LTFT boundries

[pknowles]

What do the LTFT boundries do? I did a search and it didn't yield much. The only thing I can find is in the help file which says something about the FTC cells "covering" the table. I under stand the breakdown for cells 0 through 15, but I don't understand how this effects anything. The PCM has trims for each of the smaller individual cells on the VE table, so why do the larger cells set by the LTFT boundries matter?

[AnotherUser]

What do the LTFT boundries do? I did a search and it didn't yield much. The only thing I can find is in the help file which says something about the FTC cells "covering" the table. I under stand the breakdown for cells 0 through 15, but I don't understand how this effects anything. The PCM has trims for each of the smaller individual cells on the VE table, so why do the larger cells set by the LTFT boundries matter?

Good question. I know it makes a difference because I have moved them around, but I don't really know the dynamics behind it. Anybody?

[MNR-0]

LTFT boundaries specify the size of each Fuel Trim Cell. Fuel Trim Cells control fuelling under different loads over the entire RPM range.

I recommend changing the RPM boundaries by the same scale you changed your final drive ratio and leaving the MAP boundaries alone.

I went from 3.07 to 3.46, about 12% higher which sees me using 12% more RPM. So I scaled my RPM boundaires by approx. that much. Made a world of difference to the Fuel Trims.

You want to scale your boundaries to give you the finest levels of granularity for your normal driving style and RPM. This will give you stable closed loop fuelling characteristics.

At least, that's what I observed.

[pknowles]

Well stock mine are 2500, 6502, and 6503. Since the fuel cut off was around 6200~6400 that means I never cross the mid boundry. I guess I could change them to 1000, 2500, and 4000 to split the tabel up a little, but I'm still not sure what I'm doing by changing them. I have noticed that the trims in the large cells specified by the boundries tend to be similar over the cell.

[luv2spd]

What would you recommend for me since I now idle at about 50-55kpa? I like the idea of splitting the rpms to something like 1000, 2000, 4000 or something like that, but I'm not sure what I should set my map boundaries to.

[BQuicksilver]

Great Question! I've always wondered the same. There is a lot written about these boundaries, but I really never understood what functional purpose they served.

I'm *guessing*, but i think they have something to do with the tables the PCM refernces for it's instructions.

[MNR-0]

What would you recommend for me since I now idle at about 50-55kpa? I like the idea of splitting the rpms to something like 1000, 2000, 4000 or something like that, but I'm not sure what I should set my map boundaries to.

I would recommend setting your MAP boundaries to 57/67/77 as anything above 77 is close to WOT and goes to open loop. As for RPM, it depends on whether your PCM uses the extra fuel cells in its calculations. If its set to 6400+ factory then I'd say it doesn't, and you therefore less fuel cells to maintain fuelling in closed loop than later models.

[AnotherUser]

My theory has always been to change the MAP boundaries on 98 SD tunes because it uses a lower resolution VE table, and the stock MAP boundaries are way off.

[BQuicksilver]

Has initial question been answered yet?

[Scooter70]

I'm bringing back this old thread.

While doing some SD logging/tuning yesterday, I noticed that LTFTs lock to 0 anywhere above the lower level MAP setting. So I moved the boundaries up a bit to be able to get LTFTs in some higher areas so that I can get some # to change the VE table. Does this make sense? Since my car idles around 50kPa, I have my MAP boundaries at 65, 75, 85 and the RPM boundaries at 1500, 3000, 5000.

Does this sound appropriate?

Thanks,
Matt

[jimmyblue]

There is no point to fuel trim learning in power
enrichment mode. So wherever youe PE enable
MAP and TPS are kicking you out of FAM=1.000,
is about the highest MAP boundary you'd want.
Divvy up the space between your lowest normal
MAP and the PE boundary evenly. Similarly for
RPM, my opinion is you want the upper RPM
boundary a bit above your highway cruise RPM
or right at 4000 (where the airflow scheme
settles into MAF-only).

The advantage to finer FTC boundaries is that
each cell is then a "tighter fit" to a smaller area.
Stock F-body only uses 8 of the 16 and throwing
away one for PE leaves 7 to do all the work. If
you get 15 of the 16 being useful then that means
less slop in each, less wandering-about of the trim
values in response to driving history.

So log FTC, FAM, MAP and RPM over a normal-mix
drive, scatter-plot the FAM vs MAP and FAM vs RPM
and see the point in MAP, RPM where you have no
use for trimming. Then compress the range to be
sensible and all cells used.

Or just copy the late trucks or something, they seem
to be more sensible.

The tuning advantage comes in as relates to the
MAF / SD "blend". At idle it's mostly SD. Above 4000
it's all MAF unless you're blipping it. With a lowest
FTC boundary of 2500 you can't make out much
about either MAF or VE table fidelity. The one cell
is pushed back & forth accumulating a mix of VE
error (idle & off-idle) and MAF error (2400 RPM
cruise has a good chunk of MAF input). How can you
be sure which is contributing to the trim, today?
Answer is, chop it up finer for starters. Then you
know the idle cell is all about idle, not about idle
plus highway, etc.

[redhardsupra]

what's FAM? I can't find it on any acronym list.

[Dirk_Diggler]

Fuel air multiplier