Citroen SM injector points

Hi Graham,

well using a dwell measurement for trigger contacts is also something that I recommend in my compendium. I can tell you that the ECU ONLY looks for falling edge of contacts. So as long as you do not have overlaps in closing contacts, The ECU does not care when contact opens and for how long. However if contacts are unevenly worn it could happen that they overtake each other in 8-cylinder trigger contacts.They only have 90° distance. I have only seen such a case once.

Volker,
I only know from my own experience and that of others on the forums that worn trigger point rubbing blocks CAN cause a rich mixture. I also understand why, in theory, that should not be the case.

In practice, something does cause it. The Porsche forums have talked about trigger point bounce. Perhaps as the rubbing blocks wear (and closing angle increases from 100deg to 150deg and more) there is less spring pressure and therefore the points start to bounce? Just a theory of why that could happen. But if it did, the triggers could be opening and closing several times in half a revolution. What would the ECU make of that?

Don't want to prolong this. Just passing on my experience

Howdy all,

Thanks for all the feedback.

I had an hour to spend on the car today with my gas analyser.

Note: I am using a Sun MGA 1200 analyser re-calibrated 6 months ago. No doubt the exhaust system will contain a fair bit of combusted crap after the car had run excessive rich for some time.

At idle, I could not get the CO reading at idle to go under 7/8%.

Smokes a bit, some miss on acceleration. Oddly enough I substituted the MAP sensor from a late MB unit, and the thing actually ran better and a little leaner – even at idle? (a little spurious and unconvincing result though)

I appreciate the link to ECU pin chart from Nordfisch. As suggested I’ll take all readings and report back.

Regards,
Rhys

Hi Rhys,

not being able to go below 7-8% means a far too rich mixture. Assuming that you made sure that idle-run contact reaches ECU, this could be caused by

• false air
• injectors leaking ar wrong spray
• broken ECU
• engine temperature sensor wrong or disconnected
• ignition timing is said to have no effect on CO
• MAP sensor (if not completely leaking should not effect CO in IDLE-RUN as well as it is superseded by potentiometer

So yes, you should always measure on ECU connector as Norbert pointed out. That's what I made appendix B for.

Hi Graham,

I did have the experience of what happens to a worn trigger contact during this year myself. 2 of 8 cylinders that were still running when cold, ceased to work when engine got warm.

When plastic cams get worn, the cam on distributor axle will not be able to push them open any more. Should one have some dirt or whatever on either axle or plastic cams, then it would be possible that they open and close several times: i.e. bouncing. That would not really confuse the ECU. It has a Flip-Flop reading these contacts and normally interpretes each closing as a signal from T.C. But it would need more than one contact to bounce as the flip-flop needs one bouncing contact after the other to change its state.Just one bouncing contact would not create that effect. Only if we would see bouncing on both contacts in alteration, it would create an additional injection pulse and it would also wrongly interprete engine revolutions. A flip is like two push buttons. Pushing one button sets the output, pushing the other resets output. In between you can push the first button as often as you like.

On the otther hand if we would come close to 180° closing angle, the 2 contacts on a 4/6 cylinder might overtake each other. However there is a huge tolerance on this injection timing. Fuel just waits before it gets sucked into cylinder.

As a summary: A bouncing would only create additional injection pulses if you had it in such way that it would be bounces in alternation on both contacts. I would see that as rather unlikely.

An 8 cylinder is actually like 2 4-cylinders in parallel, both in T.C. and in ECU.

Hi Volker,

I am proceeding with checking values from the ECU terminal block.

I’ve dismantled an old (Volvo) ECU to manufacture a break-out box as I find the multimeter leads on the ECU terminal contacts leaves some margin for error.

In the mean time, I’m having some difficulty with the pin out sheet with regards to pin 17 (the idle contact on throttle switch). The instruction is to measure against pin 12, but it would seem to me that pin 14 would be the correct pin. Pin 12 is indeed the central distributor trigger point, but with the ECU disconnected, their paths should not cross? (poor extract of Citroen wiring diagram attached).

Do you agree - or am I seeing this wrong?

On a separate issue.....

I understand that the idle switch activates the ECU to run on lean idle mode independently of the MAP sensor. When I remove the MAP sensor of an idling Volvo P18000ES at the workshop (both vacuum or electrical connector) it dies. What is the relationship of the MAP sensor to the idle circuit?

Cheers,
Rhys