

thx

In this case, when the flash are different, is unsafe clone whole eeprom. It can work, it can have some unpredictible errors/problems, it can stop working.

It depend on flash program.

flash different because one ecu from automatic transmission another from manual

engine type and all over are same

what do you think about it?

You can try to make it virgin as in the instruction attached to the second posting in this thread.

i download it last night)

will try do it
thx

Make it virgin as redtop say is safe way. It will work. ;-)

I do not understand how to calculate chechksum ???

Who can help?

thank you

original eeprom in file

Seems like you have calculated the checksum correct in your file! (F9DA)

redtop  you are the best!!

write eeprom to ecu

car dont start

no engine error message

you changed gearbox from Selespeed to manual, right? Maybe it's the wrong flash file for your configuration?

yes, change. But this eeprom (from donor ecu) work fine with donor body ecu installed on my car,  now i want use my old body ecu with donor engine ecu .  

It was an Alfa 147? Maybe you need to do a "proxy alignment" in your body computer?

yes,i do it first)

seems like there is no gasoline, but gasoline is ok)

upd

it was a bad mass on motor ecu when i tried

now all perfect!!

one more ecu

attached original eeprom and virgin

and there is immo error after virgin

thx

As far as I know you got the checksum right. I have no idea why it doesn't work.

first ecu work right
second ecu does not work

i tried

third and fourth ecu and 'virgin' work good

Can someone check if is this eeprom from me7.3.1 0261206716 virgin?
This dump is readed from ecu that not been virgin.I dont know how  in lines 058h and 068h is DAh instead 25h?

I'm looking for a eeprom dump of a 0261207967 Alfa 147 GTA ECU.

Can someone help me with this, not the engine MAP.

I need help, I'm looking for a eeprom dump of a 0261206707 Alfa 147 2.0

For what exactly ?
You can just fille the EEPROM with 0xFF and the ECU will recreate the content after ignition is switched off in the afterrun.
Afterwards the ECU is virgin and will pair with the IMMO again.

If you remove the K-line wire it will stay virgin forever,
but the code lamp is on as the Body computer will detect no communication.

 Hi, I needed to clone the 2.0 ecu to the 1.6 after the engine change. But I managed to find it and it's all right. It works.

Exactly I needed the eeprom of the alfa romeo gt 1.8 twin spark (0261208725-1037370360) because the engine is 1.8, I found the flash file but I miss the eeprom file. But until I find it I have to be satisfied with this.

You dont need the EEPROM data at all.
Flash the bin matching your engine to the ECU (almost all 4 CYL ME7.3.1 are the same hardware).
Then write a file created in the hex editor with all FF to the EEPROM (CSpin4.7).
Turn off ignition, wait ~10 seconds, turn on ignition.

The EEPROM has now been recreated by the ECU, as this is exactly what happens in factory.
You then have a "brand new" virgin ECU.

I've just tried it but the machine doesn't start, or at least it starts but has errors and turns off. I tried to read the eprom after this and the file was encoded up to a certain point then there are all "00". I attach image.

Can you check if I did it right?

Eeprom was recreated correctly, the empty sections are normal (they are unused),
but they have the correct checksum to the right.

Also the ECU already paired itself with the IMMO correctly which you can see on the 0x25 bytes (which are after or before the immo code, dont remember which way round) on page 5&6.
If its unpaired its either 0xDA or empty.

Did you delete all the error codes and try to start again ?
If it does not work it is more like hardware around your ECU than the ECU itself,
everything looks totally fine.


If it starts and then turns of it is NOT the immo, if the immo is active it will never start for these ecus as there is NO injection at all then.

I made a diagnosis and the program did not give errors. But it appears in the instrument panel. I also tried to load this eeprom virgin on the 2.0 ecu that normally works and that doesn't work either. Then I reloaded the eeprom of 2.0 and it works again.