Tuning With An Air/Fuel Gauge

[jonbill]

when it is in this "lean" state, does it feel normal to drive, or hesitant? 

16s is prety lean and I'd expect you'd feel it, so if you don't feel any difference, I'd think it must be an air leak in the exhaust before the sensor. 

[Captain Obvious]

OK. I had a couple goofy ideas.

First thing... The original EFI systems were fully analog and there were no computers used. But then some time along the way (early eighties), they started putting embedded microprocessors in the ECU's and using software to do the control instead of analog voodoo.

Why does that matter to me? Because if it's got a microprocessor doing the control, then this kind of latching behavior would be easier to explain. So I know your Z is a 78, which should be fully analog, but do you know for sure? Have you ever had the cover off the ECU? Got any pics?

Second thing... There's a circuit inside the ECU that is supposed to provide a little extra richness boost immediately after starting the engine, and that extra richness gradually tapers off (30 seconds or so) after the key is released. Maybe that section of the circuit is not functioning correctly, or maybe that circuit is being fooled into thinking that the key is always in the START position.

When you turn the key to START, a couple things are supposed to happen:

The starter engages (duh)
The floor temp lamp light is supposed to light up (lamp test for CA models only)
The cold start injector gets the opportunity to spray (depends on how warm the thermotime switch is)
The ECU get's the signal to begin the START enrichment boost

So here's what I'm thinking... Next time the car lean boots, try this:

1) WHILE THE ENGINE IS RUNNING, pop the hood and disconnect the small Black/Yellow wire going to the starter.
2) Turn the key to START again with the engine still running in lean-boot. Don't worry... With the B/Y wire disconnected, the starter will not engage because you have removed it's signal to engage. But all the other "harmless" functions will be activated. The floor lamp should light up, the cold start injector should give a squirt (if it's cold enough), and the ECU should get a fresh "START" signal.

The idea is that you can flick the key between ON and START a couple times to see what happens to your mixture readings. See if your lean boot symptom goes away without having to shut the car off completely? Maybe it'll go away with just a flick of the START signal.

Like I said, I'm just throwing some stuff out there! 

[chaseincats]

31 minutes ago, Captain Obvious said:

OK. I had a couple goofy ideas.

First thing... The original EFI systems were fully analog and there were no computers used. But then some time along the way (early eighties), they started putting embedded microprocessors in the ECU's and using software to do the control instead of analog voodoo.

Why does that matter to me? Because if it's got a microprocessor doing the control, then this kind of latching behavior would be easier to explain. So I know your Z is a 78, which should be fully analog, but do you know for sure? Have you ever had the cover off the ECU? Got any pics?

Second thing... There's a circuit inside the ECU that is supposed to provide a little extra richness boost immediately after starting the engine, and that extra richness gradually tapers off (30 seconds or so) after the key is released. Maybe that section of the circuit is not functioning correctly, or maybe that circuit is being fooled into thinking that the key is always in the START position.

When you turn the key to START, a couple things are supposed to happen:

The starter engages (duh)
The floor temp lamp light is supposed to light up (lamp test for CA models only)
The cold start injector gets the opportunity to spray (depends on how warm the thermotime switch is)
The ECU get's the signal to begin the START enrichment boost

So here's what I'm thinking... Next time the car lean boots, try this:

1) WHILE THE ENGINE IS RUNNING, pop the hood and disconnect the small Black/Yellow wire going to the starter.
2) Turn the key to START again with the engine still running in lean-boot. Don't worry... With the B/Y wire disconnected, the starter will not engage because you have removed it's signal to engage. But all the other "harmless" functions will be activated. The floor lamp should light up, the cold start injector should give a squirt (if it's cold enough), and the ECU should get a fresh "START" signal.

The idea is that you can flick the key between ON and START a couple times to see what happens to your mixture readings. See if your lean boot symptom goes away without having to shut the car off completely? Maybe it'll go away with just a flick of the START signal.

Like I said, I'm just throwing some stuff out there! 

Interesting theories, I'll try and answer them in order!

- I haven't opened the ecu before but there is a number on the side which should denote which it is - I can grab that tonight

- I haven't tried this but am a bit confused.  If the starter doesn't crank and the engine isn't running, how would the a/f ratio be readable?

Sidenote: I forgot to mention earlier that before I shut it off last time when I returned to the garage, I popped the hood and unplugged the TPS to see if it leaned itself out further as it should.  It indeed did go leaner at idle but noticeably less of a change than if it was not in a lean-boot mode.  We're talking going from mid 14s to upper 14s instead of low 14s to 16.  Not sure if that data point is useful or not but I figured it couldn't hurt bringing it up!

 

-chase

[Captain Obvious]

2 minutes ago, chaseincats said:

If the starter doesn't crank and the engine isn't running, how would the a/f ratio be readable?

The engine should already be running. Let me try again to explain.

Next time the car starts and you notice this lean-boot condition. Let it run, but while it's running, pop the hood, get out of the car, and disconnect the small black/yellow wire from the starter solenoid. 

Then get back in the car (all with the engine still running), and turn the key to START. The engine will already be running (because you never shut it off), but the starter will not engage since you have that wire pulled off the starter. Flick the key between ON and START and see if it drops out of lean boot and into regular running mode.

Does that make sense?

You can drive like that with the starter wire disconnected, but just remember that when you shut the car off, you'll need to reconnect the starter wire to get the car to start again.

[chaseincats]

45 minutes ago, Captain Obvious said:

The engine should already be running. Let me try again to explain.

Next time the car starts and you notice this lean-boot condition. Let it run, but while it's running, pop the hood, get out of the car, and disconnect the small black/yellow wire from the starter solenoid. 

Then get back in the car (all with the engine still running), and turn the key to START. The engine will already be running (because you never shut it off), but the starter will not engage since you have that wire pulled off the starter. Flick the key between ON and START and see if it drops out of lean boot and into regular running mode.

Does that make sense?

You can drive like that with the starter wire disconnected, but just remember that when you shut the car off, you'll need to reconnect the starter wire to get the car to start again.

Oh gotcha, that makes sense.  I'll give that a shot next time but this literally happens every few months so it may be a while before I'm able to test that theory.

[Captain Obvious]

Understood about the timing. Maybe you'll get lucky and it will never happen again. (Suuuuuuure).

And I'm not proposing any theories yet... I'm just trying to come up with something, anything. to maybe point in the right direction. I'm thinking that if you do the "START with engine already running" a couple times and nothing changes, then maybe it REQUIRES the engine to actually come to a stop to reboot. Injectors need to stop injecting, coil needs to stop coiling... all that.

On the other hand, if you do that test and the lean condition DOES go away, then there might be some info to be gleaned from the fact that the engine does not need to a complete stop in order to make the lean boot go away. Vibrations don't need to stop, starter does not need to suck a big hit of current from the battery. Alternator doesn't need to bootstrap. That kind of stuff.

Just fishing at this point.   

[chaseincats]

26 minutes ago, Captain Obvious said:

Understood about the timing. Maybe you'll get lucky and it will never happen again. (Suuuuuuure).

And I'm not proposing any theories yet... I'm just trying to come up with something, anything. to maybe point in the right direction. I'm thinking that if you do the "START with engine already running" a couple times and nothing changes, then maybe it REQUIRES the engine to actually come to a stop to reboot. Injectors need to stop injecting, coil needs to stop coiling... all that.

On the other hand, if you do that test and the lean condition DOES go away, then there might be some info to be gleaned from the fact that the engine does not need to a complete stop in order to make the lean boot go away. Vibrations don't need to stop, starter does not need to suck a big hit of current from the battery. Alternator doesn't need to bootstrap. That kind of stuff.

Just fishing at this point.   

Haha sounds good.

I just went to check and this is my ECU's version label like we were talking about: 

[Captain Obvious]

I'll see what I have for ECUs here. I don't know if I can tell anything from the label, but I'll look.

So here's to hoping the lean boot thing just doesn't come back, but in reality what will probably happen is that it'll occur when you're dressed nice and don't want to go popping the hood and pulling wires off and all that.  Haha!

[chaseincats]

On 6/29/2021 at 8:07 AM, Captain Obvious said:

I'll see what I have for ECUs here. I don't know if I can tell anything from the label, but I'll look.

So here's to hoping the lean boot thing just doesn't come back, but in reality what will probably happen is that it'll occur when you're dressed nice and don't want to go popping the hood and pulling wires off and all that.  Haha!

I removed/opened my ecu for my new problem here: https://www.classiczcars.com/forums/topic/65578-sudden-engine-shutoff/?tab=comments#comment-624753.  But since we were discussing what my ECU looked like, here it is too lol

 

Edited July 9, 2021 by chaseincats

[Captain Obvious]

OK, so that's an original 78 ECU and there's no software in it. In other words... I don't think there's anything inside that can that would cause a lean-boot that would last the length of the drive and then be reset by a reboot. Still sounds like a software issue to me.

And with that in mind... Something we haven't discussed in depth... Jonbill had asked earlier "when it is in this "lean" state, does it feel normal to drive, or hesitant?"

Your A/F meter is the only entity attached to the car that has any software in it. Let's consider the prospect that it's a measurement phenomenon and not "really" occurring?

So can you tell from driving the care while it's in the lean-boot state... Does it feel lean, or does it feel the same as when it's not measuring lean?

[chaseincats]

37 minutes ago, Captain Obvious said:

OK, so that's an original 78 ECU and there's no software in it. In other words... I don't think there's anything inside that can that would cause a lean-boot that would last the length of the drive and then be reset by a reboot. Still sounds like a software issue to me.

And with that in mind... Something we haven't discussed in depth... Jonbill had asked earlier "when it is in this "lean" state, does it feel normal to drive, or hesitant?"

Your A/F meter is the only entity attached to the car that has any software in it. Let's consider the prospect that it's a measurement phenomenon and not "really" occurring?

So can you tell from driving the care while it's in the lean-boot state... Does it feel lean, or does it feel the same as when it's not measuring lean?

Gotcha

It absolutely feels hesitant and is a real dog to drive so it shouldn't be the gauge in that case.

[Captain Obvious]

Bummer. Was hoping it was just an instrumentation issue.

We're back to where we were before. Keep us posted!