1987 3.2 Carrera Backdate project - Sladeys hotrod

[Bruce M]

If you can establish serial comms with the ECU the microcontroller “brain” is not 100% fried. It may have lost the two output channels used for the injection pulses but perhaps just the high current semiconductors that amplify the microcontroller output. It should be repairable, in that situation.

[jjeffries]

I just caught up with this, read the last 8 pages or so. I was glad to read when the noid light had been utilized; very useful information. I went through an similarity dispiriting diagnostic process with my erstwhile 87 924S with the same Motronic (but no Camens), a story for another time (or not at all), but when it was finally deduced and resolved, I realized how much double/triple back-tracking I’d done on tests I’d already verified. In other words, I hadn’t believed what the car and my diagnostics were telling me and as result, was spinning in circles. When this is fixed, you’ll have a much more robust knowledge of the system(s) and a far greater belief in your own abilities.

Best, John

[Jonny Hart]

Isn't the number one cause of laid up 3.2 non-starts gummed up injectors? Could it be that the Canems unit pulse width isn't wide enough to open the blocked injector, whereas holding the injector to ground does?

I realise you tested this with the noid LED, but I would revisit that.

[sladey]

If you can establish serial comms with the ECU the microcontroller “brain” is not 100% fried. It may have lost the two output channels used for the injection pulses but perhaps just the high current semiconductors that amplify the microcontroller output. It should be repairable, in that situation.

Thanks - that’s what I’m hoping as well


Sent from my iPad using Tapatalk Pro

[sladey]

Isn't the number one cause of laid up 3.2 non-starts gummed up injectors? Could it be that the Canems unit pulse width isn't wide enough to open the blocked injector, whereas holding the injector to ground does?

I realise you tested this with the noid LED, but I would revisit that.

Hi Jonny

It’s hardly been laid up - I was driving it before Christmas and drove it a fair bit through last year. Also we did re-try the noid bulbs after we’d grounded the connection, and we tried it on 3 different injectors. Not a dickie bird from any of the injectors.

With regard to the injectors themselves, during previous work I had them all sent away and tested and cleaned. That will be probably 10 years ago but they aren’t unchanged from 1987.


Sent from my iPad using Tapatalk Pro

[sladey]

I just caught up with this, read the last 8 pages or so. I was glad to read when the noid light had been utilized; very useful information. I went through an similarity dispiriting diagnostic process with my erstwhile 87 924S with the same Motronic (but no Camens), a story for another time (or not at all), but when it was finally deduced and resolved, I realized how much double/triple back-tracking I’d done on tests I’d already verified. In other words, I hadn’t believed what the car and my diagnostics were telling me and as result, was spinning in circles. When this is fixed, you’ll have a much more robust knowledge of the system(s) and a far greater belief in your own abilities.

Best, John

Thanks John and yes you’re right. I would still like to drive it on my forthcoming trip though if possible - just in the hands of Canems now


Sent from my iPad using Tapatalk Pro

[Nige]

Isn't the number one cause of laid up 3.2 non-starts gummed up injectors? Could it be that the Canems unit pulse width isn't wide enough to open the blocked injector, whereas holding the injector to ground does?

I realise you tested this with the noid LED, but I would revisit that.

When we grounded the injectors they put enough fuel into the engine to chuff into life, if only slightly, so I can't see them being blocked. I can't imagine all would be gummed up enough to cause a total no start, it doesn't even try, just cranks over.

Every test we did just pointed to no injector signal/pulse from the ECU. You can't feel them pulsing either when you crank it over.

My own 3.2 sometimes has a sticky/gummed injector but it will always start and clear itself.

[Gary71]

Out of all the 911s out there I think this one has had enough fuel flushed through it

[Bruce M]

Is Canems a 1-man-band?
Internet suggests he is good & helpful but hard to get hold of. I know I’d have the case open & oscilloscope probes out to trace back the signal, but that’s not for everyone.

[Bruce M]

One idea (a stretch).

The ecu has a number of earth pins. These may be linked internally within the ECU or may be a “star” earth design where each earth pin has a separate function and they are all linked at one place (as per the wiring diagram, outside the ECU). This prevents earth loops / noise.

If the injector circuit uses only one of the earth pins & that has failed (outside the ECU), then Potentially the injectors could not fire.

I’d suggest disconnecting the harness plug from the ECU & check all of the earth pins in the plug have continuity back to chassis earth. The ECU manual & wiring diagram give you the PIN numbers.

[sladey]

Thanks for that Bruce I’ll give it a go

[sladey]

Actually thinking about it Nige and I did that - we did don’t we Nige

[sladey]

But I’ll have another look

[sladey]

Is Canems a 1-man-band?
Internet suggests he is good & helpful but hard to get hold of. I know I’d have the case open & oscilloscope probes out to trace back the signal, but that’s not for everyone.

I’m game for anything but need guidance and can’t spell sillyscope

[sladey]

Ok so ran them again. They all worked except pin10