Could someone point me to a link or explain just the very basic concept behind the electronic spark control by the ECM? I'm thinking of 7747, but they're all probably similar.

Specifically, how exactly does the ECM control the spark timing? Mechanically, the distributor should be at zero degrees, and any change from there is done electronically, correct? Does the ECM send a control signal to the module in the distributor to advance or retard the timing (like simply telling it "advance 15 degrees"), or does it actually send the pulse to fire a plug?

If it's the latter, it seems like by the time the ECM got the pulse from the distrib, it would be too late to do anything but retard the timing. Unless I misunderstand what the reference pulse actually is.

Thanks for any help.

 

Take a look at the document "Turbo_P4_Doc.pdf" here :

ftp://ftp.diy-efi.org/uploads/

Although this document goes through specifics for the 7749/$58 code, there are a number of sections (like Section 11) that are pretty generic an applicable to any number of ECMs. If you're at all interested in the details of the algorithms within the ECM, you probably won't be able to stop reading this!

The ECM sends "pulses" to the module in the distributor that tell it how long to charge (dwell) and when to fire (spark). The signals to the ECM are constant for a given distributor position. And, it knows how quickly the pulses are coming (RPM). So, it can figure out when to fire based on the "last" pulse as compared to waiting for the "current" pulse to get there.

Consider the engine running at a constant 1000 RPM. The distributor pulses will occur every 15ms. This means the crankshaft rotated 90 degrees at each pulse. (Taking a V8 as the example with static timing of 0 deg.) So, the crankshaft is moving at 6000 degrees per second. We want to fire Cylinder 8 at 10 degrees advance.

In order to fire Cylinder 8 correctly, we want to "fire" the spark 10 degrees "before" the Cylinder 8 pulse or 80 degrees "after" the Cylinder 1 pulse. 6000 deg/sec = 0.000167 sec/deg. 0.000167 sec/deg * 80 deg = 0.0133 sec after the Cylinder 1 pulse.

Hopefully, this will look OK....

Distributor pulses to ECM:
Cyl 1 TDC Cyl 8 TDC Cyl 4 TDC ...
______|---|_____________|---|___________|---|__________ ...
90deg 180deg 270deg ...

EST Pulses from ECM:
_____________|----------|________|----------|_________.....
<- 13.3 ms -> <- 13.3 ms ->
spark spark ............

Ideally, this will make some sense. There are additional considerations like what to do when the engine speed isn't constant (dynamic dwell). And, if I've screwed up the math, I'm sure someone will let me know!

 

I noticed my previous reply didn't show the graph very well. Here's a better picture of what I was trying to show:

 

There are two conditons to spark timing.
EST
non-EST

Non-EST is when the engine is just using the dwell that the module has, like when you first crank the engine. So for *most* SBCs you want about 8d initial timing. So that the engine doesn't just windmill during cranking, nor is too hard to crank when hot.

EST mode comes into play at about 400 RPM. At that speed the ecm thinks the engine is running an it's OK to actively control the timing.

The reason for the two modes, is that if the processor in the ecm dies, the engine will still run on the ignition module.

BTW, there is also a little advance built into the module's design.

 

Thanks for the information. This is starting to make more sense. A few questions: What is the relevance of the pulse width from the ECM? Does it matter, or it it just the falling edge of the pulse that matters?

Regarding the modes (EST, non-EST), can you explain more detail? Don't the TBI 7747 GM's specify that you pull the connector to the distributor and set the timing to zero degrees? I know some people push it up to 6-8 or something, but factory is zero.

Anyway, how does it work electrically? Somehow the ignition module in the distrib can either work off of it's own pulses or take pulses from the ECM? How does this work? I guess I could see it if the ECM box only took in pulses and delayed them going back to the distrib - but if the ECM actually read the pulses coming in and then generated new output pulses, then if the ECM went south the whole system would break down.

Maybe I should back up. What is that connector that you open up to set the timing? Is it the "ignition module bypass", pin D5 on the 7747? Maybe it is just a signal to the ECM to not mess with the timing so it can be set?

 

When the pulse is high, the coil is charging (i.e. dwell). When the pulse goes low, the spark is generated.

The ECM effectively has two outputs to the distributor. The EST line and the bypass line. When the ECM wants to control timing, it pulls the bypass line low. The distributor module then takes its cues from the ECM via the EST line. When the ECM doesn't want to control timing (or, when you disconnect the connector at the distributor), the bypass line is left alone. The distributor will run off of its own regardless of what the EST line does.

Depends on the failure. If the ECM really goes out to lunch, it probably wouldn't know enough to tell the distributor to take over. In that case, you are correct that the system would completely fail and likely not run. For things the ECM does detect, like overvoltage or lack of distributor pulses, it can and will tell the distributor to take over.

 

I realize this is standard stuff I can look up elsewhere, but does the dwell need to change vs. RPM (or anything) or is it more or less fixed?

Okay, that makes sense. Looking at a wiring diagram for a 7747 vehicle, I see four lines going from the ignition module to the ECM pins B3, B5, D4 and D5. They aren't labeled on the diagram, but I found some names in the GM-ECM archives:

B3: Dist ref low
B5: Dist ref pulse
D4: Spark timing out
D5: Ign module bypass

So, I'd translate your description to say that when the ECM wants the ignition module to control its own timing, it leaves D5 open. When it does want to control timing, it pulls D5 to ground and then sends pulses out on D4. "ref pulse" on B5 is clear. I assume that B3 is essentially the ground reference for B5, because the reference pulses aren't relative to regular vehicle ground?

And the line to D5 is the one people open up when they want to set the mechanical timing at the distributor.

Makes sense.

 

The answers you seek are contained in the document I mentioned in a previous reply......

The rest of your post is right on the money!