[Dave2302]

OK,

Now I have 2 decent uncorrupted bin files to work with, from the RUTH and EXIT ECU's, what I need to know now is................

What do I do next ?? What is the procedure ??

I have deliberately avoided asking should I do "x y and z" because I'm probably wrong

So any and all advice from you guys who have created def files will be most welcome

Cheers Dave

[decipha]

first you'll need a disassembler that works with your binar/y(ies)

2nd you'll need to familiarize yourself with the intel 8065 opcodes

3rd it would be wise to run through the 8061 disassembly of the gufb / a9l calibration since yours are very close to those code wise, theres a GUFB strategy doc floating around here somewhere that can make following the code a little more predictable

4th, if your making the def for BinaryEditor then you'll want to open up the gufb.xls def file and look at how the def file is organized so you know how to put one together

just a side note, be sure to keep plenty backup copies of your progress along the way, if you make an accident somewhere or loose some of your work you'll be so disgusted that you won't bother with it for a while, at least thats how i get when i loose a bunch of work

[mpaton]

Everything decipha says is true. However, I think you need to ask yourself what it is you really want to do.

If you just want to tune your car, then you should see how much sailorbob or popsracing want for a def file, pay them, and get on with tuning. Even that is a lot of work.

If you actually want to lear how to make def files, then that's fine, you can do that, but it may well take you about 800 hours of learning.

For example, do you currently know how to read assembly code? For any CPU? Because you need to learn that.

Have you read any of the documentation on this site about how Ford structures its control algorithms? You'll need to understand them, because you'll be trying to relate the code you disassemble to the algorithms you expect might be there.

Do you actually know what def file is? If you don't, how are you gong to decide what you want in yours?

You CAN get help from people on this site, to learn all this stuff, but some of the people who may help you can also build you a def file for money.

People on this site have helped me, and I've some people here. But for me, they keyword is HELP. I would get fairly far, and get stuck on some subtle point, and then I could get help. I read all the docs here, and I used the very basic disassemblers, and I got hints and advice when I got stuck.

My point, harsh though it may seem, is that you've been here a few months, and it's always been "What do I do next?" People who could make you a def file in 200 hours or less (because they've done a few before) don't really have the time to lead you step by step through the process of making a def file if they don't see you've been making some progress on your own. Now maybe you have made progress on your own, and you've been getting hints in private mail; in that case, more power to you.

But I think you'd get more help if you told us here what'd you'd learned and done yourself, only there's just this little sticking point. Or if you write a convincing post about why you really want t odo it all yourself, and how you plan to learn how to do it, and how much you'll enjoy spending 800 to 1000 hours on it, before your engine ever starts.

[cgrey8]

He's not kidding about it being a long and steep learning curve. I delved into disassembly analysis just enough to understand what the problems were with disassembly and why disassemblers had such a hard time with Ford BINs. But once my curiosities were answered, I was less motivated to continuing the pursuit. At that point, I felt like I was learning something without a purpose given I didn't have the goal of developing a def. In fact, the only lofty ideas I had were to write enough assembly that a J3 based device with a powerful ARM/FPGA processor could take over as the EEC's brains making the EEC's Intel processor more of a co-processor. That's still something I think would be a neat project. But again, without real problems to solve that I'm not able to solve with conventional means, there's little motivation for me to take on such a daunting task. Although decipha, using his d.e.c.enstein module is probably closer to that than anybody. And then of course, there's MegaSquirt.

[decipha]

heres my notes for your strategy N4AG9 while i was looking for the imrc control, that will at least get you started on making a def file

fn018c c208
fn1361 37bb
act koeo max /12800 volts =c3e4
act koeo min /12800 volts =c3e2
act koer max /12800 volts =c3e8
act koer min /12800 volts =c3e6
act max /12800 volts =c3b6
act min /12800 volts =c3b4
adapt table =d4fa
fnbapxfr = c1f0
crank pw vs throttle table =d1a6 (2byte)
crank pw vs time table = d178 (1byte)
fn348a =d182
dashpot decay =cce2
ect/act transfer fn703 = c170
fn074a = cf0a
fn1362 = c4fa
fn1360 = c48e
fn072c = c2d8
fn072c = c2bc
fn022b = c288
knock retard x=rpm y=bap z=adv =c6b4
inj offset = cf40
idle ect adder = cc9e
spark act modifier x=rpm y=bap =e324
spark ect modifier =e378
fn1328 =d0dc
mbt spark =e148
spark mbt mpg table =e1b4
bdl knock table =e220
bdl knock mpg =e28c
spark table 3 =cdcc
spark table 3 alt = e3cc
spark table 4 = ce38
spark table 4 alt = e438
act default =c134
frcbft act/ect fraction =c45e
adapt max ect =d4f1
adapt min ect =d4f0
adapt max change =d4f3
adapt min change =d4f4
airmass flow multiplier =cef4
check_rom = 200a
ncnt crank exit min pips = cdaa
nrun crank mode exit min pips = cda6
crank timer limit = cfde
default crank time = cfdf
trans stall rpm speed for lhm = df54
sarchg = cef2 /6144.4688
idle dead band rpm = e5e8
dashpot gain = e5e0
idle spd cold start adder = c980
idle spd drive alt a/c = c97a
drbase = c978
nubase = c97c
idle spd neutral alt a/c = c97e
idle time to keep cold start adder = c9c0
injector slope = c42a /582.5422
inj min pw = c428
rpm = c00c
minmph = cd6e
delta tps pt thres = cdb4
bp max = c3fe
hysts = cdb8
thbp2 wot thres = cdba
hysts2 wot hyst = cdb6
rev lim drive = c446
rev lim neu = c444
rev lim hyst = c448
splclp = e0c0
spuclp = e0c1
spk_undsp = c414
pid_x bi fuel sw = c338
fn071 map scaler = c31c
table scaler abs exhaust pres bp/row = c25a
fn070 = c2a0
fn1320 ve table = cf52
??fn305 act ve mult = cf36
??vslim = c450
??vslimh = c451
??fn153 wot spark adder for rpm = e0ee
??fuel mult = c476
smarklock enable sw = c013

its an 88kb speed density eec-5 running a lot of the left over eec-4 code

if sailorbob chimes in he knows more about it than I ever will

good luck

oops my mistake, wrong thread!, thanks for pointing it out derek

[Dave2302]

1) Everything decipha says is true. However, I think you need to ask yourself what it is you really want to do.

2) If you just want to tune your car, then you should see how much sailorbob or popsracing want for a def file, pay them, and get on with tuning. Even that is a lot of work.

3) If you actually want to lear how to make def files, then that's fine, you can do that, but it may well take you about 800 hours of learning.

4) My point, harsh though it may seem, is that you've been here a few months, and it's always been "What do I do next?" People who could make you a def file in 200 hours or less (because they've done a few before) don't really have the time to lead you step by step through the process of making a def file if they don't see you've been making some progress on your own. Or if you write a convincing post about why you really want t odo it all yourself, and how you plan to learn how to do it,

5) and how much you'll enjoy spending 800 to 1000 hours on it, before your engine ever starts.

Apologies if my "simple steps" are offending anyone or if I seem to be taking a long time with this, I just don't think I want to be rushing these steps, would rather check and double check to make sure for example that I am working with a decent uncorrupted Bin file, which we have now established.
Also I'm not working full time on this, and yes this is all very new and puzzling to me, but we all have to start somewhere ;)

To answer some of your points:-

1) What I really want to do is to be able to fine tune an ECU on an engine that already runs fairly well but would benefit from changing a few things.
Sadly, that means I need a def file for at least the RUTH ECU (the one the car is running on atm). This means realistically that I will have to do it myself, with help from others.

2) I have asked lots of people including the 2 you mention "How Much" and so far only one says he is prepared to do it, although atm he is tied up on another project.
One of the guys you are quoting says he don't do defs for money anymore.
I do understand exactly what a def file is and does, and also have a full working knowledge of the EECV strategies, and which parameters it is that I need to alter to acheive my goals.

3) I've been here a few months, and the slow progress is because I can only spend time on this "part time" and also I was laid up Ill for a while in the new year.
However in that time I have learned an awful lot and made far more progress than I ever thought I would.

The reasons I want to do it myself is that realistically no one has given me a price to do it for me, or has the time to do it all, and also I will be learning a new skill, rather than just "giving up" and using say Megasquirt, Emerald, DTA etc.
That is the reason I do most things "In House" learning new things along the way.
I have made a fair commitment by buying the equipment and software needed rather than simply saying "If I send you my ECU can you read the bin, disassemble it etc etc"

4) My car runs and starts pretty well now (2nd place out of 30 car entry last time out), however it will definitely benefit from some tweaking.

When I started to scratch build the car lots of folk said "It can't be done" (Ford Puma Stage Rally Car........4X4 Running gear, V6 Cosworth BOB, etc etc) yet apart from the final paintwork and the alloy fuel tank ALL work was done by myself including for example buying a decent tube bender and building the Roll Cage from scratch.

My point is if someone is prepared to create me a def file for the RUTH then I am happy to pay them a reasonable sum...........If not then I will have to do it myself, however long it takes, and however much I need to learn, or give up (not words that exist in my dictionary)............
I thought the purpose of Forums was to share info and learn.
I have certainly helped many folk out via Forums, step by step sometimes, on transmission builds, engines, PATS systems, EFI Wiring etc etc over the years..................so if anyone doesn't have the time to reply to my "simple posts" and help me out, then fine, I'll not be offended, but equally all and any help is gratefully received................
And a big thank you to Sailorbob, Adam and Decipha and any others who have all helped me out so far ;)

I have done some of what Decipha said above and will post up a reply soon, but am busy with customers work atm ;)

Cheers for now,
Dave

[sailorbob]

One of the guys you are quoting says he don't do defs for money anymore.

I actually said:

I don't create one off def files, the time spent would be so long it means no one would pay the price.

Which is a little bit different

[Dave2302]

I actually said:

I don't create one off def files, the time spent would be so long it means no one would pay the price.

Which is a little bit different

In that case you have a pm

[mpaton]

My point is if someone is prepared to create me a def file for the RUTH then I am happy to pay them a reasonable sum...........If not then I will have to do it myself, however long it takes, and however much I need to learn, or give up (not words that exist in my dictionary)............
I thought the purpose of Forums was to share info and learn.
I have certainly helped many folk out via Forums, step by step sometimes, on transmission builds, engines, PATS systems, EFI Wiring etc etc over the years..................so if anyone doesn't have the time to reply to my "simple posts" and help me out, then fine, I'll not be offended, but equally all and any help is gratefully received................
And a big thank you to Sailorbob, Adam and Decipha and any others who have all helped me out so far ;)

I have done some of what Decipha said above and will post up a reply soon, but am busy with customers work atm ;)

Cheers for now,
Dave

I for one am not offended by anything you've done. I think it's great you're so determined; not many are, and many genuinely don't understand how big this task is, and so they don't understand how a one off def file could possibly cost 200 hours @ say 40 pounds/hr.

Assuming you haven't just hired Sailorbob to produce a Ruth def file, what symbols have you identified so far?

decipha gave you some addresses (although he seems to like bank addresses). Oops, no he didn't.

Have you found more?

Have you located and decoded the SCP tables? These are always a useful find on bins that have them.

If you want to tell us what your plan for disassembling is, then the community here including myself may have some suggestions for you. I was in the same place as you in 2008 and i had to completely disassemble a speed density bin with E4OD transmission control, and make my own def file.