L-Jetronic ECU fragility? was V6 stalling - slightly off topic!

["John Fielding" <johnf@xxxxxxxxxxxxxxx>]

Hi All,

George Graves today on the AD made a comment about L-Jetronic ECU's being susceptible to
probing with a test lamp.  He intimated this can cause damage to the internal electronics.

Well, having repaired numerous L-Jet and Motronic boxes and seen the sort of circuitry
used in them I would very much doubt Georges comment.  (I also design and manufacture
automotive electronic modules so I have also been aware of some of the problems you can
experience).

Bosch are not idiots when it comes
to designing vehicle electronics and they do a comprehensive "what-if" scenario as part of
the design effort, like most vehicle electronics designers.  A 12V battery is a hostile
environment, one semiconductor manufacturer calls it "the supply from hell" as it has so
many glitches and transient spikes due to the many inductive circuits being powered from
it.  A Defence Standard specification exists for 24V vehicles.  One paragraph I remember
is the sort of transient caused by simply blowing the horn.  Voltages typically to be seen
on the 12V supply can exceed 60V spikes!  On fuel injected vehicles I have measured spikes
as much as 300V due to the injectors being switched on & off.  Normally the battery is
considered to act like a large capacitor and should damp these transient to neglible
levels.  However, lead acid batteries are rather poor when they have to deal with high
frequency transients, they appear like a soggy resistor with a lot of series resistance or
impedance and do little to clamp the spikes.  The further away from the battery terminals
the worse this becomes.  As the wiring in a vehicle is a tightly bundled set of parallel
wires running for considerable distances, if a wire carries a transient generated current
the wires also running along side this one will pick up similar voltages.  Because of this
great care and additional circuitry is normally needed to prevent the transients being
induced into other lines from reaching damaging proportions.

The Bosch L-Jetronic have a wire which connects to the ignition coil for triggering the
fuel injection ECU.  In the L-Jetronic the connector pin is normally #1.  This pin
experiences transient peaks as much as 400V every time the coil is fired.  In the
L-Jetronic engine harness this is carried in a shielded cable to reduce the chances of the
violent pulses getting into the other sensitive inputs of the ECU.  But this line also
runs through the main vehicle harness from the coil -ve terminal to drive the rev counter
on the dash, this wire is often a plain unshielded type.  It doesn't take too much to
appreciate that this same transient is going to pop up in all sorts of other places
throughout the vehicle.  On later versions of the Motronic a separate tacho driver pin is
provided which only has a 5V peak, these later rev counters cannot be safely driven off
the coil directly.  Porsche, BMW and Alfa amongst others also provide a pin on the
Motronic ECU to drive the "fuel efficiency" indicator on the dash.  This is simply a mimic
of the injector pulses and the needle indicates low mpg when the engine is idling and the
needle swings further across the scale when the injector pulses get longer under heavy
acceleration.

All input lines to the L-Jet ECU run through transient clamping devices.  This means that
short of applying 12V positive directly from the battery they are able to withstand some
very big transient spikes.  This is a common practice with most ECU's.  The designers are
very well aware of the possibility of the incorrect voltage being applied due to
accidental connection or other means.  Those inputs normally only expected to experience
5V signals, such as the AFM or TPS analog to digital converters (A/D) have additional 5V
clamping devices.  Motorola, a typical ECU processor manufacturer, guarantees the A/D
converter inputs to withstand at least 30V without damage even if no clamping devices are
used.

A test lamp with, say, a 5W lamp can only pass about 1/2A maximum, the clamping devices
used are easily able to clamp many times that current indefinitely.

The sort of things that can damage the ECU are fortunately quite rare.  One I know of is
when the alternator regulator goes AWOL and jams the alternator into maximum charge mode.
This is a catastrophic failure mode and usually the battery boils and then goes open
circuit.  This takes quite some time to happen.  However, when the battery finally lets go
the rail voltage then rockets to maybe as much as 80V.  What usually happens then is that
the main digital power supply regulator in the ECU goes pop and the 5V microprocessor
devices instead of being fed a nice stable 5V gets supplied anything up to 60V or more.
That sort of overvoltage will definitely cause extensive damage to the electronics
throughout the vehicle, not just the ECU.  I recently saw a  4 year old Hyundai which
suffered this failure, everything from the window motors, lights, ignition, fuel injection
ECU etc etc was fried.  The vehicle was eventually written off by the insurers as the cost
to repair exceeded the book value.  The faulty alternator was a small fraction of the
total
cost required, the ECU was a NLA item as the vehicle was a limited edition version and the
factory had no stock of the ECU. A perfectly good vehicle went to the knackers yard!

Most ECU's can also withstand a reverse battery connection for a short time.  I once had
two Renault ECU's in for repair.  The stupid owner had already blown the first one, so he
took the other vehicles ECU and plugged it into the car.  Only after this also made a loud
bang did he check the battery was connected the right way, of course it was reversed!
Fortunately the ECU designers had the forsight to install a reverse polarity diode and a
weak piece of track which acted like a fuse.  The repair was not difficult and the owner
got off lightly!

John
Durban
South Africa
Alfetta 1.8L turbo
--
to be removed from alfa, see /bin/digest-subs.cgi
or email "unsubscribe alfa" to majordomo@domain.elided