[s-cars] New ECU Chips from Mihnea - Ned's comment

FvAMI at aol.com FvAMI at aol.com
Mon Oct 20 02:21:47 EDT 2003


--
[ Picked text/plain from multipart/alternative ]
In a message dated 10/18/03 1:11:28 AM Pacific Daylight Time,
Q at IntendedAcceleration.com writes:

> On one occasion I have taken a pair of chips and used the same chips in
> 6 computers in a couple of hours.  Two of the six had the check engine
> light on in a matter of minutes.  So we deduced that the issue was the
> computers.  Of course in my two 91 200 cars the computers were hand
> selected to have the light out, but mind you the programs were the same.
> I've found that even a pair of chips for the 3B from Heinz Lehmann will
> turn on the light is some computers.
>
>

Ned,

I agree with the response that it may also be due to the characteristics of
each computer -- this does not make a check engine light acceptable.

Once again your example is a prime example of tolerances built into every
component on a vehicle. I would expect "Generic" code written by the likes of
Hanz Lehman to have enough knowledge of tolerances that it would not trigger such
an event.

My original prototype for the "eight" code selection device was thought to
have enough propagation delay that when the ambient temperatures exceeded a
certain amount and you were accessing a higher address it would cause the check
engine light to go off. The problem was solved by using a different EPROM from a
different manufacturer.  Upon careful analysis I found that the rise times
were faster on the previous devices causing higher switching currents ...
causing minor ringing on the output stage ... this in turn was solved by a more
elaborate bypass capacitance scheme.

Also some ECUs use the 80535 from either Siemens, Intel or AMD. There are
differences in I/O designs ... they all use different processes ... and there may
be some incompatibility between different suppliers of the EPROM and 80535.

Also cold solder joints add a nice bit of inductance causing some fascinating
waveforms.

But then again ... we could just ignore them.

Best regards,
Feico van der Laan



More information about the S-car-list mailing list