What is an Acceptable KAMRF?

vristang

15 Year Member
Mar 31, 2005
4,933
108
124
Seattle
How close do you guys get to 1.00?

+/- .01

+/- .05

+/- .10

Just wondering when I can call it "good enough" on the MAF Transfer?

I don't have a WB yet, so KAMRF is about all I have to go on, correct?

Thanks,
jason
 
  • Sponsors (?)


my datalogs have been around like .9998-.9989

I think you want to get as close as you can it. but .999x should be preaty dang hard to beat.

FWIW- just remember to reset the eec after a change to the maftransfer and give the eec some 15-20min. of idle/drive time to let adaptive kick back in and make its changes.
 
Man O Man Jason

I could write a book on your Q and just get started :rlaugh:

I spent a lot of time :(
Burned tanks of gas :( (50% cheaper then) :banana:
Generated tons of datalogs :(
BUT
Learned a bunch of invaluable stuff :nice:

The info in this post is based on using EA to adjust slopes/bp and evaluate those changes using the PW vs KAMRF chart.

I have produced charts with results hugging the ENTIRE 1.00 line with a variance of no more that 1% ... :hail2: to me ... NOT :nono:

That means absolutely ... not a whole lot :rlaugh:

You can waste so much effort here striving for the perfect CL tune :bang:

You start to make headway with the K's hugging the line and then, all of a sudden, it all goes out the window for no rational reason you can find :eek:

The porb is simply this .........................

There are too many variables that can change which will be unknown to you that will skew your results. At times, more than one of them can be in place and your results will be ....... really out there in left field :notnice:

Let me list some stuff that will effect your results to give you an idea and to make you think a bit so we can discuss this. I can assure you of this. You can save a lot of :bang: here by not doing the same stuff I did which was unfruitful
BUT
you gotta tell my where you are now are with this, what you have tried, etc, etc, etc :)

pcm reset
time of test run (dlog) from pcm reset
ect temps
state of batt charge
number of accessories active during test run
inj offset values
maf curve
driving conditions

Bottom Line here is ..................

You may get a chart that is close to perfect
but
stuff from the list will hose around with all your hard work
and
make you think you don't know squat about that little silver box

:banana: one day ... :fuss: the next

You end up tuning with a compromise to handle ALL the variables
but
You gotta discover them first :nice:

Let the discussion begin :rlaugh:
AND
Me thinks this one is ... gonna be good :D

Grady
 
final5-0 said:
Man O Man Jason

I could write a book on your Q and just get started :rlaugh:

...Grady

First, have to say that I love reading your posts :nice:

Between your post and blksn955.0's post, I should be able to get down around +/- .01 pretty easy, and if I actually know what I'm doing (haha) I may get +/- .001?

Here is the process I use for altering the MAF.
In the morning I clear out the KAM before leaving for work. It is a 20 min. drive to work and I data log the whole way.
On the way home I data log again.
Once I get home I load the afternoon datalog into EA and let it produce a new MAF transfer.
I load the new MAF in the morning and clear out the old KAMRFs, and start the process over.

My concern is that I have not been seeing very large changes in my KAMRF.
KAMRF1 is about 1.12x
KAMRF 2 is about .91x

Should both of these be moving in the same direction? Does this indicate a bad O2?

So, I am a long ways from where I need to be.

I have found out that I started with the wrong MAF transfer. The whole C&L meter mixed with a Cobra EEC has been a pain.:nonono: I'll skip the details as it will take a while to rant on this.

I am at work now, but can post more details when I get home tonight.

Thanks guys,
jason
 
I'm leading you astray here Jason :doh:

You plainly said you were working on your maf curve.

I hosed up big time and gave you info about :(

Dialing in your inj's

So Sorry about that cause you need a good maf curve before doing any of that stuff I talked about above.

My curve was not much of an issue and this is why :)

I used a ProM with it's provided flow data sheet to generate the 30 point curve. Many other peeps used that maf so plenty of examples were around to validate my efforts with the curve.

Back then, EA was not all that old and many peeps were using it to dial in their inj's. The peeps I deemed at that time to be in the know reported they had successful results with dialing in their inj's from using maf curve values which were unchanged from their 9 to 30 point conversions.

After the inj's were dialed in reasonably close, you could fine tune things with small maf point adjustments but the changes were slight.

I just said that to show my curve was not something I spent any time with for the most part.

I can say this about working with the maf curve to tune :D

My very first CL tuning efforts years ago (way before EA btw) was based upon maf point changes but I found I could get a more stable CL tune by working with the inj values.

Again Jason ...

So sorry for the fact I muddied up the waters here :notnice:
BUT
You do need a maf curve which is pretty close to start with :nice:

I'd much rather suffer looking a bit red faced on the forum than cause you a lot of wasted work by following info that is not correct for you at this particular time in your tune.

I'm guessing you are using EA to help you find a good curve.

In your first post you asked about not having a wb.

The K's and lambse values should be plenty to tell you what you need to know about CL. After all ... you want the pcm to do its adaptive thing to keep things at 14.7 but you want it to do it with the least amount of effort. This is where the oem narrow band O2's can be trusted.

One more quick thing :D

I've seen on more than one occasion, Clint talk about the order in which to tune and he also has much C&L experience.

When it comes to C&L mafs

I believe you need take this kind of reasoning ..........

Clint = :hail2:

Grady = :Zip2: :chair: :nono: :owned:
 
Grady - No need for apologies, I greatly appreciate the help. :hail2: :hail2: :hail2:

I think I have my MAF transfer under control, or atleast I will soon enough.

My problem was that C&L did not have any MAF transfers published for the Cobra MAS sensor. As I understand it the MAS from an A9L and the MAS from the X3Z are 2 completely different animals.
So... tomorrow I will be installing the stock MAS sensor I had on my 90GT. Then I can load up one of the A9L MAF Transfers and be fairly close to a good tune. I hope?

I tried to do a write-up/FAQ type article on it the other day. If anyone reading this has any experience in this area, please let me know if I understand things correctly.
Here is a link. http://www.freewebs.com/vristang/faqs.htm

Too bad that C&L doesn't make things easier for guys with Cobra computers :bang:

I think I understand what you are saying about making changes to the injectors instead of the MAF to get the tune right.

I'll will play around with EA tonight and see if I can figure out all the details of doing this.

:Word:

jason
 
Ok, I finally sorted out the MAF Transfer.
I ran a datalog this morning. From a stop light I ran WOT through 1st, then hit another red :(

ECT 184-188F
RPM 616-5896
VBAT 13.43v - 14.1v (headlights were on)

KAMRF1 0.972-1.07
KAMRF2 0.878-1

PW1 .623-17.14
PW2 .504-17.14

LAMBSE1 11.478-15.98
LAMBSE2 11.478-16.797

HEGO1 .200-.795
HEGO2 .275-.852

MAFV - ranged from .722v to 4.83v

TPS .98V - 4.26v (I may not have had the pedal to the floor?)

I know you guys can't do much via the internet, but do these numbers seem reasonable. I sure coud use a sanity check right now.

I am mostly concerned with the KAMRF #s

jason
 
A thought or two first, then a Q for you Jason :)

The K's are all about Closed Loop.

For the most part, Closed Loop is all about drivability.

Open Loop is all about, for the most part, WOT and you should not see K activity during OL.

I feel you tune for CL first and OL last or you could make for yourself, more work.

Your maf curve can be broken down into two parts, CL & OL.

My dl's show me I never see anything higher than 2.6 mv in CL. I have deemed that point as my breakpoint for CL & OL in the curve. That tells me to never hose around with the curve at 2.6 and lower when tuning for WOT. This is a must due to the fact I tuned for CL with maf curve points & inj values until the K's were hugging the 1.0 line.

I think it would be helpful for you if you determine all the driving conditions that make up CL. You want to do this as those conditions are all that matters when dealing with the K's.

The OL part of the curve is where your wb will show you how to tweec your maf curve and OL fuel tables to match for setting your final af ratio.

You need to get your maf curve on the money before other tuning as was said before.

Here is the Q :)

Are you using EA to setup your maf curve?

Grady
 
My average KAMs cruising go from .97 to 1.03 now.

At idle, they are around .93. :shrug:

As time permits, I will continue to tweak my MAF transfer to try to get closer to the holy grail of 1.00, but I am OK with it as is.
 
Grady - your right about the OL /CL tuning.:owned:
I realized that on the way to work this morning. Dang Noobies :bang: :)

I know better. :nono:


What I hadn't considered is effectively tuning 2 MAFs, one OL and one CL.
I'll look where my wot breakpoint is set, and work everything under that.
Should I consider adjusting the breakpoint up or down?

Per my numbers and those of blksn955.0 and BlackVert, I have a little MAF tuning left to do.
Grady, can you provide a little more detail on how to make adjustments with the injectors instead of the MAF Transfer?
I promise I won't use wot this time. :D :bang:
Actually I should have some good data after the drive home today. :nice:

jason
 
The maftransfers "breakpoint" seems to be based on several things... like tps and load with in the whole engine combo itself. I dont think its a set number like the inj. breakpoint that can be fudged alittle.

Since I run a factory inj. that came on the eec strat. (j4j1) I have a realy good starting point to go from as far as inj. go. EA seems to make realy small changes in the slopes and breakpoint when I ran a log threw it. That may change since I found out about my little mech. issue this winter though.

I would want to get the MAF as close as possible so that the inj. change would be more noticable or the only thing your realy putting a focus on...you may have to make a little maftransfer change but prob. not as big of a change as you do while first tuning the transfer.

Anyway, if I am off please correct me as this is how I see it working together.
 
Like Greg talked about

My using the word "breakpoint" in the above post when talking about the maf curve SHOULD NOT be confused with the scalar ... inj breakpoint.

I was only using that word to show there is a point on the maf curve that you won't see maf point voltages in Closed Loop. Above those voltages the K's have no bearing.

I can't say about Fox stuff, but for 94-95, load comes into play in determining how high you will go up the maf curve and still be in CL. Throttle position is also a factor here and I'm pretty sure that is in effect for Jason's Fox pcm.

As for shifting the point of CL/OL in the maf curve (or breakpoint as I called it above) it will just naturally occur. I just was attempting to show there is a point in the curve, and below, where you don't wanna tune for OL.

Hope I cleared that up :)

The most important thing I wanna point out here is
Establishing a maf curve to start tuning from
and
Using inj values to dial in your inj's

Those two things are different
but
They can and do work together


About the maf curve first ................................

I know EA can help in getting a good curve
HOWEVER
I've never done it so I leave it up to the EA notes AND other members who have experience here to show the way :)

About dialing in your inj's ................................

You need to have your maf curve established before this procedure is attempted. Look at the EA notes for the details but you are working with slopes, breakpoint, and offsets HOWEVER since Jason be a Fox kinda guy ... I'm not sure about the breakpoint for him :shrug:

At this point in the discussion ...........................

I'm not too sure if I've helped or hindered the most :rlaugh:

Grady
 
I am preaty sure the fox pcm does not allow for a inj. breakpoint to be changed...I remember it being talked about as "hard wired" or not changeable in the fox eec.

yet another small detail/reason not to go from the 94-95 (HORRIBLE :rlaugh: ) eec to the fox one...but I am not getting into that debate. Just making a funny:SNSign:
 
Let me try to clarify one thing that Grady said that might be taken the wrong way.

There are 4 things interacting with each other in this part of the tune: the injector slopes, the injector breakpoint, the injector battery offsets, and the maf. if any of them is changed, the KAMs are affected.

For example:

- increasing the injector slopes will cause the KAMs to go UP, because the computer thinks it is providing more fuel than before, so the adaptive logic raises the KAMs to account for that.

- increasing the values in the maf transfer will cause the KAMs fo go DOWN, because the computer thinks that more air is coming in than before (so it gives more fuel), so the adaptive logic lowers the KAMs to account for that.

- changes in the breakpoint and offsets are more involved; let's not talk about that for now, ok?

So you take a data log and the KAMs are not near 1. What do you change, injector slopes or the maf transfer? How do you decide which to change first?

Changing both at the same time is not a good idea. Nuff said.

So you have decide what you think is the most accurate ... the maf transfer or the injector slopes.

The maf transfer is usually considered to be accurate, because the companies thar make them run them on a flow bench and provide flow sheets that should be pretty much right on.

So the maf transfer is usually the one that people leave alone at first, because that is believed to be correct already. Make sense?

So when Grady says "establish a maf curve" i think he means enter the most accurate maf curve you can find for your maf (which comes from the manufacturer), and then leave it alone. I do not think he means changing it to try to get the KAMs close to 1. At least not yet.

Then take logs and change the injector settings to get the KAMs as close as possible. Start with slopes and breakpoint, then battery offsets. Naturally, start with values for those things that are believed to be good starting points.

Then make minor tweaks to the maf transfer to bring the KAMs even closer to 1.

Sorry about the book ... this is the one area of tuning I have researched very thoroughly.

Do you want me to spill my guts about the slopes, breakpoint, and offsets?
 
blksn955.o said:
I am preaty sure the fox pcm does not allow for a inj. breakpoint to be changed...I remember it being talked about as "hard wired" or not changeable in the fox eec.

yet another small detail/reason not to go from the 94-95 (HORRIBLE :rlaugh: ) eec to the fox one...but I am not getting into that debate. Just making a funny:SNSign:
the fox takes a different approach. it has a fixed a breakpoint of 2ms, but there is a function that it uses when calculating the desired fuel. this approach was change for our computers.
 
BlackVert said:
Sorry about the book ... this is the one area of tuning I have researched very thoroughly.

Do you want me to spill my guts about the slopes, breakpoint, and offsets?
No need for apologies, that was a great read:nice:
By all means, spill away.


I want to thank all three of you guys for the repeated replies. I may not have found the perfect tune yet, but I am learning alot. The help is greatly appreciated. :hail2: :nice: :hail2:

Some differences that seem to exist between the fox and the sn95.
I do believe that the fox computer enters WOT based solely on TPS. I have a scalar called "WOT TPS Breakpoint Voltage" Current setting is 2.7v, and I see no reason to mess with it right now.

However, is WOT the same as OL? Or is there some subtle difference between the 2? Does WOT trigger OL?
I have always seen these terms as somewhat interchangeable. :shrug:

I do have an Injector Breakpoint Function as well, which the computer uses to select either the High or Low Injector Slope.
What I don't have is a Minimum Pulse Width setting (I think you guys do?).

With that out of the way...
My KAMRFs below 1,000rpm are very steady at 1.121. Above 1k rpm, the K values vary, but mostly stay above/around 1.
I thought I read that there was a 12% limit to what the adaptive strategy could do, is that right?

Question...
KAMRFs over 1.00 indicate the tune is lean, and the adapt. strat is multiplying the pw by a larger than 1 KAMRF, to increase the pw? I hope I atleast have that right.

I don't know if I fully trust the support from C&L but I will leave the MAF alone for now.

Since the idle rpm range seems the worst, I will focus on that first.

The order in which to make changes.
1) Load a proper MAF Transfer
2) High/Low Injector Slopes
3) Injector Breakpoint
4) vBat Offsets

Once I have dealt with these, then I will turn back to the MAF Transfer.

Thanks for tolerating my random writing style.

jason
 
Yes Jason

You have the K's correct, > 1 is the adaptive adding fuel.

Your stated K values < 1krpm says the pcm is adding 12%

You can get to a total range of 6% without too much effort by using the dialing in inj's procedure we talked about. The closer you get to around 3% ... the more effort on your part is what I have found.

DO YOU HAVE EA?
DO YOU HAVE EA?
DO YOU HAVE EA?

If you do, look at the notes and a lot of this stuff will be more clear.

Search on the Tweecer support site for inj offsets for the size inj's you're using.

I'd load the offsets and work on the slopes first. See the stuff I posted above about the offsets and how accessories can change things.

Other than tps voltage, I won't say a lot about Fox OL/CL transition.

You can certainly find info on the Tweecer site as for every 10 threads posted, 8 of em is for those confounded Fox Stangs :bang:

That used to make me so mad in those early days :rlaugh:

Grady
 
oh, i see you have a fox. i assumed you had a 94/95 because of where the forum is. my bad ... :oops:

yeah, the details are different for a fox, but the approach is still the same. sorry, i'm not nearly as familiar with the foxes as i am with the 94/95, so my comments could very well not pertain to your situation

i also see you have EA. i have never used it, but i hear it is great. and it was written by a fox guy, so i would think its injector stuff and pretty much everything else would be especially good for people with fox computers.

there are at least 2 ways to get into OL, startup and WOT. WOT is one of the ways to get into OL, but not the only way.

and yes, over 1.00 means lean and the computer is adding fuel.

my advice? put in the known maf transfer and leave it alone for now. take some datalogs under similar operating conditions (ie - same accessories on, lights either always on or always off, this helps keep changes in the battery voltage out of the equasion) and run them through EA for its injector slopes calculator. once you are as close as you can get with that, then start on fine tuning the maf.

one other thought ... does the EA help file talk at all about an approach to tuning?