Ford EU EEC IV&V specificities

[jsa]

It seems really simple, but adding 40% (real injector difference) to what I think is Injector slope, generates really lean idle or stabilized mode (if not stalling).
Updating constants to raise [214] value, changes nothing, the main result is done by raising Battery correction [2a6] by 80% to stabilize everything.
On my daily stable version (which is not really clean), only VE tables and expected AFR are changed.
OK, to expect AFR 14.8 at idle to provide 13.5 at real and 15.3 at WOT to provide 12.5 is not the best thing to do.

The description of how it works sounds fair, but I am not clear on what you have done with injectors and slopes.
Are you saying you put injectors in the car that are 40% larger flow rate, then adjusted the slope 40% and got lean running?
Or are you saying you adjusted the slope by 40% on the OEM injectors and got lean running?

Is a fuel value being clipped or capped somewhere?
Are you able to log the various registers involved?

I agree, fudging battery etc is not a solution.

Feel free to send me your listing If you want another set of eyes to look over it. I will have a look as time permits.

[decipha]

if you want to post it here i'll give it a look when time permits

did you setup datalogging yet?

[Pym]

I am using 265cc injectors instead of genuine 190cc ones, on a self modified binary which I have based on a known stage 2 chip (still available for purchasing).
This stage 2 chip was written for 240cc injectors (without modifying slope) and with a maximum boost at 0.9bar. With this setup it is already too much lean and with mine clearly bigger injectors were needed.
As I have said, this binary is not well tweaked, my changes get better results, but I have reached the limit like this, this is why I want to restart from original binary and to update right information.
So the whole work is done on current injectors, 265cc ones.
Logically, I was thinking like this, raising by 40% the right single injector slope only, on original binary, should give good results and perfect results should be achieved with battery correction, but not at all.

You can have a look at 0FAB (0SLTXA) which is on the first page of the topic, it will match described registers.
Yes, datalogging is in place on mine, but I have to add more registers to detail different values in ticks, because I do no understand why battery correction has this impact, compared to other constants.
On my last tests, on normal voltage, original value at 123 was giving a 17 AFR at idle, for 297 it was a 10 AFR, function goes from 123 to 1426, for sure I have not tried the maximum one.

[jsa]

Ok understood.

Injector offset (volts>injector dead time) will mostly effect low pulse widths, so idle and low load stable running. What does your Lambda read as your load increases and more injector pulse width is demanded?

Say for idle, 2ms dead time + 1ms of fuel > 66% is no flow time. Injector offset dominates
Some load, 2ms dead time + 6ms fuel > 25% is no flow time. Injector slope dominates.

From your numbers
123 gives 1.15 Lambda
297 gives 0.68 Lambda
174/0.47=370.2
370.2*0.15=55.5
123+55.5=178.5

Enter 178.5, see what you get.

Is it possible the OEM MAP transfer is skewed to account for injector characteristics?
What is the Fuel Pressure Regulator rating and injector flow rate specs for OEM and your new injectors?

You said you expect AFR of 14.8, is your fuel really 14.8? More ethanol lowers AFR for lambda=1.
Do you have an AFR scalar?

[Pym]

The best thing I have obtained (in the garage) is 250cc equivalent on slope and 225 for battery correction, with other things updated too.
It was normal at idle and stabilized at 2500 rpm, but my feelings were not good enough to run like this.

Last week on road, I had only updated slope, at 210cc it was ok at idle and bit lean stabilized and I had 10 or 9 on AFR at WOT.
At 220cc car was not stable at all.

I do not think MAP transfer reflects injectors characteristics, it seems accurate and conform to specifications.
For fuel pressure, originally it should be 2.5bars at idle, I use 2.7bars because I can not go lower anymore with my fuel pump, knowing fuel line is limited to 5bars at boost.
S2 chip was created for 3.0bars.

Car was produced to use leaded fuel RON 97 MON 86 and unleaded fuel RON 98 MON 88 (and probably RON 95 MON 85, never tried).
Today in France RON 98 includes ethanol between 5 and 10%, RON 95 between 10 and 20%, so it has clearly an impact.
But I am talking about a CVH engine, AFR 14.7 is not a target for it, a stable idle for it has always been around 13.5 (CO between 1.5 and 2%).
14.7 could be a target in stabilized mode (for fuel consumption only) and hopefully it is not a Subaru, we can go over AFR 11 at WOT,
in secure mode (long run) it will be 12.5 and 13.5 for maximum results.

I have not seen a global or base AFR scalar, that is activated, but all related tables are easy to manage, standard results will always give a x1 multiplier.

It is interesting to talk about ethanol, because more and more people are now running with 100% of ethanol and I know some on this car with great results,
but with an aftermarket management and hours and hours (or days) spent on its setup.

[Pym]

Hi all,

no more work was done on 0SLTXA, since last year.
But I have worked on other strategies starting from 1992.

Where it is impossible to find equivalence for 0SLTXA (except on identical engines and cars),
this is not the same thing with zetec or duratec engines.
Zetec 1992=>1994 definitions are really near mustang ones, like CBAZA.
Duratec 1996=>2000 (like probably zetec) are compatible with mustang ones, based on years.

It makes the work easier.

[Pym]

Does anyone have examples about Pulsed Secondary Air Injection ?
It is used (at least) on zetec engines between 92 and 98, with some effect on exhaust.
I have removed the hardware, but it could be smart to do the same thing on software.

[Pym]

I am still searching for source of early European EFI strategies,
even if they were probably created in Germany or Spain,
because VSBAR is managed directly through Km/h.
But its first source could also be related, with the US Escort GT / Exp Turbo (1984/1985),
which are sharing the same engine and which are really near first EFI ones in Europe 4 years later.
Escort GT seems to be very, very rare especially for these years.
Have you ever seen one of this EEC IV ecus somewhere and why not its binary ?

[decipha]

I try to stay away from the 80s stuff when possible. I couldn't imagine the 80s code to be too difficult to figure out.

I don't have any early escort bins.

[Pym]

I try to stay away from the 80s stuff when possible. I couldn't imagine the 80s code to be too difficult to figure out.

To stay away from 80s stuff, I am looking at LUX0 strategy, which is semi sequential like the one I am working on
and because I have heard about DA1 turbo projects based on this ecu.
Do you know more about this one ?

[jsa]

LUX0 defs if you don't have them already.

https://eectuning.org/forums/viewtopic. ... x0#p132340

https://eectuning.org/forums/viewtopic. ... x0#p132336

[Pym]

Thank you, but I think this one is much more up to date.
viewtopic.php?f=99&t=3222&p=44957&hilit=lux0#p45104

[jsa]

Cool

[Pym]

Anyway John, after analysis, LUX0 has nothing common with early Euro EFI strategies, but is a good base for naming elements.
The most interesting strategy is for me 1WNBX6 (this is you, which had provided me a binary),
it is an Euro one too, but based on Pinto engine, using batch firing or semi sequential injection too,
with probably all injectors opening at WOT. This is Escort RS2000 MKV.

[jsa]

I will look at what else I may have for it tomorrow.

Xanthe comes to mind, maybe.

[Pym]

I will look at what else I may have for it tomorrow.
Xanthe comes to mind, maybe.

Thank you, but I am pretty sure that you will find nothing.
Xanthe has worked on zetec engines, which have strategies probably coming from US ones,
and which have given the base for GHAJ0.
This one is a pure European one and because,
it was not sold in a big volume (thanks to RS Cosworth ones), not so much people have had a look at it.
But we can have a look at the Scorpio, which is using the same engine and why not at the 2.3L used on Scorpio and Transit.

[jsa]

Well I can recall stuff from 5years ago........
I shamelessly downloaded 1AGB from a post by xanthe, here;
https://eectuning.org/forums/viewtopic. ... 90#p118654

I see FALL now as well;
https://www.eectuning.org/forums/viewto ... 73#p118973

http://xanthe.arkku.net/?p=09

*RS2000 AFAR or DADS ecu can�t be replaced by older ecu 1AGB (nonPATS)

I did a disassembly run on the 1AGB but nothing further as it was quite different to others I had looked at.

[Pym]

Just discovered a strange Escort MKV near my home.
A 1994 car, using the same type of engine, that had theoretically disappeared end of 1992.
ECU is "89AB 12A650 BC", "9ABC", but this one manages a lambda sensor, compared to 9AAA.
This type of strategies (the ones I am working on since a long time now),
seems to be able to work with adaptive things, but I am not able to get its software for now.
I have also discovered the same type of engine
(with some updates and also known as 'LEAN BURN') with a 1994 (94AB...) ecu
and I will try to see if this one, is still based on the same strategy type or
how it was updated to next generation appeared in 1992.

[jsa]

Looks like Fiesta and Sierra cvh as well.

https://www.ebay.co.uk/itm/Ford-Fiesta- ... .l4275.c10

https://www.ebay.com/p/6013254695

Hardware code turns up a range of uses
https://www.google.com/search?q=efi-sd102

[Pym]

Yes, Fiesta, Escort, Sierra and Scorpio are in the same case,
from 1985 to 1994-1995, when using EFI on Pinto and CVH engines.
A dedicated type of strategies, for European cars,
which was replaced on Zetec engines arrival, by a mustang based strategy.
You will say that Escort Cosworth uses a Pinto engine,
but with the Zetec strategy type and this is right, because of year probably or injection type.

[Pym]

Just ended first analysis on this 94 management (based on SME-105).
For information, engine is a CVH one with full sequential injection,
using a MAF sensor and a camshaft position sensor.
It is the only CVH (in Europe) with 4 injectors fired distinctively.
No surprise, it is based on Zetec management.

As conclusion, I can say that early Euro EECIV managements, designed for EFI,
on Pinto and CVH engines, were coded to work only on a MAP sensor base and
on half sequential injection systems.
These strategies were used since 1985 until 1992.
They were replaced by Zetec style ones centered on MAF sensor and managing fully sequential
injection, based on a camshaft position sensor.
Escort Cosworth EECIV management, Zetec style, seems for me,
the only one to use both MAP and MAF sensors.

If you find software for ECU 89AB 12A650 BC (9ABC), it will permit to confirm,
where and how Lamba sensor is managed on Pinto/CVH type.
Another software 95AB-12A650-DA (AFAR) or 95AB-12A650-DB (DADS),
used on a Pinto MAP sensor based engine, could confirm
usage extension of early type until 1997.

[jsa]

Half sequential sounds like bank fired.

MAP on the ESCOS is largely for fault monitoring of MAF, not for primary control.

I have a RS2000 DADS box here. 95AB-12A650-DB, finis 1011493, ESD-402
Also Zetec 1.8 CAST, SME-405, 95AB-12A650-EA box
Also RS2000 BINS for 1AGB and FALL
Do you want any of these BIN's?


The 90's RS2000 is not a pinto based engine, it is the I4 DOHC.

[Pym]

I say half sequential, because injectors are paired and it is not batch firing,
so probably semi-sequential or bank firing is the right expression.

Yes if you can read the DADS software, it will be a great help, because RS2000,
even on DADS in 1995-1996, was still not using a MAF sensor.
It could be the last of its generation.

For the others, this is ok, I have them.

When I talk about Pinto, I talk about its bottom end.
If I remember well, Sierra/Scorpio 4 cylinders SOHC or DOHC (not CVH),
Escort MK5/6 RS2000, and Sierra/Escort Cosworth
are all based on the pinto 2.0l bottom end, on its bare block, no ?
2.3l looks like that too.

[Pym]

I have updated my post about the SME-105 / CVH.
I have new information, it is a full sequential injection one and CVH engine has a camshaft position sensor.

[jsa]

Ok will dig dads out over the weeked.

The 2.0 EAO SOHC 'pinto' was fitted to the Mk1 & 2 Escorts, Cortinas, Sierras and what ever else from the 70's and 80's. Also fitted to a certain disliked USA namesake.

The 2.3 LIMA SOHC was fitted to USA cars. It looks similar to the EAO but is longer with a wider bore pitch.

Cosworth developed the EAO into the Cosworth YB series by strengthening the block and internals, and adding there 16V DOHC head.

All the EAO, LIMA and YB engines are belt drive cam timing.

The I4 8V and 16V are all chain drive cam timing. They are not like the EAO, LIMA or YB. They are a different design. So the RS2000 engine has very lttle in common with the RS Cosworth engine. Things like coil packs and some sensors are the same between YBP Cosworth and the RS2000. The YBT and earlier use a different sensor package.

[Pym]

Really interesting, I was putting everything into the same 'Pinto' package, now it is more detailed.

[jsa]

DADS bin attached.
Minor differences compared to 1AGB and FALL.

[Pym]

DADS bin attached.
Minor differences compared to 1AGB and FALL.

Thank you John. Yes, I am surprised too, to see it is still the same strategy 1WNBX6.
So 95AB-12A650-DB (DADS) is certainly the last EEC using this pure MAP management.
Not sure, this ESD-402 has real hardware improvements, compare to 91AB-12A650-GB (1AGB) on EFI-SD202,
apart from its external PATS ability, which should use a dedicated set of chips, because code is not affected.
So ESD-402 ends this famous family, composed essentially with EFI SD1XX and EFI SD2XX EEC,
started in 1987, following VM1XX EEC (started in 1985 and finised in 1990).
Now it is time to finish their definitions.

[jsa]

Certainly a long running strategy. Bean counters obviously curtailed more money for the RS2000. How far back have you found the strategy goes?

This particular one is a brand new service part, manufactured sometime on or after 2004. So IC1 and IC2 might not be representative of those fitted to the cars originally.

You can see the PATS chip in the pics.

[Pym]

The first version, for this type of strategy, from my stock, was built in 1988 for a Sierra DOHC.
This type requires EFI, MAP Sensor and EDIS.
I have no Scorpio dedicated ECU, but I know EFI started with them on VMXXX.
But early VMXXX I have seen, are the previous version of this type of strategy.
They are badged EFI, use a kind of MAF sensor without EDIS, ignition is done through the distributor.

[Pym]

1WNBX6 strategy is really interesting, I though this type of strategy was not able to manage HEGO sensors,
but this one manages 2 HEGO sensors and this pulsed air thing.
First time I see this type of strategy working with adaptive table. I just try to understand why this one uses 3 times 13x13 tables.
The code for that has some common parts with zetec versions, which helps.
Knowing that at least one 1989 strategy (9ABC which I still have not) was managing 1 or 2 HEGO sensors,
I have searched the same type of code in another 1991 strategy, and it is present using 1 table only,
even if HEGO was never used with it.

I will continue back to 1988 to find which ones are managing that.

[jsa]

Is the hardware present switch on for both hego's?

[Pym]

I have not seen any dedicated hardware present switch for hego's, even on more recent strategies.
On 1991 strategies, a flag can activate/deactivate hego's/adaptive processing, yes, but it is a global one.
It is on on 1WNBX6 and off on others were hego's are not present. It is B6_R91 <=> B6 on 0x2403.

[Pym]

Definitively, this type of strategy has no real interest.
It works, but engineers have lost much time to setup cars properly,
just to adapt what software can not include properly.
Hopefully, this strategies have only be used with 2 different injectors sets,
they have probably abandoned with the 3rd one.
A basic injector upgrade can be done in 2 ways.
1. Update what looks like a single injector slope, but this is not enough.
FN1320 table has to be reviewed entirely, because this 'slope' has to be changed, based on load.
Finally, when using injectors really different, FN1320 use will have to be patched,
because it is basically near its limits.
2. Play with this slope or not, but update on desired lambda tables is required,
still because this 'slope' has to be changed, based on load.
In all cases the result (visually) is terrific, slope is never really related with injectors,
VE table means nothing or desired lambda value is really far from reality.

I have now to replace this type of strategies, with another one, not so easy.

[decipha]

numego is the o2 present switch its on all fords from the 80s til current 2020s

[Pym]

numego is the o2 present switch its on all fords from the 80s til current 2020s

I can find NUMEGO present in this type of strategies, since 1988 (first version I have).
Even on cars without HEGO sensor, values are 1 or 2, which is what was possible on some versions (very rare).
Use of adaptive routine and globally everything related with HEGO sensors, is always activated / deactivated by the B6 coming from 0x2403.
Some strategies (still with this type) have no adaptive routine at all and no NUMEGO too.
Do not forget catalytic converters were not required before 1993 in Europe (globally).
The cars produced before 1993, using HEGO sensors were designed to stay on market until 1996 probably,
the others had not this requirement and were never equipped with HEGO sensors, their production was stopped end of 1992.
In my opinion this type of strategy, is a pure European one, created in Germany or Spain (I do not know where was the R&D par, at this moment).
The best proof, is just speed, with an output in Km/h and the temperature, with an output in °C.

[George 92]

In ANTI binary, sensor conversion function works on 2 bytes, 16 bits, I can not say more on it.

I add another photo for EECIV SD130, from a Sierra 2L DOHC 8GAB (real EFI this time). Hardware is different, it could do the job, but Pin-out that I can see is not the same.
Strategy is not the same than for 0FAB, but it's close to it, really. Tables, functions & scalars are not in the same place, not with the same size, but everything is organized in the same way.

Hello!
I've tried to open this strategy with EEC Editor but it requested me the Definition. Do you have the Definition?
Thank you!

[Pym]

No validated definition is available for 8GAB for now, for now, but main parts are identified.

[George 92]

It sounds good!
When I start to learn doing definitions I want help you with them, if you wish!

[George 92]

[img]

IMG-20200618-WA0001.jpg

[/img]
Hi! I have offers for 2 differently calibrations for ford escort 1997, 1.6i zetec Cat. 1st is JEEP. 2nd is STAG. Both are EEC IV with PATS and 1 HEGO, MT. What ecu should I choose? Which calibration is the most closely to the Mustang ones? Thank you!