A 3.4 DOHC Build then... F40 Turbo (Page 32/57)
sleevePAPA NOV 19, 10:29 PM
what is the PW? Have you tried single firing for idle?
Fierobsessed NOV 20, 12:46 AM
Its firing between .12 and .15 ms to get me to 14.7:1.

I believe I am in single fire, but I have to check the logs.

Admittedly, it now idles and runs great at 14.7:1 open loop. But as soon as the slightest bit of fuel is trimmed by closed loop... it just shuts down.

I'm a bit curious though about idle timing. At idle I found if I start throwing a few degrees at it, the AFR's start to drop considerably even with the exact same amount of fuel going in. I assume that it's a good thing, its bringing up its idle fuel efficiency, and asking for fuel to be trimmed in return, which I am unable to do.

Is there an "every other revolution" mode? That might help... SFI works on that principal if I'm not mistaken. But it could cause instability I would guess.

I can't be the first to try to tune idle on a GM P4 ECM using 60 lb injectors. Makes me wonder what everyone else is doing, or what I am missing right now. These same injectors are used on some ULEV vehicles, so it has to be possible. Maybe the lack of SFI is the reason I can't get the PW's low enough to idle correctly. I can't fix that without rewiring the whole thing...
RobertISaar NOV 20, 01:13 AM
non-SFI shouldn't be a problem..... some of the code59 guys are running 80 lb/hr without issue IIRC.

i've forced single-fire mode in place of double-fire in A1 before..... i really didn't see a difference in terms of any output or idle stability or ......

but i was also playing with 3.1 liters and 16.7 lb/hr injectors, so i didn't have such critical of timings. i tried for a few weeks to try and get the transition from single-fire to double-fire correct, but i must not have understood the code that operates it well enough, roughly 50% of the time it would go off without a hitch.... the other times a calculated single-fire burst of fuel would have a double-fire burst of fuel added on top of it, causing the AFR to shoot way rich for a cycle. feels odd with the TCC locked. i seem to remember the double to single transition being less of a problem.
Fierobsessed NOV 20, 06:16 AM
Did some code searching. looks like this code only sets the "Deliver fuel on every reference pulse" flag during cold start. I'm assuming that's 3 pulses per rev.

Otherwise, I can't find anything showing single or double fire mode anywhere.
Joseph Upson NOV 20, 07:26 AM

quote
Originally posted by Fierobsessed:

Its firing between .12 and .15 ms to get me to 14.7:1.

I believe I am in single fire, but I have to check the logs.

Admittedly, it now idles and runs great at 14.7:1 open loop. But as soon as the slightest bit of fuel is trimmed by closed loop... it just shuts down.

I'm a bit curious though about idle timing. At idle I found if I start throwing a few degrees at it, the AFR's start to drop considerably even with the exact same amount of fuel going in. I assume that it's a good thing, its bringing up its idle fuel efficiency, and asking for fuel to be trimmed in return, which I am unable to do.

Is there an "every other revolution" mode? That might help... SFI works on that principal if I'm not mistaken. But it could cause instability I would guess.

I can't be the first to try to tune idle on a GM P4 ECM using 60 lb injectors. Makes me wonder what everyone else is doing, or what I am missing right now. These same injectors are used on some ULEV vehicles, so it has to be possible. Maybe the lack of SFI is the reason I can't get the PW's low enough to idle correctly. I can't fix that without rewiring the whole thing...



Have you tried following the Integrator on the Tunerpro dash to see what it's doing at idle when closedloop is activated? It does not always match what the wideband reads. That helped me a lot with the direction I needed to go for adjustments as well as when loads were initiated at idle; cooling fan, A/C.

You may also need to start closedloop activation at a higher temperature for now, if it is as low as the 8F OE setting of ~80 deg. I had a fairly easy transition into closedloop at 135 deg and part of this problem is not knowing what all the coldstart tables are doing, when and for how long at startup to tune in consistency. Anything lower than 135 deg and an idle hunt set in along the way to full operating temp. So a coldstart areas state of tune varies as a result to the point of having a headache at a very low activation temp, to no trouble at a higher activation temp with such a large injector change.

You may also need to restore the characteristic idle spark advance for the tune, as well as find out what the typical idle spark advance is for that motor. When I zeroed out the idle spark correction tables a couple days ago as you had done, along with my already flattened sparkadvance area, I actually wound up with an idle hunt in openloop that I did not have with them in place before. So I reasoned that with my high compression ratio that perhaps I need to increase them a little above stock. So far so good although I'm still running openloop at the moment.

You may also need to consider your idle rpm setting considering the tune 8F anyway, was setup for a smaller motor with about 40 lbs of coverter, flywheel and fluid weight. I have the best results around 925 rpm. It may sound irrelevant but after wrestling with this subject for the better part of a year before making good progress you will want to consider all possibilities given the enormous number of tables in the code to accomplish the same thing "brainless" carbed engines did with little effort.

[This message has been edited by Joseph Upson (edited 11-20-2013).]

Will NOV 20, 09:27 AM
With more displacement and a lighter flywheel, the 3.4 will gain and lose RPM more quickly than the 3.1/auto. That affects idle stability significantly, as a given load will have a much larger effect on idle RPM and require a faster correction to keep the RPM from changing significantly.
RobertISaar NOV 20, 09:51 AM
i took a look at all stores to 3FD0.

the section of code starting at A4C3 looks to be the one involving single vs double-fire. there are two stores, A4F1 deals with the no-fire cycle, A528 deals with the double volume cycle(due to the LSLD at A501).

if single-fire isn't allowed, the code immediately jumps to A52E.
Fierobsessed NOV 20, 03:36 PM



9.2.2 Quasi-Asynchronous Fuel Delivery
Quasi-Asynchronous fuel delivery is used when the synchronous fuel base pulse width becomes
so small that the fuel pulses cannot be accurately delivered.
Enabling quasi-asynchronous fuel mode will cause the fuel injectors to be energized every other
reference pulse but for twice the duration (of the normal synchronous base pulse width). This
results in the same amount of fuel being delivered, but with an accurately controllable injector
pulse width.

It looks like what I really need is quasi-asynchronous mode. Problem is, It's not activating it for some reason. It appears that all the conditions are being met, but it's not going into that mode.

C0D0, Battery voltage normal
C0DF, Engine running
C0E3, DFCO not active
C0F3, MPH not exceeding L86C2 (255)
C0F9, BPW high hysteresis qualifier L86BA (1.28 MSEC, I'm WELL BELOW!)
C102, BPW low hysteresis qualifier L86B8 (0.98 MSEC, Again, WAY BELOW)

So it should be setting the quasi async flag. but I'm not seeing it. I need to get this one figured out!
joshua riedl NOV 20, 05:46 PM
How About commanding 13:1 afr and see if it will idle that way. It should at least tell you if it's in the tune or injectors.
mkman NOV 20, 07:45 PM
I would like to ask you a quick question that might help me with my build. I noticed you did not use the jackshaft intermediate axle that the F40 tranny uses, what application of inner cv cup are you using that will go into the F40 trans directly? I want to do my build without the jackshaft, but I haven't gotten to the axle research just yet.

Thanks, Mike