[xj-s] Marelli question

XJ-S
21

In reply to a message from CJ95 sent Thu 29 Aug 2013:

John:

Yes, the results are useable indeed. What I meant to say, more
precisely, is that the problem did not switch banks when the amps
or the coils were swapped. This vindicated both the amps and coils.

It’s a bit hasty to point to the ECU next, at least on this car,
and this is because it has a transplanted ignition harness. Yes,
you heard correctly: this is not the original harness.

That is why I have been stepping through this harness very
carefully with Gene. He’s got the S-57 wiring schematics and the
rest of the Marelli documentation I pulled from the shop manuals.
And we’ve stepped through this pretty carefully.

Never-the-less, the testing and inspection done thus far have
convinced Gene and I that the ECU must be at fault.

Your point on Gene checking the firing by looking at the coil wires
is excellent- when I first discussed the idea of the rotor
advancing away, he quickly informed me that he was looking at the
coil wire behavior. So this nixed the distributor argument. I
still like this, which is why I mentioned it earlier (even though
it can’t possibly be acting in this case, at least at this point in
the game).

Again, we are pointing back to the ECU.

Lockheed: your comment on teeth on the flywheel is well taken.
When I converted my car to a 5 speed, I had to fit the Marelli
sensor to the bell, and I did this by using a die grinder to notch
the bell for the sensor element itself, and then I created a flat
surface on the periphery of the housing for the mounting flange and
then drilling and tapping to mount the sensor. I was worried about
clearance with the teeth, and had read Bywater’s ‘unfussy’
clearance comment and specification. Never-the-less, I mounted
this sensor such that the clearance was at the far end of the
range. Guess what happened? When the car got warmed up (and the
bell housing expanded) the engine quit. When I got it home, the
next morning, it fired right up. I knew then that the speed sensor
gap had to be adjusted. I changed the clearance to be mid-spec,
and since that time, it’s run flawlessly.

Fun.

-M–
The original message included these comments:

Any result is usable. What happened to the spark when the

–
Mike, 1990 5.3 XJS Conv., 5-speed, SE-ECU, TT Extractors
Lakewood, OH, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

22

In reply to a message from mike90 sent Thu 29 Aug 2013:

Mike:

Hmm, I believe this is the first time this issue is mentioned
here. Someone has been there before, and there obviously has
been a damage of some kind. Do we know more?
What required the ‘transplant’?

At this point, we better wait until your test results come.

Steve–
The original message included these comments:

It’s a bit hasty to point to the ECU next, at least on this car,
and this is because it has a transplanted ignition harness. Yes,
you heard correctly: this is not the original harness.

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

23

In reply to a message from sbobev sent Thu 29 Aug 2013:

Steve:

All good questions, and perhaps when Gene reads this he will chime
in.

What I know thus far is that this car was bought on EBay, with the
ignition behavior disclosed. The prior owner had done a lot to
rectify ignition problems, including amps, caps, rotors, harness,
and even sourced a replacement used ECU, all to no avail. The cost
that must have been associated with that exercise makes me blanch.
The car was EBay’d and became Gene’s challenge.

My own opinion, having spent a great deal of time on a car which
came to me as a fire victim, with an ignition that was a complete
mess, and after patient, persistent work to return it to
operational condition, is that the Marelli system, in terms of its
parts and interconnections, is not that complicated, and he can
solve this with a little help.

Now the used ECU that is in the car is of unknown provenance (Gene
and I have reason to believe it is not the car’s original ECU), and
these are most often taken from vehicles in the wrecking yard.
Jags end up in the wrecking yard many times because of consequences
due to ignition faults, among other things. What I am trying to
say is, he may have gotten a bad ECU. Unless the party sourcing
the part can verify its integrity, we simply don’t know.

As I said, we’ll have a look later this evening, when I get a
chance to mess with putting this ECU on my car.

-M–
The original message included these comments:

Hmm, I believe this is the first time this issue is mentioned
here. Someone has been there before, and there obviously has
been a damage of some kind. Do we know more?
What required the ‘transplant’?

–
Mike, 1990 5.3 XJS Conv., 5-speed, SE-ECU, TT Extractors
Lakewood, OH, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

24

In reply to a message from mike90 sent Thu 29 Aug 2013:

Well, I wish this had been disclosed at the beginning.
Probably all of us who responded would have approached
things differently…

The fact that a ECU had already been swapped to no avail is
puzzling. With verified amps, caps, rotors, harness, and
ECU (keeping fingers crossed), what’s left?

Good luck tonight/tomorrow with the ECU. Let’s hope the test
provides the clue we need.

Steve–
The original message included these comments:

What I know thus far is that this car was bought on EBay, with the
ignition behavior disclosed. The prior owner had done a lot to
rectify ignition problems, including amps, caps, rotors, harness,
and even sourced a replacement used ECU, all to no avail. The cost
that must have been associated with that exercise makes me blanch.
The car was EBay’d and became Gene’s challenge.
The original message included these comments:

Hmm, I believe this is the first time this issue is mentioned
here. Someone has been there before, and there obviously has
been a damage of some kind. Do we know more?
What required the ‘transplant’?

–
'95 XJS V12 6.0L, saphire/tan
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

25

In reply to a message from sbobev sent Thu 29 Aug 2013:

Fair enough, except that Gene and I have been over this harness
with the S-57 wiring schematics in hand, and apart from some
details here and there, it’s a match.

Of course, those details mean something.

But, the fact remains, A bank is firing. B fires for about 3
seconds.

Looks like the basic details are intact.

Will get back to Gene and this list with the findings after I’ve
had a chance to mess around a bit after work.

-M–
The original message included these comments:

Well, I wish this had been disclosed at the beginning.
Probably all of us who responded would have approached
things differently…

–
Mike, 1990 5.3 XJS Conv., 5-speed, SE-ECU, TT Extractors
Lakewood, OH, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

26

But, the fact remains, A bank is firing. B fires for about 3
seconds.

So, is the engine sitting there running on the A bank?

That firing for 3 seconds is a puzzlement. If I were playing with
this car, I’d connect a VOM directly to the power supply into the B
bank ign amp and see what that voltage does during that 3 seconds.

– Kirbert

// please trim quoted text to context onlyOn 29 Aug 2013 at 13:50, mike90 wrote:

27

In reply to a message from Kirbert sent Thu 29 Aug 2013:

After reading all of this, I’m feeling extremely fortunate
to have so many of you considering my problem.

Concerning Kirberts thoughts on the white wire going to the
Lucus CPU: This is where its really dicey. There was a wire,
with alligator clips, spliced into the Marelli harness,
going to this white wire mentioned. Since this controls
fuel, I KNOW that this will have to be addressed once I know
I’m getting fire. I certainly feel that this car has sat for
a little over a year and a half, as the transplanted Marelli
ECU had a label on it of one of the VERY well known
suppliers, dated January 2012 (I think). This car has very
low miles on it. All of this gunk and grease mentioned
simply isn’t there. A very clean engine pretty much
everywhere. But, it has obviously been at the hands of parts
changers, verses ‘‘mechanics’’.

I feel sure that I may have to address fuel related
problems once I overcome the ignition related problems.
Other than that one splice, everything in the harness looks
virtually new. None of the hardened wires mentioned
throughout this thread. I will however, be checking the
sensors MUCH more closely now however.

Now the next favor, just for knowing. Can anybody check, and
see if there is ANY voltage on the 3 wire connection on the
A coil, other than the 12 volts? Again, when I turn the
switch on, I have 12 volts as I should on one wire, but 3.4
volts on the opposite wire- the tach signal wire. If I am
reading the schematics correctly, this simply shouldn’t be
(I don’t think). I am not much up on circuitry of ECUs, and
even though its on the A side, isn’t there still a chance of
some sort of weird backfeed, that somehow ‘‘may’’ be effecting
the proper operation of this car?

One hurdle at a time, as time allows, I will nurse this car
back to life (I hope)–
Gene Holtzclaw
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

28

In reply to a message from Gene Holtzclaw sent Thu 29 Aug 2013:

Gene:

I sent you an email on this, and here is what I wrote:

…I DID swap out the ECU, and…it works. Fired right up, smooth
exhaust, both banks.

I am convinced that we need to open the connectors at the B bank
amp and at the ECU, and look at the wires themselves where they
enter the pins that are located in the connectors. Then, I think
we should ohm out the cable itself on those two lines. Finally,
I’d like to know for sure that there is 12v at the B amp all during
the cranking process (beyond the 3 second point at which B quits
firing).

I think Kirby’s comment bears a look: I can tell you that during
the diagnosis of a much worse situation on my car, I eventually
found that the B bank +12v line to the amp was broken, in the
harness, in the section that runs from the amp to about where the
harness approaches the front of the block. It was broken well in
the middle of that section, in what was an otherwise unmolested
jacketed bundle, with absolutely no evidence of mistreatment.

I am beginning to wonder, now, if you are suffering an intermittent
+12v feed to that amp in this harness.

In any case, it’s got to be between those two ends in the harness,
because A works fine.

I will return your ECU tomorrow, and I will throw in a spare
reluctor sensor that I removed from the crank end of my engine when
I was diagnosing it back to life (guess wrong on this sensor, I
think this one is still good. You can measure the resistance on it
yourself).

This sensor is the same P/N, whether for the crank or the flywheel.

We WILL solve this, Gene!

-M

P.S.: that coax you are looking at is almost surely the Lucas
trigger input, from the Marelli ECU. On my car, this evidently had
proven problematic, because there is a separate line spliced into
the ECU connector and running directly back, on its own, to the
Lucas fuel ECU. But problems in THAT line are not responsible for
the ignition problems. When the ignition is fixed, we move to
fuel, and slay that one, too.–
Mike, 1990 5.3 XJS Conv., 5-speed, SE-ECU, TT Extractors
Lakewood, OH, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

29

In reply to a message from mike90 sent Thu 29 Aug 2013:

OMG!

When you envolve a nutty PO that changed out the harness on a low
mileage car, absolutely anything is possible!

You have cleared the coils, amps, pickups, and ECU…You know what
that means!?!–
The original message included these comments:

It’s a bit hasty to point to the ECU next, at least on this car,
and this is because it has a transplanted ignition harness. Yes,
you heard correctly: this is not the original harness.

–
John. '95 XJS 6.0L convertible. Southlake, TX
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

30

In reply to a message from CJ95 sent Thu 29 Aug 2013:

Not really…I get what Mike is saying. Start at the B
coil, and work back, to see where I am losing signal?–
Gene Holtzclaw
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

31

In reply to a message from Gene Holtzclaw sent Thu 29 Aug 2013:

Has anyone thought of advising Gene to go
over the ignition looms grounding point
on the left hand side of the firewall? If
that is loose one might get the sort of
behaviour you are getting. Also a long
shot but this contact if loose might have
chewed through your rotor (it’s a tiny
hole in one of the corners of the star-
shaped hole on the underside of the
rotor). But check that ground first and
as Kirby says disconnect the dashboard
instrument panel (or just the tach).–
'91 xjs v12 5.3 early facelift in BRG
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

32

In reply to a message from Pele sent Thu 29 Aug 2013:

Pele:

Actually, based on the symptoms, I have been thinking and advising
Gene to not only check continuity on each connector to connector
point, but also check continuity to ground, as I am concerned that
somehow the firing signal to B amp is grounding out somewhere.

Gene knows the ground point on the A bank manifold, and he knows
the ground points on the firewall.

It would be good to verify all the ground points once more, too.

Here’s part of what I wrote to Gene by email after doing this test
last evening:

I am not convince that the flywheel sensor is at fault- I know the
stuff that’s been written on the 'list, esp. from Lockheed, and I
have yet to pull the connector while the engine is running, but I
do have one data point of experience with this: When I did the 5
speed, I had to refit the flywheel sensor to the new bell, and
because the bell was not machined to accept the flywheel cover
plate from the Marelli/TH400 setups (and this cover has the
mounting boss for the sensor), I had to fashion my own mounting. I
did this by using a die grinder to grind out a semi-circular slot
in the bell into which the sensor could be mounted, and then I
filed a flat on the circumference of the bell over that ground out
slot, and drilled and tapped screw holes for the mounting screws
for this sensor. Of course, the clearance between the sensor face
and the gear teeth had to be established, and I first set it at the
far end (large gap) of the spec. When I ran the car for the very
first time after getting all this back together (had pulled engine
and trans in order to remove TH400 and fit the Tremec), the car ran
great right up to the point where it was fully warmed up. Then,
abruptly, the engine died. I had to flatbed the car home (first
time!). The next morning, on a lark, I hit the key, and it started
right up! I knew then that this sensor clearance was set too wide,
and the thermal expansion of the bell moved it out of range of the
flywheel teeth. I reset the clearance to mid range of the spec,
and the car has not missed a beat since.

So, when the OEM manuals say those are primary sensors and the car
won’t start without them, I believe it.

Moreover, the argument used by Lockheed and subscribed to by others
is going like this: the front sensor gives TDC for the A bank, and
the flywheel is giving engine speed, so that B bank firing can be
computed. This means that they think that firing on A comes
directly from the trigger points on the crank sensor. IT cannot be
this way; first, the sensor is set for some 8 degrees After TDC
(ATDC)! You can’t fire the A bank cylinders ATDC! Why is it set
this way? Well, because the ECU needs time to compute the firing
points, and setting it JUST AFTER TDC gives it all the time it
needs. It MUST use a speed signal to compute A bank firing points,
BECAUSE these must have advance in them as well as the B points.
(see next post, because I am out of chars on this one)–
The original message included these comments:

Has anyone thought of advising Gene to go
over the ignition looms grounding point

–
Mike, 1990 5.3 XJS Conv., 5-speed, SE-ECU, TT Extractors
Lakewood, OH, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

33

In reply to a message from Pele sent Thu 29 Aug 2013:

Part II:

So, logically, this notion that the rear sensor failing would
somehow permit A to continue firing and B to not just doesn’t hold
up to scrutiny. Moreover, I have an actual experience of losing
the flywheel sensor, and the car didn’t simply drop in power; it
shut down, like, right now. So, these are primary to the ECU
functioning, period.

What I think you have in your system is a wiring fault, and I think
we will find it, but it will be potentially very tricky, but can be
found. When I was diagnosing my car, I too, thought, well, maybe I
had a sensor failure, and that is when I refitted a crank sensor.
But the eventual cause of my B-bank not firing at all lay in the
harness. And it was very very subtle: the harness section that
runs from the front of the engine, through the flange above the
radiator, to the amps was where the fault lay. Not at the
connectors, and not in the section that lay on or behind the
engine, but in this very nicely preserved section. When I peeled
the insulation of the bundles back (from amp connectors on back),
and was carefully, by hand, inspecting the wiring, I noticed an odd
look to one of the wires- the insulation looked whiteish in a very
small area (you know how plastic, when it is heavily deformed can
give a whiteish cast? Like this). I found that the conductor
INSIDE the jacket insulation was broken! And this cable section
had not been kinked or otherwise abused. And it was in the factory
oversheathing as well. This was a materials quality failure from
the factory.

The symptom I had been chasing: no +12v power to the B amp.

So, when this ECU reaches you, I would recommend that you put a DVM
on the +12v at the B amp, and have someone crank the motor, and see
if this +12v holds. If it does, do the same measurement at the B
coil connector: make sure you are getting a steady +12v at the B
coil. If both those hold up, and you are still seeing your
symptoms, then, look carefully at the other two inputs to the B amp
(the firing signal lines from the ECU) and the output of the amp,
which is the fire signal to the B coil.

Pele suggests carefully going over the ground points, and I agree,
this is sound advice, and could be responsible for this behavior
(esp. ground on B amp and B coil).

One of these is the problem, I am sure of it, now. We have
vindicated everything else that can inhibit B bank firing.

-M–
The original message included these comments:

Has anyone thought of advising Gene to go
over the ignition looms grounding point

–
Mike, 1990 5.3 XJS Conv., 5-speed, SE-ECU, TT Extractors
Lakewood, OH, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

34

In reply to a message from mike90 sent Fri 30 Aug 2013:

Also don’t forget the braided cable that
goes from the block (or sump) to the
crossmember and body at the front left
side. I believe the combination of loose
firewall ground and low quality contact
of this braid leads to the rotor
failures. There is a special ‘‘toothed’’
washer that goes between the body and
braid to ensure good contact and if it’s
missing…well another headscratcher for
you.
I believe Kirby has driven this point
home countless number of times by now but
it never hurts to repeat it.
Also in the wiring schematics you have
the exact test values for the ignition
system like what should you get at coils,
amps, sensors (700 ohm on both I think)
etc so go through that carefully. It
takes about 5 minutes to go through all
of them with a vom.–
'91 xjs v12 5.3 early facelift in BRG
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

35

In reply to a message from Pele sent Fri 30 Aug 2013:

Pele:

Thanks for those tips. Gene has the documentation that gives the
expected indicated readings for the sensors.

The ground strap to which you’ve referred is a good thing to check,
too. I am sure Gene will have a look.

Interesting theory on the rotor failures being related to
poor/loose grounds…I don’t think I’ve heard that one before your
post.

Thanks-

M–
Mike, 1990 5.3 XJS Conv., 5-speed, SE-ECU, TT Extractors
Lakewood, OH, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

36

What I think you have in your system is a wiring fault…

Agreed. And since the amps and coils have been confirmed to be OK,
I’m betting on either power or ground. One or the other has a weak
connection that looks good when checking the circuits but actually
has high resistance and drops too much voltage under load. These
cars are full of connections like that.

– Kirbert

// please trim quoted text to context onlyOn 30 Aug 2013 at 6:52, mike90 wrote:

37

first, the sensor is set for some 8 degrees After TDC (ATDC)!

From this fact alone, we can conclude that this trigger is not used
to fire 1A – even good electronics cannot predict the future.
Rather, this trigger would be used to fire the NEXT pair of cylinders
on the A bank – as well as the FIRST pair on the B bank – after
some computed time delay, or perhaps after some computed number of
teeth on the ring gear pass the flywheel sensor. The 1A cylinder
firing will be keyed off of the THIRD trigger, the one that happens
about 112 degrees BTDC.

If anyone knows how many teeth are on that ring gear, we could
hypothesize how many it has to count to get from 112 degrees BTDC to
10 degrees BTDC. Not that it would make any difference to anyone.
It’s also going to have to count off 42 degrees to fire off a pair of
cylinders on the B bank.

If it DOES count teeth – which I doubt, it more likely just uses the
tooth passing rate to compute a time delay – then some missing teeth
would retard the timing. If it’s computing a time delay, perhaps
missing teeth would also retard the timing, perhaps by even more, or
perhaps it’d just confuse the ECU and cause it to switch time delays
back and forth by a factor of 2 very quickly, which of course might
mess up all sorts of stuff.

These things DO lose teeth. I’ve been in contact recently with an
owner who is aware of ONE missing tooth – but it’s on a Lucas car,
so it doesn’t have a flywheel sensor to worry about. It just
occasionally jams the starter and he has to get out an manually move
the crank a hair to get past it.

– Kirbert

// please trim quoted text to context onlyOn 30 Aug 2013 at 6:47, mike90 wrote:

38

In reply to a message from mike90 sent Fri 30 Aug 2013:

Ask me how I know (wipes silicone off
fingers)…–
The original message included these comments:

Interesting theory on the rotor failures being related to
poor/loose grounds…I don’t think I’ve heard that one before your
post.

–
'91 xjs v12 5.3 early facelift in BRG
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

39

In reply to a message from Gene Holtzclaw sent Thu 29 Aug 2013:

It means you are down to the harness. Unfortunately, at
this point it could be broken wires, bad grounds, cracked
insulation making other grounds/connections, bent
connector pins, pin sockets pushed in, missing grounds.
The list of what a PO can screwed up is unending.

Like you are saying, Gene, the harness needs to be gone
over wire by wire, connector pin by connector pin. I
would recommend not even trying to crank it till you have
the harness squared away. Something is shorting or open
somewhere.

Lure some friends with a cooler of beer and get them
working with jumpers, the wiring diagram, and a
multimeter!–
The original message included these comments:

Not really…I get what Mike is saying. Start at the B
coil, and work back, to see where I am losing signal?

–
John. '95 XJS 6.0L convertible. Southlake, TX
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only

40

In reply to a message from Kirbert sent Fri 30 Aug 2013:

Agree-

I did know the tooth count once, and maybe even have it written
down somewhere.

The reason: when you fit an manual box to a Marelli engine, and you
fit a new flywheel, you better be sure that the tooth count is the
same, or the ignition (and injection, too) will be operating
erroneously. I remember counting the teeth on the flex plate and
then the teeth on the flywheel, to verify this. Why? the trans I
put in was part of a Keisler kit, a very early one, and was ordered
for a pre-89. I bought it from a fellow lister who never got
around to fitting it. I learned many things about the differences
between >89 and–
Mike, 1990 5.3 XJS Conv., 5-speed, SE-ECU, TT Extractors
Lakewood, OH, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

// please trim quoted text to context only