lambse, kams... rich?

Mustangless said:

Got a little datalogging with BE and now have a few questions. Most of the entire log the kams were .99x-.83x (rich?.) My lambse was anywhere from 12.x-18.x.

Does 1 or 0 is CL?


Does this mean I am running rich? My wideband is not working, so I can't use it.

Click to expand...

Here is some brief simple info along the lines you ask of

14.7 is considered the standard for Closed Loop

> 14.7 is leaner
< 14.7 is fatter

LAMBSE is not your fuel ratio

LAMBSE is the value that the pcm is commanding at any given moment

KAMRF is a fuel trim OR your adaptive strategy at work is how I see it

KAMRF value of 1.00 is considered to be the standard for Closed Loop

< 1.00 means the adaptive is pulling fuel
> 1.00 means the adaptive is adding fuel

Now that you understand how it works

Lets see if the values you've seen in your dlogs make sense

Your LAMBSE values are fat
thus
Your KAMRF values show the adaptive is pulling fuel to compensate

Does this not make sense ... that is if you're a 94-95 pcm

OK ... OK ... I'm getting a bit ahead of things here

Trust me ... if you focus on the data events where you see the K's
active ... you are gonna see the LAMBSE commanding fat values

I'd have no comment on the 0 or 1 thing as I don't run that software
but
Inactive K's are usually and indication of Open Loop

Let me know if I made things better or worse for you

Grady

Mustangless said:

It really seems that I am running on the rich side. I was looking through my datalog and it looks like I get CL for a few seconds.(unless the CL is not datalogging right) That is, if 1 = Cl. I guess that could make my run richer.

I posted something about CL on eectuning.

Click to expand...

I saw your dlog on the .org site and was gonna take a quick look

Can you put up the REAL file

I usually just look at a dlog in its raw data form when I wanna get a quick
big picture of things

Grady

Mustangless said:

I dataloged with BE and that is how it saved it.

If you have eec anaylzer you can watch it like calcon shows it.

Click to expand...

Now I understand

Thanks for the info

I bet Wes or one of the other guys who have later versions of software
will get involved and lend a helping hand.

Grady

I bet your "Fuel - Injector Offset vs Battery Voltage" is throwing you off some, it looks like the stock J4J1 settings - is that what your tune is based off of?

Make another datalog with the params below. Get some good cruise, stop, start and into the throttle a bit (WOT not necessary):

• LOAD
• VBAT
• PW2
• PW1
• KAMRF2
• KAMRF1
• LAMBSE2
• LAMBSE1
• MAF
• RPM
• SPARK
• TP
• MPH
• ISCDC
• STATUS1
• AFR/EVP (if you have it)

EDIT: Start the datalog a couple minutes after you have started the car.

How confident are you that the MAF curve is right?

You are bottoming out your KAMRF's, something looks pretty far off.

Wes

Post up the MAF transfer or tune with the MAF transfer in it (BIN format if you can).

Looks like you just cleared the KAM and the car was relearning?

Do some more driving and then take another datalog. This log does look closer than the other, however.

Wes

Be sure you are filtering your data in EA when you log from CalCon - I filter RPM (0-7000)

Wes

The car thinks it is running lean.

The KAMRF is the adjustment it has learned (so 1.05 would be giving you the programmed fuel multiplied by 1.05 -- or 5% more).

The LAMBSE is the AFR that the car is actively trying to hit at that moment. When your tune is correct, the wideband and the LAMBSE will be very close to agreeing all the time.

Wes

You will see lambse active during CL and OL

The K's only come into play during CL

If you are wanting to work on the K's, it would not be of benefit to work with the
upper end of the maf curve as that is OL territory

Also ... You should not see the K's being active at throttle positions of 1/2 or more
as the load it too great to sustain CL at that much opening

I usually slip from CL to OL at throttle positions of 1/4 to 1/3 maximum

Here is where I think your frustration may be all about

I feel you are not quite grasping the basic functions of adaptive strategy which is
all about CL and how the K's are an indicator of being rich or lean.

You can work with the maf curve or the inj slopes but I feel the inj slopes are
the best way

I really never paid much attention to lambse when working with dialing in my
CL tune. I never paid all that much attention to what my wide band was doing
when working with my CL tune.

In a nutshell ... you dial in your CL tune as such

1) You load in your maf curve
2) Then you work with the inj slopes, offsets, and break point

You work with 1 then 2 ... but not 1 AND 2
If you do 1 AND 2 at the same time you are going to go in circles

The objective here is to bring back in line the adaptive strategy to the point it
does not have to make large adjustments.

You want the pcm's adaptive strategy to maintain 14.7 at all conditions during
CL. It will have to make adjustments to do that but you tune to were it makes
small changes.

The pcm controls CL ... not you or not some table ... Make Sense

1.00 K's are considered not fat or lean
< 1.00 and the K's are pulling fuel cause you are too fat
> 1.00 and the K's are adding fuel cause you are too lean

BUT ... UNDERSTAND THIS

You are not controlling this ... You want the pcm to do it

The whole CL tune thing is about small throttle openings and very light loads
and
Driving conditions that don't meet those conditions are all about OL
and
Then the adaptive falls away and the tables come into play
and
Yes ... this is the part of the tune you do want to control

Have I made things more clear or I did just add to the confusion

Hey ... this stuff can be a bit mind blowing from time to time

Grady

Brett, I emailed you a new MAF curve based off of your datalog and the MAF you posted.

I forget, did you say if you had the flow sheet for the MAF you are using?

Wes

I would follow this path of action

Load in the curve from the sheet
then
Move over to working with inj values

Use info from past threads in this forum and the notes in EA to work with
inj slopes
inj offsets
breakpoint
min pulse width if you run large inj's

After you get things fairly stable you may see a point of two that is a bit fat
or lean and you can tweec the maf curve at those points but most have
found the amount of change needed to be very small.

It can take a while to dial in your CL tune using this method
but
It will be the best way as it will produce the most stable results

Grady