I’ve got one of those ideas in my head, you know the ones, the ones that are just freaking brilliant or stupid, but I can’t decide. Maybe someone can throw some information my way so I can decide if I’m going to look into it further.

What would happen a fairly stock setup with a 165ecm (’87 L98) if you installed injectors that flow 2x as much and then only routed half the air that the engine is breathing through the MAF?

My instinct says that if it only sees half the air it will only open the injectors half as long so you’ll essentially idle and maybe even run just like you did with the original injectors, right?

Problems that I could see:
- will the MAF read half the idle air flow accurately?
- How is the MAF information used? Is the correct fueling looked up on a map or is the MAF value use to compute the correct fueling? Basically, I’m wondering if half the air flow would put the air flow value out of range of something that the ECM could do something with?
- How would this mess up the timing (not as big an issue, more accurately, one that could be programmed around)?


Thanks

 

BTW, before anyone says it, I realize that if you needed to you could burn a chip with a different injector constant (and I have talked about burning a chip for the application anyway), but that is not what I'm trying to accomplish.

 

38 or 48#/hr injectors on a stock motor?.
You'd have to have a 1,200 rpm idle to use that much fuel (give ir take)

The MAF isn't a linear deal, you have varing amounts of laminar flow as the air speed thru it changes, halfing the air flow thru it would totally fubar the MAF and scaler stuff.

 

Are you saying to get it to idle at all? Why? I've seen plenty of cars with larger injectors then that idle lower then that.

Ok, I know that this is the case for the later maf's that gm used on engines like the LT1 that had a frequency output, but I was under the impression that the voltage output used with the TPI motor's maf was at least closer to linear.

Do you have any idea what the shape of the curve is (roughly)?

I'm not sure what you mean by 'scalar stuff', can you give me a brief description? (I figured that I'd ask those that know and was hoping that you'd answer, I got a lot of good thoughts about getting my crossfire running from you years ago. I haven't done any programming with this thing yet, but trying to decide if this is a project that I want to undertake or just go with an aftermarket ECM)

 

Actually, it would be helpful to me if someone knew about what the metering range (highest and lowest) that is typically seen on a stock TPI engine and the total range (again, highest and lowest, lowest especially) that the MAF actually can somewhat accurately measure

 

On "high impedence" injectors like ours, there is a "minimum" PW that the injectors need. This is the typical problem guys running large injectors (that are high impedence) - the injector size needed for their engine are so large that the minimum PW they need for idle is too low for the ECM.

Back to the drawing board.

 

There's a discussion going on the DIYEFI mailing list that has some information that may be helpful to you. Basically, there's a guy that wants to cut his injector pulse in half so that the injector only fires when the valve is open instead of spraying it all over the back of the closed valve. Other than the MAF part of the deal, he's more or less brainstorming the same ideas as you: Bigger injector with a shorter pulse width. It seems like a good idea in theory. Whether there's any sort of real-world benefit, well, I don't know. I'm sure that with a batch-fire system, it'll be useless, but with SFI, it could be beneficial.

 

Actually, as far as that goes, I'm ultimately heading in the opposite direction, I'm trying to figure out how to get it to feed a much bigger HP engine, that's not the hard part with big enough injectors/fuel system, but maintaining a low end is. At different times I've been a member of the GMECM list, but found it difficult if not impossible to weed through the technical noise for the real meat of the answers, to put it bluntly, I have no aspirations to be an EE, I’m just good enough with a soldering iron and PCB to be dangerous, but I recognize that some of their knowledge could help me to get my mechanical systems working like I want them to.

Glenn, I understand what you're saying, but I don't entirely agree with it. I know that high impedance injectors have a slower response time, but I can tell you for a fact that I've seen two cars running 77# high impedance injectors (modified 36#, which have not such a great spray pattern so they had to be fairly close to ‘right’ for the nice idles and easy starting that I saw), that idled fine at 750-800 rpm (both fords).

Even if this would be a limitation of the GM ECM, I'm willing to cope with a 1000-1200rpm idle if I can get it to work since this would save me a lot of money over going with an aftermarket ECM, especially if I want to keep the factory niceties.

 

If you want to run low impedance injectors, LJ Stevens has built an impedance converter box for fuel injectors - plug it in line and your ECM can run low Z injectors. Couple of hundred bucks, plus the cost of injectors, but you get much better short pulsewidth control.

Just some random thoughts (aren't all of mine ?? *G*)

 

Um, I’m not sure where the confusion is, but yes, I know that a peak and hold injector driver could be built (basically just a current limiter), I know about the injectors firing 2x per complete engine cycle, and that the high impedance injector drivers react slowly limiting the smallest pulse width. Not to sound ungrateful, but I didn’t ask for any of that information.

I want to know about the 165ecm and the MAF. Really, I could answer my own question if someone could tell me
- what are typical MAF readings, (even just the minimum seen on a running stock engine, say for idle, I could see something like reversion in the intake having the value drop to close to 0, but that is not what I’m looking for)
- what range the stock MAF can actually read reasonably accurately (again, even just the minimum value)
- at what point will the ECM go, ‘whoa, there’s something wrong here?’
- and what is meant by the response of the maf is not linear? Does it curve up, down, waver all over the place… and is the translation into a volume/hour done by the MAF circuitry or the ECM (I’m guessing that either way the L/Hour output that something like Diacom reports is using the value that the ECM is seeing to compute a value that means something to us, but does the ECM see a linear relationship between airflow and maf readings or does it translate a non-linear output according to some table?

Thanks