Lets Talk About Injector Offets

m0l0 said:

On the mustang-tech you get these ones:

10 1.300
11 1.110
12 1.00
13 0.900
14 0.725
15 0.725

Click to expand...

These numbers (along with the injector slopes, breakpoint and minimum pulsewidth) come from Mike Glover's latest "Getting Started" guide with his permission.

I'll post some more on what the offset is actually doing (it's an adder) later on when I have a few minutes.

My take on tuning is that the high/low slope, breakpoint (if you are using different slopes) as well as the injector offsets should be dialed in FIRST while leaving the MAF curve as-is from your flowsheet. This should get you awfully close when you have the injector slopes dead on.

Then the MAF transfer should be leaned/richened in the WOT areas.

Many ways to skin a cat but this is what I like.

Wes

Here's a good thread on wrapping your head around the other injector settings: http://eectuning.org/forums/viewtopic.php?t=8471

Well

I may ruffle some feathers by saying this

I like EA as much as the next guy
but
It has kinda gotten to be a crutch for some ... as I see it

I would propose some may have gotten to the point that tools like EA
and the like have become a substitute for the basic knowledge skills
needed to tune in a logical manner based upon how the pcm works.

The offsets are nothing more than a way to + or - pulsewidth as the
batt volts rise and fall.

IMHO Peeps make too much about trying to make their tune .....
like mine
the GT
or
the Cobra

Some say keep the slopes the same
Some say they need to be different

It has been documented for years now :Word:
the slopes and offsets have a relationship with each other.

Lets use a bit of common sense and see at how the GT & Cobra
use slopes and offsets by looking at unmodified files of each one

If you look at a GT file you will see the offsets are much higher in value
If you look at a GT file you will see the slopes are different

I wanna be the first to say .........
each will work as I have tried each method of tuning.

IMHO ... over the years ... I've seen peeps get hosed up by trying to take
GT stuff and move it over to a Cobra file and vise versa.

anyway ..........

The thing to keep in mind is as the inj slope difference becomes greater
you will most likely have to use larger offset values

Now as for using my values ... my file is kind of a conglomeration of
t4m0 and j4j1
Hey ... I did warn peeps about using some of my stuff

I've had good success by using a known t curve to start with
then
Dial in all the inj stuff like slopes, breakpoint, offsets, min pulsewidth

I almost forgot ...

When working with batt volts and the offsets
I found it takes about 30 mins for the voltage to stabilize
and
I found the best way for me to approach the offsets was .........
ensure I gather data with the same accessory load :Word:

Consider what happened to me and maybe you can see my point

I had a tune that was very close and was in the fine tuning phase
I went out one evening and I saw my K's had taken a dump

I had been doing dlogs during the day with no accessories up until
that time I should make it clear.

The night I saw the K take a dump ... it was kinda damp out
I ran the head lights as it was dark
I ran the defrost (ac) to keep the windshield dry
and
the last half of the dlog I had to use the windshield wipers

I was all like
What caused these K's to be hosed

The wipers were the tel tale give away
as the K's were all the more hosed
in the last 50% of the dlog

Hopefully ... by now ... that little story should show you that ........
My offsets at batt volt values lower than normal needed work

I guess thats enough for now
any comments, Q's, or ... Grady ... you're totally out in left field

Grady

I got to thinking about this subject
and
Believe I should expound on something I said above

I said the offsets could + or - the pulsewidth

They can't actually take away from the pulsewidth

HOWEVER

Say you you see your K's a bit more active with their < 1.00 efforts in just one
particular spot on their curve.

You could see if it happens at a reoccurring batt voltage and use the offsets
to fine tune just that one spot on the curve with a lower value

So from that perspective you would have the effect of lowering the pulsewidth
but
In reality ... the new offset value wouldn't be adding as much as before

Hope I did not muddy up the water too much

Grady

Ok, here's my explanation of tuning via the slopes, breakpoint and battery offsets.

First off, let me clarify what we are talking about and where it is so everyone's on the same page.

Definitions:

Injector High Slope (typically a low number): This scalar is primarily in use when the car is commanding a high pulse width. Such as WOT.

Injector Low Slope (typically a high number): This scalar is primarily in use when the car is commanding a low pulse width. Such as idle. This is because as pulsewidths decrease, the injectors get inaccurate, erring to the rich side. By telling the injector it is larger than it really is, the pulsewidth is decreased to shoot less fuel and allow for this.

Breakpoint: This scalar represents the point at which the calculated pulse width using either the high and low slope are equal. However, the cut over between high and low slope is not an either/or equation, but rather it is a smooth curve that takes into account calculations from both high and low slopes. If your high and low slopes are identical, this scalar becomes irrelevant.

Minimum Pulsewidth: This scalar is the minimum amount of time that an injector can open. When you have larger injectors, they don't have to open as long to deliver the same amount of fuel that a smaller injector does. The pulse width must be lowered in this case so that the larger injectors do not stay open so long.

Battery Voltage vs Injector Offset: This function ADDS pulse width at a particular battery voltage. At smaller pulse widths (idle) this adder has a larger effect than at higher pulse widths.

Discussion:
As Grady said, there are two setups commonly in use - having the injector slopes the same (J4J1) or different (T4M0, U4P0, W4H0). Neither is better than the other given the correct numbers to support what you are running.

A lot of guys run the same slopes to eliminate having to mess with two injector slopes and the breakpoint.

Regardless of what you choose, either can be made to work fine. I think MAF curves get a bad wrap not because the curves are off by a lot but because the injector settings are off in the tune causing the car to run different than is commanded.

I am very surprised that noone has come up with a silver bullet config of settings for the different injector types - I'd love to see what SCT uses/changes when they load in an injector file.

Here's another good thread for reading: http://eectuning.org/forums/viewtopic.php?p=45042

Cliff Notes:
Different High/Low Slopes
If you have different high/low slopes, the high slope should be what your injectors flow at the pressure of your fuel rail and the low slope will be moved to a value greater than the high slope to lean out the fuel at lower pulsewidths. I use this method and started from the numbers on the mustang-tech wiki for 42lb injectors on my car and it got me very close.

Same High/Low Slopes
Your injectors are both set to the same value (rated flow rate). Your breakpoint scalar is irrelevant. It is critical that you have the Battery vs Injector offset table correct.

What Wes Said

Nice Job my friend

I'd like to add something

It took me a while to see this when I first started with all of this

We talk about ... low pulse width conditions ... and naturally ........
We think ... idle

If you've never dl'ed maf volts before ... do it
and
Drive around under ALL different kinds of conditions

You'll quickly see
a steady cruise on a level road
will put you in a low pulse width condition :Word:

The maf volts are just another way to see even though the rpm is not
at idle and the mph is not at Zero ... well .............
I think you see my meaning ... I hope

You see peeps all the time talk about bucking and the like

See where I'm going with all this

I ... Like Wes ... keep the bulk of the CL portion of the t curve with factory
values ... but ... a LITTLE tweec here and there in the curve is not gonna
bring down the wrath of the Self Tuning gods upon you

I encourage you to try a thing or two on your own
to see what happens by how the pcm responds

Grady ... EA ... Wes ... & all the others ...........
they don't have all the answers

Its when you think about and try little things .......
thats when you see your knowledge grow ... big time IMHO

Grady

Great information guys!

I had to richen my maf in the upper parts 15%. I was running 13.2 at WOT, now I am closer to 11.3-11.6.

So I want to change my curve to stock and use the high slope to fix my lean problem.

Currently my high is at 41.9992.

Is it just a shot in the dark to see how much I need to change?

I am guessing I need a higher number so that it flows more lbs/hr.

Would 43.5 be a good starting point, or should I move my high slope down, to around 40.5


If all that is confusing.

Is higher numbers richer, or leaner?

What's your breakpoint, low slope and battery voltage vs injector offset? All the numbers work together.

Here is a good start:

Breakpoint 1.66
Minimum Pulsewidth 0.0004
High Slope 40.32
Low Slope 49.68


Battery vs Injector Offset
15.9375 0.81
15 0.81
14 0.91
13 1.03
12 1.19
11 1.34
10 1.59
8 2.38
6 5.91
0 5.91
0 5.91
0 5.91

m0l0 said:

All the information added is really amazing, but as to all of this tweecer thing , i have learned that there is no the perfect tune , all the outside factors can bring different results on all the parameters as also the combos they will all be different, so just get your info straight and start slowly. Each one of us have different ways to see the whole picture of the tweecer.

Click to expand...

Good info

I would have to agree

I do believe some rule of thumb kinda things are the same for all
cause
The pcm works like it works

That can't change no matter if
NA
Forced
302-347
etc

I've seen a lot of different things come and go since I got started

The first efforts most did when everybody was getting started was
working mainly with the t curve.

Hey ... it worked but was kinda crude

The next big change was when it was discovered you could work less
with the t curve and use inj values for a more stable CL tune.

I've not seen as big of a change since if you talk about methods of tuning
other than ...
Programs such as EA & BE makes it to where you don't have to manually
do a lot of the work with formulas, calculators, and the like.

So ... while everybody gets all excited about how EA can help you dial
in your inj values ... that is nothing new ... just more user friendly ...
and ... I'm all for that

As for some rules of thumb

If you gotta hack your t curve a whole lot
or
If you find your load factors really way off

One should address major things like that
before
tweecing other areas of the tune
or
One might end up doing more work than necessary

Lets don't talk about how I learned that

Grady

high 41.9992
low 47.9988
breakpoint 1.00136
min 0.000400543
15.938 0.5
14 0.5
13 .5625
12 .7812
11 .9062
10 1.2813
9 1.5
8 2
7 3
6 5
5.5 7
0 7


Those numbers are way different

You're not too terribly far out of the ballpark from me with the settings.

The offsets I am using are a tad richer but at WOT the effect of what they add won't be felt very much. The offsets you had look similar to the ones run on the cars with the same high and low slope (14/15 volts at 0.5).

The offsets are really felt at lower pulsewidths where what they add makes a bigger difference.

I'm making up numbers here but in other words if the offset adder is 0.5 and your pulsewidth is 1 at idle, 1+0.5 is a big difference. Now your pulsewidth is 15 at WOT = 15+0.5 is no so big of a difference.

If you are 15% out I would look at making sure the MAF transfer is accurate and making sure the fuel pressure is where it should be throughout the run, you could be running outta fuel pump?

Another good way to get ballpark on the MAF curve if it is of questionable accuracy is to turn off adaptive strategy and get in 3rd gear (or on a dyno) and hold the car steady in 500rpm increments - see if you are lean/rich (according to a wideband) and adjust the MAF voltage points accordingly.

As said above, every car is different and there are several ways to tune.

BUT - my thoughts on tuning are to start with the base file of your processor, load in the injector settings and as accurate of a MAF curve as you can get then tackle closed loop driving to see how far off you are with your settings. Get closed loop within 5-10% on the KAMRF's and then tackle open loop.

After that is done you can go back and fine tune things but I like to get the car in the ballpark at all throttle positions from the get go so that you can at least enjoy it.

Also, there are some tables that add spark/fuel based on IAT - if your air gets hot you could be running really rich with really high spark so be sure you disable them or adjust them accordingly.

Wes

I have read that entering the maf curve in caledit does not work really good. I tried doing it with EA but got sick of messing with it.

When I get some free time I will try again.

One thing I've noticed (even tonite messing with it) is that CalEdit will skew the numbers I was importing. Be sure to watch that. The numbers I was importing were around 50-100lb/min different to the LEAN side!

Wes

I just calculated with EA and my curve looks around 4% leaner than my curve I manually made 15% richer. According to my last log with the 15% I was running 11.2-11.6sh, so this should be close to perfect..

I guess I can try messing with my other settings if needed.

Are you using EA to try to calculate changes to the upper end of the MAF curve?

EA is using your fuel trims to make adjustments. You won't have much, if any trimming taking place in the WOT areas of the curve. Once your injector settings seem ok, you can fine-tune the MAF to get where you need to be.

If you are commanding 11.6 and it's only about 5% off, then that sounds in the ballpark.

For instance, on my MAF curve, with the injectors dialed in, I was about a half point to the rich side at WOT. I then leaned the MAF as needed at those flow points and got it where it needed to be.

Just a lotta logging, changes and relogging to see the effects of what you changed - sounds like you've got it.

Wes

No, I did the MAF fitting to make my maf transfer, instead of using calcon. (using my 9 point from flow sheet)

I am thinking this could be a little on the lean side 11.75-11.95.

Should I lower (raise, still confused what makes it richer) my high slope to richen it up at WOT, just to be safe? Or should I just go ahead and datalog and richen the curve alittle if needed?

To richen the curve, you want to RAISE the lb/min or MASS#/tic numbers (Binary Editor or CalEdit units respectively).

If you think about it, you are telling the computer you are flowing more air at a certain voltage, so the computer is going to compensate with more fuel.

Wes