[killacal]

Greetings,

I just finished building a single turbo setup for my 89 stang. I'm blowing the cobwebs off of all my tuning stuff and am starting to work on a tune for the car.

Right now I'm working on tweaking Injector Hi / Low Slopes to get the AFR close, then will move into tweaking the MAF. Since the car is now seeing boost, I'm seeing loads and loadx's in excess of 100% when in boost. This got my thinking about how the motors "LOAD" or "LOADx" is taken into account when calculating the amount of fuel to deliver. I've been reading through this forum for most of the day and went back to read through GUFB as well, to gain some insight.

From GUFB, Base fuel flow (e.g. base fuel calculation dictating injector PW) is: BASEFF1 = (0.5*AM*KAMFR1) / (14.64*LAMBSE1).

With the car forced into Open Loop (as I've done) the KAMFR will be equal to 1. The "AM" is the air charge (this comes from the MAF transfer function) and the LAMBSE is set by the base OL fuel table (FN1307) and the WOT multiplier at WOT (which I've set to 1).

So how does load and loadx play into the fuel calculation? I know load will impact the desired AFR (i.e. using the FN1307 fuel table) and spark tables (FN901, etc...), but does it directly impact or add a multiplier to LAMBSE elsewhere?

It seems to me like once I exceed the max load in the FN1301, FN901, etc... tables it will just use the max value of that table and assuming I'm "happy" with the results there's no reason to do any load table scaling (albeit I can see how this would give me more control).

Is this a "sound" understanding of the impacts of Load and Loadx on AFR... or am I missing something obvious?

Thanks in advance,
Cal

The Car setup is as follows: GUFB/A9L, tuned using EA / BE and Quarterhorse, 306CID, 42lb injectors, single turbo setup as a draw through, currently using stock MAF sensor/housing to get tune close, will be changing to C&L or 90mm lightning MAF in the near future.

[decipha]

load has no bearing on fuel at all, absolutely zero none

[killacal]

Thanks Man. This is a point that should be plastered in a sticky, or in one of the many load topics.

[cgrey8]

load has no bearing on fuel at all, absolutely zero none

I agree that Load does not directly affect fueling. But indirectly, it does.

Regardless of whether you use FN1307 (Base OL Fueling) or FN1360 (Stabilized OL Fueling), Load% (aka LoadX which is a derivative of Load) is an input to both of these tables. And the output of these tables is the basis for what the commanded LAMBSE value is while in Open Loop.

And then you have transient fueling (EFTRFF) which is very short-term fuel enrichment based on change-in-Load%. It's minuscule, short lived, and for the purposes of this discussion probably can be ignored, but it does exist.

At stable Closed Loop idle and running conditions, Load is almost absent, but not quite. The HEGOs are in the driver's seat at this point maintaining the engine at stoich. However the HEGO Bias, Amplitude, and Delay tables take Load% as an input and thus how the EEC responds to HEGO behavior is altered based upon Load%. So while it has less influence and significance while in Closed Loop, it's still not completely gone from the fueling equation.


Thus I believe Load is very much involved in the EEC's decision of how much fuel gets delivered to the engine. I don't know if I'm being too literal and academic about this. But I did feel like it was worth my posing an alternate viewpoint with my rationale.

[decipha]

thats cool

but to clarify

load being at .1 or 1.9 will not affect the fueling in any way shape or form given the same maf flow and lambse your afr will not budge

[killacal]

load has no bearing on fuel at all, absolutely zero none

I agree that Load does not directly affect fueling. But indirectly, it does.

Regardless of whether you use FN1307 (Base OL Fueling) or FN1360 (Stabilized OL Fueling), Load% (aka LoadX which is a derivative of Load) is an input to both of these tables. And the output of these tables is the basis for what the commanded LAMBSE value is while in Open Loop.

And then you have transient fueling (EFTRFF) which is very short-term fuel enrichment based on change-in-Load%. It's minuscule, short lived, and for the purposes of this discussion probably can be ignored, but it does exist.

At stable Closed Loop idle and running conditions, Load is almost absent, but not quite. The HEGOs are in the driver's seat at this point maintaining the engine at stoich. However the HEGO Bias, Amplitude, and Delay tables take Load% as an input and thus how the EEC responds to HEGO behavior is altered based upon Load%. So while it has less influence and significance while in Closed Loop, it's still not completely gone from the fueling equation.


Thus I believe Load is very much involved in the EEC's decision of how much fuel gets delivered to the engine. I don't know if I'm being too literal and academic about this. But I did feel like it was worth my posing an alternate viewpoint with my rationale.

Interesting points, and I guess they all can have impacts on drivability.

I guess I was trying to be more direct in understanding that there is no direct "load" or "loadx" multiplier and that when you exceed max Load or Loadx shown in FN1307 or similar it defaults to the max table value (unless you exceed 200, then I assume all bets are off). I was also trying to get across that there's no direct load multiplier to the base fuel or timing tables (i.e. if I'm asking for 12.0: 1 AFR at 100% loadx, I'm going to stay at 12.0 when my loadx goes to 140%, but it will go down when I change within the temperature and loadx ranges shown in the table).

In my case I don't expect to be exceeding my load tables unless I'm at WOT, so perhaps some of this may have more of a driveability impact for someone with a bigger forced induction setup.

Thanks for the replies guys, great discussion.

[AussieSOHC]

I have a returnless system - static fuel pressure at about 60psi (?)
I have found no mechanisms of sensing fuel rail pressure or manifold pressure.
A factor in the pulse width calculation is pressure drop across the injector (?)
I have a nominal VE (load%) of 60% at 0psi and 102% at 7psi.
There is a nominal -7psi at idle,
So between -7psi (idle) and 7psi (WOT) the pressure drop across the nozzle goes down from 67psi to 53psi.

There are tables that describe injector flow correct for various pressures the like this:

Pressure
Drop Correction
60.0292 1.238
54.9609 1.184
50.0292 1.130
39.1503 1.000
30.0195 0.875
20.0097 0.714

What ManifoldPressure is subtracted from the static 60psi to get the index into the first column to get the correction factor.
Is this ManifoldPressure inferred from LOAD% or something else ?

[decipha]

its not manifold pressure its delta pressure, what u would actually have for rail pressure

If you dont have a fuel rail pressure sensor then you dont have a returnless setup, most likely you have an intank return style setup with only 1 fuel line ran from the tank to the rails

those functions are deltap multipliers, there should be at least 4 of them, hi slope, low slope, brkpt, offset

the multiplier is a direct multiplier on the base scalars value

look for a fuel pressure compensation switch scalar, if its value is 0 then those functions are disabled and you have a return style setup

[AussieSOHC]

You response is most valued - as I've been hunting for a way to understand this for a long time.

There is as you say only 1 fuel line and I believe there is a regulator in the tank near the pump that limits pressure to 60 psi.
There are 3 tables altogether very similar and at 39.15 psi (the pressure the injector flow is quoted at) they are all unity.

There is a variable "Injector Comp for Pressure" and it is set to 1.
1 flags pulse width is adjusted for according to fuel line pressure measured or inferred fuel line pressure by a calculation.

I am trying to understand how the ECU compensates for variations in pressure drop across the injector ?
I can get the overall mixture correct by using the MAF but that is not understanding that is achieving an out come.

I want to understand this because I want to use e85 instead of e10 and the 47# injectors are a bit short on the required flow at 39.15 psi.

There is a parameter "Fuel Injector Pressure Drop" set to 39.15.
If I set it to 45psi does make any difference ? to what
Does it get my the extra flow margin ?

[AussieSOHC]

To my understanding and observation (on my system which is generally described as Mechanical (regulator) Returnless):
there is no sensor to measure fuel pressure but it is known that here will be 60psi in the rail (the value of the mechanical regulator in the tank)
there is no sensor to measure the pressure or vacuum in the intake manifold.

To calculate the pulse width the pressure drop across the nozzle must be known.

The pressure across the nozzle is 60 minus the manifold pressure.
The manifold pressure is related to VE and calculated vacumm/boost.
6.5 psi of boost producing a VE of 1.04% hints that this calculation is not correct.
Vacumm/Boost that does not match gauge readings hints also that these calculations are incorrect.

Are there parameters to modify to alter the VE and Vacumm/Boost calculations?
I believe if VE and Vacumm/Boost calculations are closer to real world then I won't have to play with the MAF Transfer to compensate for fuel requirements when changing from e10 to e85

[AussieSOHC]

For mechanical returnless at engine off the pressure in the rail is 64psi.
With these systems the pressure in the manifold is inferred and added to the nominal rail pressure to give pressure across the injector.
So getting the load/ve/vacuum calculation to present values that correlate with a vacuum gauge are important.
I'm close but I've heard that moving engine displacement around to get calculated load/ve/vacuum to match gauge is not necessarily he correct path.

My interest is further to trying to get the correct injector data for DEKA 80 when they are running between a 80 and 55 psi pressure drop - 25 inch hg vacuum and 10 psi boost.

I am trying to get my no load and cruise STFT to be similar.
The no load STFT are notionally dialed in are 98% - 102%,
But at cruise and load they in the range of 115 to 125 ie trimming to get to stoichiometric.

From my description and the data below if anyone has a process to follow or some suggestions please.



DEKA 80
Note engine displacement halved explains slopes halved

0.0000122 Breakpoint
0.0104000 hi slope (x 2 x 3600 to give lb/hr) 75lb/hr (80 lb/hr is at 3 bar not 2.70
0.0140000 lo slope (x 2 x 3600 to give lb/hr)
0.0004000 min pulse width seconds
39.1600000 reference psi for slopes (2.7bar)
1.0000000 pulse width calculation uses inferred manifold pressure

BreakPoint Correction
80.000 1.4199
60.000 1.3090
54.960 1.1209
50.029 1.0842
39.150 1.0000
30.019 0.8756

HiSlope
80.00 1.477
70.00 1.379
60.00 1.260
50.00 1.140
39.15 1.000
30.00 0.867

LoSlope
80.328 1.419
70.000 1.333
60.000 1.221
50.029 1.140
39.150 1.000
30.019 0.844


STOCK corrections for Breakpoint LoSlope HiSlope all same
127.9 1.4590
73.0 1.0679
70.0 1.0458
67.0 1.0231
64.0 1.0000
0.0 0.4940

0.0000122 Breakpoint
0.0068900 hi slope (x3600 to give lb/hr) or 24lb/hr injectors
0.0073700 lo slope (x3600 to give lb/hr)
0.0000915 min pulse width seconds
39.1600000 reference psi for slopes (2.7 bar)
1.0000000 pulse width calculation uses inferred manifold pressure

[decipha]

those 3 modified function u posted are corrupt

[AussieSOHC]

They have been tailored from the Ford documented specs to match the test data I have - that why you say they are corrupt ?

Else
would indicate why you think they are corrupt ?
would you indicate what they should look like ?
would you indicate what affect this may have ?

[jsa]

The top and bottom entries need to be at max and min like the oem 127.9 and 0.