What is the resolution of the injector controls? This is a general question for all GM ecms; c3 (TBI), p4 (MAF&SD TPI), etc.
Is it 16bit or should I just go open a hack file and read through the source code .

 

16 bit as far as I can see from the code.

 

Yep, 16 bit value. On a C3 TBI ECM ('747, '8746) each bit is 15.26 usec of injector PW. From the code (I've never tried it), the counters appear to be signed. So the max PW is from the value $7FFF for 500 msec of PW.

In async mode a C3 TBI ECM each bit is still 15.26 usec of PW. However, it is only advanced on every other bit. So the true async PW resolution is double that: 30.52 usec.

IOW, a value of 10 will give 152.6 usec of PW. Same as the value of 11: 152.6 usec. Then the value of 12 will increase the PW by 30.52 usec for 183.12 usec of PW.

Interesting huh?

RBob.

 

Yup... now the questions is how much will it effect tuning? I know with our motorcycle engine our tuning was every 625usec but that's batch bank firing with small injectors, high fuel pressure. TBI p&h's fire frequently enough that even 30usec of resolution should be plenty yes? I would think so. I can't imagine 30usec effecting the AFR by much, or would it....

 

I don't see the 30 usec resolution in async mode being a problem. I am finding that async mode itself has problems. Something just isn't right about it. . . .

RBob.

 

Can you elaberate on that a bit?

 

Async mode should not be used above 1200-1400 RPM. Else the manifold can dry out between async pulses.

This becomes acute at 2400 RPM and higher.

RBob.

 

I thought it would be ok up to 2400 rpm since its equivelent to the injectors firing normally at 2400 rpm but this is not the case? Is it due to the way they are firing together?

 

Yes, the way the injectors fire together and the rate at which they fire. At 2400 RPM and in async mode, the two injectors will fire together at a maximum possible rate of every 12.5 msec.

(please note that the following is intended for a broad audience).

At 2400 RPM there is a cylinder firing every 6.25 msec (8 cyl engine). So even if async mode is firing every 12.5 msec the intake stroke following injection will be rich. The next intake stroke (next cylinder in firing order) will be lean(er).

See the attached graph. The upper dark blue is the firing of each cylinder (spark). The yellow and magenta (InjA, InjB) are the syncronous firing of the injectors. The RPM for this graph is 2400.

As shown in sync mode the injectors alternate at each plug firing. The fueling is consistent between each cylinders intake stroke.

The bottom two lines (Async_1, Async_2) are two different async possibilities. The light blue (Async_1) shows what I have described above. That is async mode at the highest possible injection rate.

The bright blue bottom most line (Async_2) is where trouble really starts. In this case there wasn't a long enough injection pulse required for it to be injected (note missing pulse at 25/26). The ECM accumulated the short pulse and waited until the next loop through to inject that fuel.

See the blue bar over top to show this area of the graph. In this case there are 3 cylinders doing an intake stroke without an injection pulse. As the fuel runs out (fuel film evaporation) each cylinder in turn runs leaner then the previous one.

As the ECM loops through there is another async injection pulse. Now this pulse is going to be an accumulation of the previous non-injected pulse and the current calculated pulse. This pulse will then cause the next cylinder in turn to be very rich. With the subsequent cylinder(s) in turn leaning out.

The result of all this is a rough running engine, stumbling and cut-out, pop and burbling of the exhaust.

The scenario shown is only two of multiple possibilities. The higher the minimum allowed async PW is set to, the further apart async injection pulses will be. The switchover parameters for sync-to-async and back also affect this.

Even AE can be affected by the minimum allowed async PW. A high value here can cause a chunky feel to acceleration as the go-pedal is pressed into service.

RBob.

 

Another aspect of async fueling is maniold design: single plane vs. dual plane. In the scenerio above the discussion is centered around a single plane manifold.

With a dual plane manifold some of the problems associated with async mode vanish. Recall that a dual plane manifold has two plenums one for each bore of the TBI unit. Each plenum then feeds four cylinders.

The routing of the runners is such that each plenum feeds every other cylinder in firing order. So each time an intake stroke occurs, the next cylinder in turn will be from the other plenum.

In this case whenever in async fueling there is no stealing of fuel from both injectors. Each cylinder only receives that fuel from the one injector that feeds that plenum. This also affects AE fueling.

RBob.