[two-point-o]

This old chestnut again (know I've been here before for this)...
I've got another Spider with the ole start/run for 3-4 sec./stall problem. Each time I come across this it seems to have a different cause, this one really has me stumped. Let me start by saying I've read the Bosch L-jet link provided on similar threads and checked everything accordingly, oh yes; there is fuel in the tank. Now let me tell you what I've checked (don't laugh): * all air intake hoses- large and small- no air leaks,
* popped the AAV in the freezer and oven and replaced it with a known good one- all okay but still start/stall,
* tested continuity of leads off every sensor back to the pins on the large connector at the ECU to rule out broken wire in the harness- all okay,
* swapped out the injector ECU and spark ECU with ones from a good runner which also came in for service- no change- put them back in the other car and it runs like a champ,
* replaced both relays located at the ECU,
* replaced the Air Flow Meter from the other car- no change,
* tested both coolant temp sensors for varible resistance within spec- all is good but no run,
* replaced the throttle position sensor- no change, reinstalled it using an ohm meter to insure correct positioning
* replaced (just for fun) the altitude compensator, no change,
* replaced both flywheel sensors, no change
* replaced the battery (reaching here) with one with higher CCA
* replaced the vacume advance device at the spark ECU
* checked/cleaned all grounds and connectors (even added an extra braided ground strap from head to firewall knowing how Alfas like grounds)
* checked wiring at coil, all fine and a nice fat spark...

The car is in near perfect tune and sounds great for the few seconds it runs. It has excellent spark and fuel pressure at the rail is spot on. Cold start injector circut is working as is fuel pump until the engine stalls. Just no injector pulse as confirmed by the noid light. Please guys; what am I missing? The last time I delt with this it turned out that someone had installed the wrong fuel pump relay at the ECU, not the case here. Someone must know the answer. This thing is becoming my White Whale! Help!!!
Cheers,
Two Point Oooooo

 

[papajam]

I think one must consider that the injector pulse dropping out may not be the cause of the engine stall but the symptom. I'd try the following;

Hook up a voltmeter to coil positive
Install a spark tester on one sparkplug wire
Install noid light

Start the engine and observe the above.
If the voltmeter drops before the engine stalls, suspect power feed to the coil (ignition switch for example).
If voltage remains good but spark drops out, suspect ignition system (to include a shorted tach).
If both voltage and spark remain good, suspect fuel management.

It's difficult to diagnose a problem without knowing where to look.

 

[two-point-o]

Thanks Jim,
Diffucult, yesh. I did as you recommended and observed that the coil voltage remains steady up to and after the engine stalls and the spark continues to look fine until the engine is at rest. The noid light to the injector, however, never flashes an injector pulse. I have test the noid light via the CSI to validate its function and its fine. I can supply voltage to the injectors and they 'click' away happily and spray fuel - when instructed to do so. No joy here yet. What sensor could tell the ECU to keep the injectors off? or ??? Hope solving this will add to the considerable knowledge of these L-Jet systems as I see a lot of threads on this issue. Thanks
Ken,
'91 164L (daily driver)
'74 GTV
'67 GTV Jr.

 

[papajam]

It would appear then that the coil power source and the ignition system are in good order. That leaves fuel management. The fuel ECU though, having been replaced with a known good unit, can apparently be ruled out as well.

The fuel ECU pulses the injectors by providing a ground. This means that one of the two injector wires must be hot with key on. Is it? Injector power is supplied from main relay terminal #87.
Also check for ignition system pulse at fuel ECU pin #1. This signal comes from ignition coil terminal #1 (negative side).

 

[Elio Comello]

If you read the factory workshop manual PA3748, Fuel System 04-2 (Sect 04page 2) DESCRIPTION, you will find a list of all the inputs into the FI Control Unit.

One of them is "Starter motor signal".

I believe this a signal (wire) from the ignition switch to pin 2 of the female connector of the FI control unit.

Intuitively it would appear that the injector pulse is inhibited when the starter motor is being cranked.

There is not much of substance said about this, but if you look at the wiring diagram LOL... Section 40-16 you will see a wire connected from the ignition switch to the Injection control unit.

I believe that in the Start key position: AVV ("Avviamento" in italian) that the injectors may be silenced.

I may be all wet, but I think that you may have to look at the ignition switch and the signal to the FI control unit terminal mentioned above. You have ruled out everything else......

One last thing, please note that there is differences in FI contol units 82 to 85 (Gold bosch tab) vs 85 to 89 (green bosch tab). However I believe that they only differ in ability to drive the VVT and possibly inertia switch. So substituting like with like is important, assume you did this for your 88.

Best regards, Elio

 

[two-point-o]

Thanks guys,
I'm finding that both pins on the injectors are hot. If I use a jumper wire and ground one pin the noid light flashes and the injector pulses, indicating the ground is not being provided by the ECU, no? Pin 2 at the ECU shows about 11volts from the ing switch and pin 1 has full battery power with key on. Also pin 87 on the main relay has full power. Thanks for the heads up on the different ECUs as I had exchanged one in this '88 for one from an '82. I have since replaced it with a green tagged one from an '85. Still no go. Also switched out the ingnition switch with no success but it was a red/brown/green/2 blacks one and the one in the '88 is a red/brown/black one. Only hooded up the main 3-terminal connector on the switch to see, but no go. Hate to send this car to another shop and concede defeat (and watch the owner get blead dry). But I'm out of theories. Any other suggestions? Please...
Cheers,
Ken
'91 164L
'74 GTV
'67 GTV Jr.
'79 Lancia Zagato
2 Citroens, 2 Panhard Dynas, VW bus
divorced (go figure)

 

[Elio Comello]

Pin 16 and 17 as well as pin5 of the FI CU should be connected to ground as well as Pin 2 of the Ignition ECU.

Don't forget that the system needs at least 10.5V during cranking for the C/U's to wake up (new battery or not).

Don't know what else to tell you.

Elio

 

[papajam]

...indicating the ground is not being provided by the ECU, no?

Correct. The ECU is not providing a ground to pulse the injectors.

...and pin 1 has full battery power with key on.

Pin 1 is the pulsing signal from the ignition system (coil negative). Without it, the fuel ECU thinks that the engine isn't cranking or running and will not pulse the injectors. Please check the voltage at pin 1 with the engine cranking (a dwell meter can be used here as well). You should get a pulsing voltage reading.

The grounds to which Elio refers are attached to the cam cover just to the rear of the auxillary air valve.

 

[two-point-o]

Well you were spot on about a ground fault. It turned out to be a ground wire from the tachiometric relay to the ECU. Not sure how that became detached or even where it was supposed to ground except through the ECU (my wiring diagram is not clear on this). Splicing into that black wire and grounding the other end to the body at the ECU and the car started right up and is now running fine. Thanks again for the help. I reckon there'd be a lot fewer Alfas on the road and a lot more Italian lawn ornaments around without you guys. Cheers, Ken

 

[papajam]

That's excellent news, Ken! Congrats and thanks for the update.

 

[ghnl]

Yeah, that is good to hear. Thanks for sharing the solution - this will add to the L-jet knowledge base.

Do you recall the color of the ground wire from the relay to the ECU? (black is usually ground). Do you recall which terminal on the relay it is?

 

[Mborras1]

Good morning to all on this thread but I'm having the exact same problem went through the same exact process stated above and believe it's the ground wire. We're would I find the below relay that two-point-o ended up grounding?



It turned out to be a ground wire from the tachiometric relay to the ECU.

 

[Mborras1]

Good morning to all on this thread but I'm having the exact same problem went through the same exact process stated above and believe it's the ground wire. We're would I find the below relay that two-point-o ended up grounding?



It turned out to be a ground wire from the tachiometric relay to the ECU.

Is it the wire that runs from 87 black/white to the negative ignition post?

 

[doctorb]

Mborras.
I see we are chasing similar things from my other thread. I also see this is an old thread, but useful.

I call the tachiometric relay the "fuel pump relay" or "drive relay". On my 84 spider, the black/white wire on that relay is on pin 31b and is the tachy signal from coil (-) . Do not ground that wire. The ignition ECU switches that wire to ground during cranking/driving. That same coil terminal (-) should also have a shielded yellow wire that is the tachy signal to the FI ECU pin 1. This yellow wire carries the tachy signal that is massaged by the FI ECU to provide the FI pulse. (Tomorrow, I will spend some time figuring out if I get the tachy signal back at the ECU.)

In re-reading two-point-o's post, I believe he spliced and hard grounded the black wires on pin 85 of the main relay (which is also connected to pin 31 on the fuel pump relay). My schematic has those black wires grounded through the FI ECU. If those black wires were not grounded, the FI ECU would not get power, nor would the injectors get their 12V supply (via the main relay), and the fuel pumps wouldn't give their initial pre-crank kick. So, I think there would have been bigger issues if those black wires weren't actually grounded through the ECU. I do not see how two-point-o would have measured 12V at the injectors if these wires were not grounded, since the main relay would not have been energized. I also don't know if we're all talking about the same car/jetronic. Do you have a schematic?

 

[Mborras1]

Mborras.
I see we are chasing similar things from my other thread. I also see this is an old thread, but useful.

I call the tachiometric relay the "fuel pump relay" or "drive relay". On my 84 spider, the black/white wire on that relay is on pin 31b and is the tachy signal from coil (-) . Do not ground that wire. The ignition ECU switches that wire to ground during cranking/driving. That same coil terminal (-) should also have a shielded yellow wire that is the tachy signal to the FI ECU pin 1. This yellow wire carries the tachy signal that is massaged by the FI ECU to provide the FI pulse. (Tomorrow, I will spend some time figuring out if I get the tachy signal back at the ECU.)

In re-reading two-point-o's post, I believe he spliced and hard grounded the black wires on pin 85 of the main relay (which is also connected to pin 31 on the fuel pump relay). My schematic has those black wires grounded through the FI ECU. If those black wires were not grounded, the FI ECU would not get power, nor would the injectors get their 12V supply (via the main relay), and the fuel pumps wouldn't give their initial pre-crank kick. So, I think there would have been bigger issues if those black wires weren't actually grounded through the ECU. I do not see how two-point-o would have measured 12V at the injectors if these wires were not grounded, since the main relay would not have been energized. I also don't know if we're all talking about the same car/jetronic. Do you have a schematic?

Thanks for the response. I believe I have narrowed it dow to pin 16-17 ground from AVV
No continuity from either wire and no voltage.
Going to try some magiver on it this afternoon and try to bypass to see what happens I have the color Palauan version of the wire diagram been pretty helpful.

 

[doctorb]

My understanding is that pins 16 and 17 are two of three grounds needed for the FI ECU. The other is pin 5. Yes, those wires ordinarily go to a ring terminal that is bolted at the AAV mount. I'm most interested if that is your issue!
For me, I have continuity and grounds are good, which is a good thing, but not-so-good in that I have to look elsewhere.
Good luck. I'm waiting with bated breath.

 

[Mborras1]

My understanding is that pins 16 and 17 are two of three grounds needed for the FI ECU. The other is pin 5. Yes, those wires ordinarily go to a ring terminal that is bolted at the AAV mount. I'm most interested if that is your issue!
For me, I have continuity and grounds are good, which is a good thing, but not-so-good in that I have to look elsewhere.
Good luck. I'm waiting with bated breath.

I have nothing at 16&17 I did a continuity and volt Test and zero. Now thinking on ways to bypass might get foggy and remove the actual pins and tap in a ground wire in both and mount to the chassis somewhere to see what happens.