Fuel Injection Tuning

[clair.davis]

EFI Guys/Girls, Let's Talk Tuning.

I'm sure there are at least a few folks here that have DIY'd their own EFI or installed a kit that required some tuning. Let's get some nitty-gritty discussion going on what you've found to work for you on your particular setup. What I'm looking for is guidance or clues on where spark or fuel tables need some special attention to run well. Things you've found in your project that may have seemed strange but made a difference you could feel are important.

Let's also give some basics on our particular systems... I'll start...

Mine's GM-based, a batch-fired MPI from a 90-92 5.7L Z-28. Tuning is via software and an emulator that replaces the hard-coded chip until I get the tune right. Burn the chip, install, forget about it.

Engine is a 72 340, about 8:1 CR, cleaned-up 360 smog heads, Crane HMV272 cam, TTI headers and pipes, nothing wild at all. Made ~300hp at the crank (260hp @ wheels) with a TQ on a Stealth intake a few years back. EFI intake is a 4bbl MP M1 single plane, with a MP 1000cfm billet TB on top. 24lb/hr injectors, Ford F150 fuel pump, Aeromotive regulator, Lean Burn distributor... what else is there?

I need to sort out my spark table for sure, and do some more work on my base fuel table before I can start to fine tune my acceleration and power enrichment routines. I'd love to see how y'all have done things, and pick up any hints I can apply towards my project.

Thanks!

Clair

 

[clair.davis]

OK, for fun and motivation, I'll include a couple photos of my project as it sits now...

She ain't a show car...

Back when the parts were new and pretty...

Modified PS air box with ECM installed

Clair

 

[MadScientistMat]

I've got a MegaSquirted slant six Dart myself. Like you, I've got a Carter pump off a Ford F150 and a Lean Burn distributor with a GM HEI ignition. Currently, it's off the road for a turbo upgrade.

I usually pull a bunch of spark timing and get the main fuel table dialed in with a wideband. On a naturally aspirated motor, 12.6:1 is pretty good to aim for at full throttle, 14.7:1 or a little leaner at cruising. Then get the ignition dialed in. You can kind of copy a distributor advance curve, but I haven't found any real substitute for dyno time there. You also need the main fuel table dialed in before fine tuning acceleration enrichment. I don't think early GM ECUs do any model based acceleration enrichment, which really demands a well tuned fuel table, but even the regular sort will be off if added to a poorly tuned fuel table.

 

[clair.davis]

Woohoo! A real reply!

Hey, Matt,
That's my general scheme at this point as well. I used DeskTopDyno to model the engine as it sits, and that was pretty well backed up by my one chassis dyno experience. I took those curves and built the WOT end of my fuel curve, and it seems to work pretty well if a bit on the rich side. The engine runs GREAT at WOT, with AFR in the 11.8-12.0 range. I figure that's safe until I get the spark side sorted better.

In general, my tune will idle the engine well (14.5-15.0:1, 600-700rpm), and spark advance at idle is ~20* in that range. The cam really likes 20*, and would probably take more, I just don't want to have too big a change from idle to just off idle. Steady-state cruise is another issue, though. Even small transitions in throttle result in lots of bucking and popping, which makes me think there's a spark advance issue, but AFR is swinging wildly, too, according to the LM1. I think one of my next tuning steps is to get a fuel gauge that I can read on the dash (Auto Meter electric), and make sure I don't have a fuel delivery problem. I'll also pull back some timing, although I believe my timing curve is pretty tame. This is where I think comparing notes could be an advantage. Not that anyone's timing curve will be the same, but the general shape would be good to show trends. I'll see if I can save some images from my tuning software to show what's going on. Maybe I can get to that in the next day or so...

Clair

 

[MadScientistMat]

I could write a book on this... actually, I have, but it won't be out for a couple more months. A misfire will show up as a lean spike, so it's hard to say for sure if it's fuel or ignition, but I'd probably check the distributor phasing. As there's no advance mechanism physically moving the rotor, it's possible to line up the rotor in such a way that when the timing curve advances too far, the spark jumps to the adjacent cylinder. If pulling timing makes it go away, that's likely what is going on.