348 2.7 Motronic chips

I am in the midst of performing a test installation of Dave Helms' connector kit for the 348 2.7 system and during the course of the install I rectified a chip degradation issue. Dave's connector kit has already yielded significant improvements in how my car runs, and that deserves its' own thread when I have completed it - this thread will deal with what I have learned about the Motronic chips.

Background: I have had many intermittent electrical/sensor issues on my 348 (if this sounds familiar, keep reading). Most were sorted by the best practices owners have shared: Eliminating battery quick disconnects, checking fuses and relays, having good grounds, testing sensors, etc. One piece of conventional wisdom has been the use of Stabilant 22a; although it appears to deliver some initial benefit, over time I (and others) are finding it creates a lacquer-like finish on the pins which probably hampers conductivity instead of improving it. Recently I went through all my connectors and using wire brushes and contact cleaner, removed the Stabilant build-up.

Which brings me to the last week or so - although I have swapped Motronic ECU's on several occasions to test the integrity of them, they had never delivered a consistent set of error codes between banks (which might indicate a failing ECU - for example, if I get 1113 consistently on bank 5/8 and the throttle sensor checks out, switch ECU's and then it shows up in bank 1/4 that would be an indication the ECU is at fault). So I was pretty confident my ECU's were okay - No Doubt has a great thread on chips and checking them:

I believe I have uncovered another fault indicator - when the chip starts to degrade and certain memory locations fail to be read correctly it appears it will show up as sensor faults (and stored error codes for those faults). I was still having the occasional sensor fault pop up, and every once in a while it would cut a bank out for a second or two, but nothing consistent which made troubleshooting difficult.

To test this theory out, I purchased an eprom programmer and a batch of automotive grade chips (our 2.7 systems use 27C256 eproms). I burned a copy of a known good chip and used the file to produce two cloned chips which I inserted into my ECU's. I've put about five hours of drive time on the car since the chip swap and it has run without one hiccup. No more error codes. No more bank sporadically cutting out for a few seconds here and there.

From what I have read about the 27C256 chips, automotive grade ones have a life expectancy of about 20 years. Some may fail instantly; in my case, it appears it was starting to gradually decay and inhibit reads of certain memory locations. The good news is the fix is cost effective and all the info to replace the chips exists on this site. No Doubt's thread, quoted above, is a great place to get started.

If you choose to work on your own ECU's exercise caution as static discharge can fry them. If you do not have experience working with computer boards/chips/etc. get assistance from someone who does have experience.

Now about Dave's connector kit - I can't recall my car running as strong as it is now. I'm not done yet and the results so far have been impressive. More on that soon!
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login

 

interesting info. Maybe I'm being dense, but what does Dave's connector kit do? It seems like you took the chip out of the ECU unit on to your USB programmer, it doesn't seem like you need anything else...

 

That's a seperate project; I probably wasn't clear on that, thank you for pointing it out. Dave's connector kit addresses issues with the wiring harness, specifically the connectors that attach to the pins of our sensors, fuel injectors, etc. The specification for the OEM connectors is for five cycles (connects and disconnects). As the connectors wear from routine maintenance they can introduce a number of issues with the performance of our cars. Dave explains it better than I'll ever be able to:

 

Is this thread about maintaining your chip prongs? As a geek hobbiest, I certainly go along with that. Yes, the best thing one can do is to make sure that those little prongs are nice and clean and shiney and dry and not coated with anything. Hey I wonder if we can still get those gold plated ones and just clone the stock ones...

Stabiliant is not intended for use in a computer, I'm pretty sure, especially one that gets real hot. There are too many things that can go wrong when you use a conduction enabler on a board. In the engine bay on your electrical connectors, sure, but not in the computer.

 

No, it's about another indication of Motronic chips going bad and what can be done about it (chip cloning). Stabilant is for wiring connectors, not computers boards/chips - two different things.

 

EPROM chips can lose/erase random data at random memory locations over time. These chips are reasonably reliable, but time and harsh environments take their toll.

The loss or change of *data*, however, will fail the Motronic 2.7's CRC test. This was *probably* done to dissuade hackers from programming their own chips because the M2.7 boxes will turn on (and won't turn off) their CEL on your dash for that side of the engine when a chip fails the CRC test. However, when you "pull the codes" on the M2.7 box you may not see anything. The change of a *single* bit of data will fail the CRC (hackers for these chips must change other data to offset any changes). It's very sensitive, which is good as far as insuring chip quality goes over time.

EPROM chips can also short circuit over time, impacting the motherboard (e.g. a short to ground will impact the board, as will a short to power). This *can* throw a CEL M2.7 error code. It certainly will impact your performance! This is why your first test of a "dead" M2.7 ecu should be replacing the EPROM chip. You can easily fix such a "dead" M2.7 ecu with a fresh chip if a bad chip is the cause of its ills.

I have.

So if a CEL is always on...check the EPROM chip on that side's M2.7 ECU first. If you get lousy performance that hints that you have a completely dead M2.7 ECU (especially if the CEL *never* comes on)...swap the EPROM chip first. If you get the 1313 error code...swap the EPROM chip first.

Moreover, there's no downside to upgrading an early M2.7 to the 348 Spider's (& 348 SS) EPROM code. Slight hp boost if you upgrade to the pink chips, too. So there's a nice bonus if you have to "fix" a dead M2.7 ecu, anyway. Bit more power for your trouble.

 

I disagree.

Stabilant, as we are all using it now and in the past, is proving FAR more of a problem than a solution. When used on anything short of a perfectly clean connection it absolutely assures a failure point. That's said, there is no possible way to assure the female connectors are perfectly clean at the contact points, they must be replaced.

Any loss or corruption of data PRIOR to the ECU processor completely assures that there will be problems. The age old...Garbage In, Garbage Out directly applies here.

Everyone is checking the components at the component terminals, this is not what matters. What does matter is the component readings at the ECU terminals. Check the readings at the component itself and you have ignored the single highest failure point in the whole system, the plug contacts themselves. Ferrari saw fit to load the plugs with non-conductive grease and then cover the connection with a boot that only lasted a few years where it meets the harness sheath. A good connection of the boot to the plug then assures that we have a nice water and dirt trap holding moisture and swarf right on the plug connections themselves.

In this equation, the sensors themselves are very high quality and have proven to be extremely reliable. The connection points to the sensors are under designed for the application and have been failing for years now. My kit addresses this problem using Gold Plated contacts to assure a positive contact at the highest failure point in the system. If one cant get accurate data from the sensor itself to the ECU chip....no chip made can process corrupt data and give accurate outputs....Garbage In = Garbage Out.

3fourty8 has implemented changes to the ECU while in the process of installing my Gold Plated Connector Kit. With the single end goal of making the car run right this is completely understandable and justified. In a perfect world I would have preferred that the ECU's remained completely intact until the very end of the SR Gold Connector Kit install so each system and repair procedure could be judged on its own merits but we have to also remember the end goal here. 3fourty8 has been very helpful in bringing the 348 kit together early by identifying the needed connectors and boots for this model as well as an accurate critique's of the kit itself. As I do not have a 348 on the schedule for another month for me to do it myself, his assistance has been instrumental in getting this done quickly.

 

How is this done? Where do you get the 348SS EPROM Code? I am thinking of backing up my chips anyway, so doing and upgrade would be great. I assume there is nothing more than buying the eprom programmer and a batch of automotive grade chips and installing the code?

Thanks,

Mike

 

Correct.


You have a very good idea to back up your Ferrari EPROMs. You own those chips, yet they could fail (given enough time, perhaps hundreds of years, all EPROMS will fail eventually). Backing up your own chips eliminates the legal risk of being sued for copyright infringement, as well.

 

In my enthusiasm to share some information I've probably created more confusion than anything. I sorted an ECU issue prior to Dave's kit installation - my thought process being that results from the kit install may be flawed if the ECU issue wasn't addressed first.

ND is our resident chip guru, and I would be happy to help anybody locally who wants to clone their chips as I have the equipment and software from doing mine.

I am still working on installing Dave's kit and I will leave that for a seperate thread. To clarify: The ECU fix got rid of some issues I was having and gave me a stable platform to begin installing Dave's kit. Dave's kit has so far resulted in a smoother running car and I feel some "seat of the pants" improvement - although subjective at this point, the car just feels more lively to me.

 

Eric,

How do you know the problem was with the ecu chips? maybe the fact that you put in a new chip into the female ecu board scrapes off the corrosion and make a better connection and now it works? Did you try putting the old chips back in to see if that is the case? A couple of pullouts and putins will clean the pins and get clean surface on them just from activity.

 

I followed ND's suggestion from another thread to swap ECU's and chips. The errors followed the chip, and I have not bothered to put the old ones back in after getting the new ones cloned and installed. Could it be corrosion? Certainly a possibility as I only swapped chips once before putting new ones in. I say we stick my old chips in Vince's car and see if it starts burping and farting - then we'll know beyond a doubt.

 

Even this is an old post I hope to get some response. I own a 1991 Mondial T with same engine like 348. My ECUs seems to work fine but I would like to have a spare new EPROM chip for each one of my ECUs. I am leaving in Greece, so the question is could you Programe two EPROM chips for my ECUs with the original Ferrari code so to keep them as backup in case of failure? If yes how much it could cost and what additional info is needed?

TIA, Aris