WeathermanShawn

iTrader: (1)

Running Closed-Loop, MAF enabled.

Have read through search, but obviously need some fresher insight into any new methods or ideas to determine LTFT RPM/MAP boundaries.

The car has no vacuum leaks and is in excellent mechanical shape.

Stock RPM settings are 2500, 6502, and 6503, which makes no sense.

More issues determining low MAP values. The theory of setting your lowest MAP Boundary to just above idle does not make sense. I idle at 45-48 kPa with a 228/232 110 cam. But yet light-throttle and cruise can be as low as 28-34 kPa, part-throttle at 35-50kpa. Setting the 1st MAP boundary based on idle is missing a lot of cells.

To further complicate the issue, I am at altitude (5220'), so MAP's out here are much lower than sea level.

Do you set up RPM/MAP boundaries using engine load, airflow rates (g/s), or any other valid parameter?

Is the idea really to 'get them all into play', or does that approach leave you with FTC cells that have decel and accel both sharing the same cell?

In my case and the driving I do, the switching from one FTC cell to another causes me more problems with fueling than solutions.

And yes, I have just thought of turning LTFT's 'off', but for now I am just looking for some expert insight in a sound method to make it work.

Any constructive thoughts?

Thanks.

..WeathermanShawn..

jimmyblue

iTrader: (11)

You just want to slice it fine, and if possible put
your main driving load-points at the center of cells
(not straddling an edge). You might take a look at
the Commanded AFR histogram in Counts mode and
see where the closed loop "action" is (ignoring any
non-1.00), and draw the lines in a way that makes
eyeball-sense.

pHEnomIC

iTrader: (14)

What Jimmy said, also I believe I heard that the resolution of trimming depends on how big the range is between two intervals. For example putting it like 800-2000-5000 or whatever, there will be more resolution in the 800-2000 interval than the 2000-5000 one. One thing to keep in mind.

WeathermanShawn

iTrader: (1)

Thanks for the replies.

I am having more trouble with the concept of where to set the lower RPM/MAP boundaries.

I do not like the idea of setting your first MAP boundary based on idle MAP. With bigger than stock cams, your idle MAP may be ~40-60 MAP. But, driving at light or part throttle your MAP may be as low as 25-40 MAP.

Same issue with the lower RPM boundary. Coast to a stop in neutral, and I find those lower FTC cells trimming at no-load also trimming part-throttle driving behavior.

Perhaps I am over thinking this, but for 2002 LS1's there really is no sound documentation on how they set the FTC Cell boundaries. So, add a cam and attempting to engineer new boundaries is a challenge.

While I like the idea of having more cells involved in trimming, you are also faced with a lot of FTC cells switching as you are driving normally. Does it make for more accurate fueling, or does it ultimately complicate it by having more FTC cells switching more frequently?

Again appreciate the replies. It is somewhat of a advanced topic, so it is always helpful to gain further insight into how all this works. The rest of the tune is dialed in perfectly, so this is really the only issue that I can not figure out.

Thanks again.

..WeathermanShawn..

PlatnumStatuz

iTrader: (6)

why don't you run semi-open loop?

WeathermanShawn

iTrader: (1)

That is a good thought.

I tried it last summer, but had a few issues with the idle going lean. However, that is an option now that I have the fueling and air down.

Thanks for the suggestion. Perhaps I will give it a try again.

..WeathermanShawn..

jimmyblue

iTrader: (11)

My recommendation to use the histogram of
(say) WBO2 or NBO2, and look at the Counts
mode, aims at putting the most-common load
points smack in the middle of a FTC.

Idle gets its own cells, so you do not have to
worry much about the discrepancy between
cruise MAP and idle MAP with the cam. But you
would like to not be continually toggling between
two cells at cruise, just because you put the
cut-line smack-dab on your 80MPH motor RPM
or something. Every time you switch cells you
zero the STFTs and start over with a LTFT value
that's "yesterday's news". Try and make the
pockets be where you live.

WeathermanShawn

iTrader: (1)

Thanks Jimmyblue.

I'll try that. I am probably getting hung up on the word 'Boundary'..as if it has to be at that perimeter of the RPM/MAP matrix.

Your method essentially centers the FTC cells into the areas of most usage.

As it stands now 95% of us never use FTC cells 0-3 & 12-15 with the 32/57/77 MAP and 900/1400/2200 configuration.

Are you wanting to get those FTC cells into play also? You could do it by raising the lower RPM matrix higher, and lowering the MAP thresholds. Or are we starting to get into problems using FTC cells (occasionally 15) where PE may fall?

I will certainly try your method. I just question if the GM engineers really intended for all of them to be used. Certainly the 2002's LTFT boundaries are very different. It is sometimes surprising what you can change on a software tune package, and later find out that some functions appear to be 'hard-wired' into the PCM. I have seen that phenomena on Commanded Fueling and Lean Idle limits for example.

Once again, appreciate your help.

If you think of anything else, let us know.

..WeathermanShawn..