Well after looking at this (new feature in EFILive v6)



and knowing from version 1.3 of LS1edit LTFT boundries

RPM MAP*10
Low 900 370 or (37KPa)
Middle 1400 570 or (57KPa)
High 2200 770 or (77Kpa)


I think I figure out what fuel trim cells 0-15 mean

Main Ve table is sperated in to 16 quadrants.

cell 0 being below 900rpm and below 37kpa
cell 1 being above 900rpm, below 1400rpm, and below 37Kpa
cell 2 being above 2200rpm, and below 37Kpa
cell 3 being below 900rpm, above 37Kpa, and below 57Kpa
cell 4 being above 900rpm, below 1400rpm, above 37kpa, and below 57kpa
cell 6 being above 1400rpm, below 2200rpm, above 37Kpa, and below 57Kpa
cell 7 being above 2200rpm, above 37kpa, and below 57Kpa

I verified this with my EFILive Log data and it matches 110%

I hope I have the point accros!!!!

so what i'm planning to do tonight to fix my ltrims is average out the ltrims for each cell and adjust those coresponding quadrants with those percentages instead of the IFR.

If this prove right, I hope somebody can explain this more cleary because I'm sure I havent lol. :tongue:

 

maybe this image will help you understand what I mean better

starting from the top left going right,
first row, cell 0 - 1 - 2 - 3
second row, cell 4 - 5- 6- 7
third row, cell 8 - 9 - 10 - 11
fourth row, cell 12 - 13 - 14 - 15

 

DAYUM! Is this the holy grail?? I need to update my EFILiveV to VI and get Edit 1.3 going (it locked up my WIN98 laptop)so i can take a look at all of this. Sensory overload!! Congrats! Thanks for the info.
joel

 

I think you just blew my mind

 

After looking at this for awhile it sure sounds and looks solid. I wonder why EFILiveVI RPM boundaries are slightly different?? Example- 899 v 800(VE), 2200+ v 2400(VE). Maybe a dedicated FTC table?? Just thinking out loud(Typing as I think....how can I think out loud but say zip??)
joel FWIW 3.39 kPa = 1.0 " HG

 

Joel, the boundries are set in LS1edit 1.3 it's 900, 1400, and 2200. btw thanks for the kpa=hg conversion, I was wondering what it was

What I also thought about, is that everyone(maybe) averages the ltrims from cell 6-14, because that's the most operating ranges we see on our cars and everyone(or just me) wondered what cell 0-5 was used for. it's the least used area of our cars, that's why you don't see it all the time. my car idles at around 49kpa.

 

Oops, my bad - now I see it. Thank you.
YellowZ06- where can I get the V6 EFILive upgrade?? Is only available as a Beta?? I checked Starr and B&Bs websites and can't find it? Thanks in advance.
I looked at FT cells last fall and rarely saw 0-5 logged with EFILIVE. I was looking at RPM and MAP ranges by FTC, and manually writing them down in columns, when I realized very few entries 0-5. I just figured too few datapoints to include in averaging. We shouldn't need to average now!!!

My car idles 70 -85 kPa....I'm still trying to get it set up.......Actually I've made headway since asking NoGo for help. NoGo is THE MAN
joel

 

Very interesting work. I've always been a fan of ignoring the "ltft based on trim cell" approach, since we didn't really ahve a good understanding of the cells themselves, and we had no way to alter them individually. Looks like more stuff to play with. Good job.

 

Since I saved that pic for reference, but wouldn't know what cells were which later, I took the liberty of modifying it.

 

Talk about great timing! I have been suspecting this sort of relationship since I had noticed that it would tend to jump by four on cell changes, like 0->4->8->12, as you pressed down on the loud pedal. It also looked like the 4->5 transition was based on RPM. So, I decided to play on the way to work this AM and log the data. I also intended to post and see if anyone else had noticed or decoded or whatever. And I find this!

I haven't reduced my data yet, but I did notice that my car runs in cell 4 at 2050 RPM and 11 inHg, which doesn't fit your model. If I understand, your model says that should be cell 6.

 

This isn't making any sense. I made scatter plots of my data from my trip to work. I got nothing in what you call cells 0, 1, 4, 5, 8, 9, 12, or 13. Instead, all of the reported cell 0 use falls in what you label as cell 2, cell 1 use in your cell 3, cell 4 use in your cell 6, cell 5 use in your cell 7, cell 8 use in your cell 10, cell 9 use in your cell 11, cell 12 use in your cell 14, and cell 13 in your cell 15. In other words, my data is shifted right two columns. I wouldn't think the logger is reporting FTC usage incorrectly. I dunno.

 

Well that is very wierd, I just logged another one this morning coming to work and it verified 110% correct for me... can you email me your log file so that I can take a look at it? abedrossian@logicmate.com

Thanks

 

EFILive Home Page this is there new website. V6 is still beta, due to release very soon, I emailed them to get a beta copy but still waiting for it.

 

no can someone shed some light on what the numbers actually represent....

 

ugh...my brain hurts.

 

Seems like VE numbers are just a mathematical factor...i.e. no units assigned. Used in the Base Injector Pulsewidth Equation.
joel

 

Thanks- Big help.
***

Cells 3, 7, 11, 12, 13, 14, 15, and of course 22, are PE cells. So we are in PE Mode (Open Loop) for each of these RPM - MAP ranges????

***
Cell 22 differs from 15 in that it is TPS% dependent- WOT. Hmmmmm.....looks better and better.
joel

 

To convert the VE numbers on the tables that we see in edit simply divide them by 30. That will give you the actual VE number that they represent.

Just to chime in here. We should not use the VE table directly to make large changes to the LTerms. You have to calculate out your VE table properly. The VE values that we see are used in the LTerm calculation. Using the VE table to make large shifts in the LTerms does more to 'trick-f!ck' the LTerm calculation than it does to adjust fueling. Using the VE table to adjust LTerms will land the car at a target a/f ratio other than intended. As a result, you can end up with a car that sputters and such even though the LTerms are fine.

To see the actual target a/f that the car is trying to obtain at different MAP/RPM divide 14.7 by the "Open Loop Air/Fuel Multiplier Table" value. This table is actually the stoich ratio table, the title that edit has given it is misleading.

To properly determine your VE table you can use either a wideband O2 sensor, with a data logger, in corrolation with the target a/f values given above, or use your MAF and some of the math below. To deteremine the relative VE values through the use of your MAF use the following math.

Determine Theoretical Flow:

T Flow = ((RPM * (Engine Size ft3))/2

Convert to Mass Flow:

T Mass = T Flow * (MAP Kpa / 101.3) * .0780

Determine Actual Volumetric Efficiency:

VE Actual = (MAF lb/min) / T Mass

Determine LS1 Edit Volumetric Efficiency:

VE Edit = VE Actual * 3000


Drive around ALOT logging ONLY MAF, MAP, RPM, IAT. Use only those values that have roughly the same IAT (+/- 10 degrees). Additionally, average values repeatedly for cells. You won't be able to get all of the values, so you will have to interpolate some. Use the stock tables as a reference. Most values are in a linear relationship up until the higher RPM.

Once finished, if you are a lightly modded car you values should shift down in the lower RPM cells (because of your cam) and your higher RPM values should shift up slightly.

This is how I do it to remove MAF's and such from vehicles. If you have anything to add, or if you have caught a math error from above, please chime in.

Good Luck with the VE tuning.

 

I don't use EFILive, so the log probably wouldn't do you much good. At least part of the difference is that LS1Edit reports my LTFT boundries as: low - 2500 / 320, middle - 6502 / 570, high - 6503 / 770. That would explain why I saw my transitions from one cell to the next at 2500. I didn't have the guts to raise the rev limiter and see what it transited to at 6502 I have a 2000 Z28, so perhaps the difference in the boundry table is normal. I may try moving the ranges down and see what happens.

 

Thanks NoGo! That sheds a lot of light on things.