Synch vs Asynch fueling

[gregs78cam]

In my very limited TBI experience, I have gotten down to reliable 0.488ms at a warm idle, and O2 cross counts still occuring.

[907 Chevy]

I believe synch is better for larger injectors for idle. I am thinking that if asynch is allowed the ecu moves form synch to asynch in a rythmic pattern causuiing a surge. Pulse width is constanting oscillating. For PE WOT asynch I believe will work but if you have adequate inj size and or FP then synch may be better. You cant have both
synch and asynch.

Downside of using synch only is ecu may not be able to command a desireable A/F as there is a limit as to small a PW is possible. I recall discussion as to the maximum PW being around 7.0 msec but have not seen any comment on minimum PW. There is also proportional gains that may need to be reduced to allow a very short PW. Mine are about 50% less than stock L03(inj's75 lbs at 26 lbs FP).

As pointed out a surge can be caused by vac leak or lean. Other tools exist to steady idle. Try varying SA to determine what works best for you. Make all SA around your idle cell the same SA. Lock SA for idle. Consider idleing in open loop at a richer A/F. Some masks allow the above others do not.

Hmmm... looks like in BJYL $0D minimum synch PW is around 0.5 ms.

[907 Chevy]

In my very limited TBI experience, I have gotten down to reliable 0.488ms at a warm idle, and O2 cross counts still occuring.

Did you have to lower the min synch PW? I was under the impression that injectors don't work that well with pulse widths under 1 ms... maybe I'm mistaken on that?

[EagleMark]

Guess it all has to do with the Mask ID as well? Can always change min...

Then if your data can do injector pw to milliseconds and you have a value/gauge... is it compensation for Injector Bias? $42 bin, some, is .396 to begin with.

[RobertISaar]

i'd say it depends on the injector... newer high-impedance units do better than some of the old 80s multecs.

low-impedance? those can be driven pretty low before they get finicky.

[POZE]

Injectors are of either low or high impedance. Low impedance injectors range from 2-4 ohms. High impedance range from 12-14 ohms. The impedance of the injector determines what type of driver is used in the PCM. Low impedance injector drivers (peak and hold) allow more current to flow to open the injector faster, and ramp it down to hold it open. High impedance injectors use a saturated switch type driver. They will open a little slower .5. - 1 ms, and the current will stay more constant once the coil becomes saturated. The injector needs to match the ECM type.

[gregs78cam]

Did you have to lower the min synch PW? I was under the impression that injectors don't work that well with pulse widths under 1 ms... maybe I'm mistaken on that?

I did lower the Min Synch PW to something like 0.200 or 0.300, I don't remember exactly.

[1project2many]

You also need to allow for closing time and behavior. While the injector is closing and the pintle is moving through the magnet, a voltage is being created with current wanting to flow backward. That voltage can get pretty high if the driver circuit "disconnects" the injector electrically. If the driver turns on and "reconnects" the injector partway through closing, well, now you've got a bit of reverse current applied to your circuit. This has to be dissipated before the injector can be opened. The circuit needs to be robust enough to handle it and you're going to make heat in the circuit and the ecm. If you constantly and repeatedly open the injector while it's trying to close you can end up with some really inconsistent injector behavior and you may risk damaging a driver.

With the four TBI injectors on the Cavalier I think I ended up around .6ms before the pattern fell apart. Unfortunately I needed substantially less. I finally worked it out, as mentioned, using the quasi-asynch mode but even still that ecm case could get pretty warm at times.

[RobertISaar]

this one of the reasons you see diodes going to ground on the circuits that control other inductive stuff? IIRC, that has to do with a magnetic field breaking down and creating negative voltage when power is removed?

i've actually experienced that while testing a CCP solenoid... i was using a booster pack to test it by switching power on and off... every time the power would switch off, the "leads reversed" alarm would chirp momentarily. added in a diode next to the solenoid that doesn't conduct when voltage when the 12V side has positive voltage(but would when voltage was higher on the negative side) and no more alarm going off.

[1project2many]

this one of the reasons you see diodes going to ground on the circuits that control other inductive stuff?

Yep. AKA "flyback current." I once created a bunch of trouble in a C3 system because I didn't put the diode in an A/C connector.

[corvairnut]

How do we disable the async mode in $OD mask? I have been working this issue and now that I sync idle and got my sync bpw's right, it idles the best it has since I started this project. Since then, Ive got a small jerk/surge verified by the Async flag being thrown each time as get the jerk/surge. Yes... its very small, but I do feel it I would like to eliminate it.

Im not really interested in fixing the async setting, just disabling and keeping in sync all the time (dumb it down), but that seems more difficult that I thought. Maxing out and mins do not work for the sync to async and async to sync. I did max out the RPM and MAP on the async settings, so it only comes on during decel and light throttle on decel.

[1project2many]

Eliminate all acceleration enrichment, deceleration enleanment, stall saver, and small pw asynch operation. Asynch is used for transitional fueling. It's part of the programming so eliminating it completely while maintaining transitional fuel could be a challenge. It's probably easier to eliminate it in the one range where you're having problems.

The question is what is causing the ecm to enter asynch mode? Is it noise on the REF line tricking the ecm into thinking engine rpm is changing quickly? Or voltage spike on the MAP or TPS signal wire? Is rpm momentarily falling below stall saver threshold? Try enabling only MAP or TPS based AE to see if it clears up. Look at reported RPM to see if it's changing right before asynch fuel pulse is delivered.

[Six_Shooter]

Why is there such confusion over sync and async fueling?

It's simple, sync fueling is the injectors pules WITH the DRP input, or at least in equal proportions. There will be some propagation delay between DRP input and injector pulses.

Async fueling pulses the injectors every 12.5 msec, regardless of number or frequency of DRP.

It has nothing to do with bank to bank or batch firing.

[EagleMark]

Why is there such confusion over sync and async fueling?

It's simple, sync fueling is the injectors pules WITH the DRP input, or at least in equal proportions. There will be some propagation delay between DRP input and injector pulses.

Async fueling pulses the injectors every 12.5 msec, regardless of number or frequency of DRP.

It has nothing to do with bank to bank or batch firing.

Well that's one part... TBI Sync is each injector firing alternatly, Async they are firing together. TBI to MPFI conversion changes to Async/Batch, Sync/Bank...

[Six_Shooter]

Well that's one part... TBI Sync is each injector firing alternatly, Async they are firing together. TBI to MPFI conversion changes to Async/Batch, Sync/Bank...

Not in any information I've ever looked into. Regardless of induction/fuel delivery scheme, the sync and async mean, or act exactly the same way. With DRP, or 12.5msec apart.