76 no start problem

[darom]

Hello,

I need your guys' input on something the PO did to my car. 1976 280z, stock, with 79 alternator (no voltage regulator).

I appreciate your time reading this.

I replaced all the injector connectors + 4 more (air valve, cold start valve, thermotime and temp switch). In addition, I took the intake/exhaust manifolds off to install the studs. The PO used a lot of wood screws to hold them. A new Nissan gasket was installed. Everything looked nice and clean. I was happy.

While replacing/soldering the connectors, I noticed the PO ran an ignition wire (START position) to the engine bay's Bosch aftermarket relay (one of the relay's contacts was connected to positive on the battery, and the relay switched wire to the Air Cold Valve's one of the connectors.) The other wire on the ACV was attached to the thermotime's switch connector (basically, like stock's setup to loop both switches together). He left the stock ACV connector loose.

In my infinite wisdom, I figured WTF he needed this part for, and removed his wiring mod to roll back to the stock setup. After that I checked per the Fuel Inj. bible book, all my ECU pins for Ohms and voltages. Everything checked out ok including the AFM pins and fuel pump operation.

Now here is the problem: the car cranks, fuel pump works, but it won't start. The engine almost starts in the first 10 seconds, coughs a few times, and then nothing happens. I can keep on cranking the starter, but it doesn't start the engine.

At this point, I got mad at myself for being adventurous, and replaced the whole INJ/ECU harness with a spare 1976 one. I thought maybe I messed up the polarity of the injectors which could be causing this. Or the PO hacked it at some time.

Well, even with the original wiring harness, the engine wouldn't start. I'd hate putting that mod back, because I don't understand why the PO provided the positive signal to the Cold Start Valve when it can be simply disconnected (one of the steps the Inj bible mentioned was to unplug it completely)?

Did the PO do it to provide the positive to the coil this way?

Now, (putting the flame suit on) since the car's ignition is messed up already, I am ready to install the GM HEI module with my stock 76 distributor to bypass the in-cabin ignition transistor to get better spark/response and drop that ugly ballast. Should I do it and then waste another couple weeks tracing the problem :-) ?

When I took the intake off, I didn't touch the distributor, each spark plug cable was labeled.

Any ideas why the PO ran that relay modification to get the positive wire to the Cold Start Valve?

Thanks in advance!

Den

Edited February 25, 2010 by darom

[darom]

Update:

- replaced the coil with another used one (engine cranks, but no start)

- installed another ECU (same result)

- retested per the Inj. Manual the ECU bulk connector pins (everything tests ok)

- tried advancing/retarding the distributor (no difference).

When the engine is being cranked, I cannot hear the injectors' clicking noise. A pulse should be coming from the ECU. 2 bad ECUs?

I am going to test the distributor's 2 wires tomorrow and the elec. ign. fuel relay.

A few questions:

If I pull the coil wire on the distributor side, and try to start the engine, I should see a spark if it is close to the ground, right?

If the ignition switch is in ON position, should I be able to see voltage below 12V on the coil's + and - posts? Do these get energized only when the ign. is in START?

Thanks!

[FastWoman]

I confess I don't know what "ACV" stands for. I'll contribute this thought, though, for whatever purpose it might serve:

If your cold start valve is working, but your injectors are not, then you will get that initial start, followed by the engine immediately dying. The lack of sound from your injectors is consistent with this.

There (-) wire on your coil should also connect to the tach AND the #1 pin on your ECU. If there is no connection to the ECU, then the ECU won't have the trigger signal it needs to inject fuel once every three sparks (once per engine rotation). I'd do a continuity check to see that the (-) on the coil connects to the #1 pin. If not, there's your problem.

[darom]

Thanks, FastWoman, for catching my typo. I fixed it - it is Cold Start Valve (CSV).

I checked with the voltemeter, on the ECU bulk connector, that I have battery voltage on the pin number 1. The problem is that I don't see the battery voltage on the coil side at all when the ignition key is in the ON position.

I am going to run a positive cable from the battery directly to the positive side on the coil and see if the engine starts. If it does, then I need to figure out why my ignition switch is not providing 12V to the coil. Wouldn't it be the ignition relay (pass. side, kick panel)?

Thanks for the tip. Would you mind please checking this out?

- I have the complete 1976 wiring diagram (76circuit.pdf). If I trace Pin 1 from ECU, it goes to:

White Cable -> Blue -> C4 Block -> Black with Blue (BL) -> Ignition Switch BL

I don't see how pin 1 connects to the - (negative) on the coil?

Doesn't the negative come to the ballast resistor (and coil) from the ignition transistor unit which sends pulses (see attached picture)?

Here is the diagram (sorry about the size, it is 10MB):

http://www.filefactory.com/file/b05b301/n/76circuit.pdf

I apologize about the crappy 20 sec wait time for a free account.

Thank you,

Den

[darom]

Another update:

- I have 12.2V at the coil (+) terminal when the ignition switch is in 'ON' position (good news). I guess no need to run a direct (+) battery wire to the coil's one.

- When the engine is cranking, I see 9.6V at the + coil post.

- The transistor's BW post shows 12.1V when the key is in 'ON'. When cranking, it shows 9.6V.

- If I remove the coil's wire to the distributor, I am getting a nice size spark.

I traced both green/red wires from the engine coil area (2 wires which also go to the distributor) back to the transistor's red/green posts. There is continuity on both.

Still no injectors' clicking noise at all.

I am wondering if the transistor is failing, time to convert it to the GM HEI module? The engine doesn't even 'cough' for the 1st time I try to start it.

Thanks!

Edited February 26, 2010 by darom

[darom]

I had too much coffee -- the ECU controls the injector pulses. I completely confused the transistor unit (spark control) with the ECU (injector pulses).

There is continuity on pin 1, ballast resistor's (-)/tach lead. I am going to take apart the ECU and see if pin 1 inside is not loose (I will re-solder it). Someone on zcars.com forums had some issues with old solder points creating bad connections.

Edited February 27, 2010 by darom

[darom]

Update: did a manual (touch 3 times the ignition lead wire) ECU injector test per the Injection Book and it failed. It appears the ECU fails to start the injectors. The wiring harness passed all the tests (continuity to injectors, to coil (-) and to grounds).

I took apart the ECU, all soldered connectors were in a good shape, tested each lead. I guess I will start looking for a ECU replacement (the one I have is from a different year).

What still puzzles me how the heck the car ran before I removed that CSV hot wire/relay and redid the manifold/intake gasket?

Edited February 27, 2010 by darom

[alhbln]

I dont have experience with your specific ECU but can give some input based on work done with Marelli and Bosch ECUs which might help.

A typical fuel injection ECU from the late seventies/early eighties has the following start sequence:

1. ECU listens to input signals indicating a start scenario, e.g. a low frequency pulse from the cam (engine rotation) and crank (ignition phase) sensor (in your case the cam sensor in the distributor for both rotation and phase) and (optional) a positive voltage from the start position of the ignition key.

2. ECU opens the cold start valve for delivering a rich fuel mix and opens the throttly body (typically the cold start throttle body bypass valves) to get air into the engine, then it starts firing the coil via the transistorized ignition. On some ECUs this happens for 1-2 rotations without igniton to prime the engine. Also, some ECUs activate the fuel injectors in combination with the cold starting value (or don't even have a cold start valve and only use the fuel injectors).

3. ECU fires the ignition according to the RPM signal (some ECUs limit the ignition to 10-20 rotations of the engine before stopping the ignition and fuel injection as a safety measure to not flood the engine if the engine did not start in that sequence. You then need to turn the ignition key to start again)

4. ECU listens to the RPM signal via the cam or crank sensor while supplying fuel via the cold start valve and triggering ignition impulses. If the RPM signal reaches idle speed (~600-800 RPM), the cold start valve is deactivated and the fuel injectors take over. The throttle body is opened and the throttle body bypass valves are then closed.

If the RPM signal does not reach 600-800 RPM within a predefined number of rotations, fuel supply is deactivated to avoid flooding the engine.

5. Engine runs in idle, ECU manages fuel, air and injection based on RPM signals, air temperature (air density), engine temperature (water, oil), air flow etc.

This is only a rough overview, as example some ECUs (Marelli IAW as example) don't have a dedicated cold start valve but enrichen the mixture via the fuel injectors by priming the engine on a cold start.

I would assume that in your setup the ECU does not trigger the injection valves as the engine never reaches running RPM (see step 4), because the cold start valve does not open to deliver the air/fuel mix to enable the engine to start.

The wiring done by the PO would open the cold start bypass valve while starting, so the engine has the correct air/fuel mixture, the ECU would then trigger the ignition until you reach idle RPMs (600-800 RPM), and only then the ECU would trigger the fuel injectors.

Maybe the PO wired it this way because the original ECU failed to initiate the cold start valve via pin 4 & 21? If you have tried another, known to work ECU, then this would indicate that the ECU is missing another signal/sensor input to start the cold start valve.

As a next step, you might want to check that the cold start valve is activated as soon as you engage the starter, by measuring the voltage on the ECUs pin 4 & 21 with a multimeter. A 12V LED (or a standard led with a 480Ω in line with one pin) connected to the cold start valve is also a good indicator.

If you can't measure a voltage at pin 4 & 21 (e.g. the cold start valve is not activated), then the ECU (if it has no defect) is missing another signal input. I would recommend to then check if the starter activation wire is connected to any pin of the ECU and if that wiring is ok and the ECU receives a signal as soon as the ignition key is turned to the starting position.

If there is a signal on Pin 4 & 21, then the wiring to the cold start valve is faulty (probably not the case as you have exchanged the wiring trunk loom already).

I wouldnt worry too much about the ignition and the fuel injectors as long as the cold start valve is not activated while starting.

good luck and let us know how you proceed,

Adrian

[darom]

Adrian, thanks for a good explanation.

I am basically at the same conclusion as you are in regards to the PO's wiring. The CSV never got the ECU's 12V to engage, thus the butchered wiring solution (The CSV on 280-z works only when the engine is being cranked. That's why the PO's wiring was off the START ignition wire to the CSV directly bypassing the ECU pins.). When I bought the car, the PO re-assured me that he had replaced the ECU with a rebuild unit. That's why I have been delaying its replacement (esp. when the car's engine ran) until I discovered the relay/CSV hook-up etc.

The cold start system circuit passed the test (page 71 of the Elect. Fuel Injection 280-z book" with the harness disconnected from the ECU, starter solenoid lead wire disconnected, ign. key in START position. Here is the reference: http://www.atlanticz.ca/zclub/techtips/efisystem/280zfuelinjectionbook.pdf

I found a used 76 ECU - I will update the thread once it is installed. Maybe someone will benefit from this in the future.

Adrian, I like your avatar signature: "Registered Ign/ECU Nurse: :-)

Edited February 28, 2010 by darom

[alhbln]

Very helpful documentation, thanks for the link. On page 24 the relay wiring for the start circuit (power relay:ignition on) proves that the starter voltage is indeed forwarded as a start signal to the ECU.

The signal input at the ECU seems to be on pin 4 (see page 71), so the first test should be to measure pin 4 at the ECU against ground when starting to make sure the signal is actually present at the ECU.

If its not, a quick fix might be to connect a wire to pin 86 of the relay socket to pin 4 of the ECU.

There is a good chance that the ECU is not defective but the wiring is faulty. If the signal is present at pin 4 though, you might have lost the input driver stage of the ECU (ttl buffer, transistor), which should be easy to replace in most cases.

p.s. could you post a photo of the internals of the ecu? i am naturally (nursely) curious.

[darom]

Adrian, you are _good_!

I ran the test on page 69 (CSV disconnected) and it registers my battery voltage on pin 4. So I think the wiring is good through the ign. relay all way down to the ECU.

I think the ECU fails to provide ground on pin 21 for the CSV (page 71).

Hmm, I just rechecked - my ECU doesn't have any leads on pins 20, 21. Now I am guessing the Nissan engineers ran the wire (21) from CSV to the ECU harness plug just for testing sakes.

The CSV gets its ground from the thermotime switch, positive off the fuel ign. relay (47).

On my circuit board the following pins are soldered: 1-18, 30-35, 26-27.

You can be right, that pin 4 might be the culprit in my ECU. I checked the resistor on pin 4 and it looks ok (24k ohm). Red-yellow-orange-gold 24k Ohm +/- 5%. Beyond that I am not much help testing the board :-(

For pictures, please find them here:

http://s24.photobucket.com/albums/c1/darom/280z/

Thanks.

I appreciate your guys help!

[alhbln]

The symbol for the thermotime switch is not complete, it seems to be a temperature activated mechanical switch, not a thermistor.

So, if activated (cold temperature?) it switches on the cold start valve by supplying ground and also forwards ground to pin 21. I think your guess is quite correct, pin 21 might have been assigned for checking the status of the thermotime switch by the ECU, or overriding the thermotime switch, but that feature then was never actually implemented.

Based on this setup i would guess that in starting mode the ignition and fuel injectors are always active, and a cold engine then activates the cold start valve via the thermotime switch to further enrichen the air/fuel mixture.

Regarding the ECU, can you trace where the conducting path after the 24KΩ resistor ends up? I would expect a small signal transistor in a TO-92 package, as 24K sounds like a line resistor to drive the base of a small PNP.

Or it might drive one of the two NCC relays next to the connector, but for that setup the 24K resistance is a bit high.

Thanks for the ECU photos, quite interesting! The ECU is a discrete circuit without any microcontroller help from what i can see. There are not even any TTL chips. Should be relatively easy to repair.