ECU rule thoughts

[dickita15]

Originally posted by Geo:
I don't know the answer Dick. Wish I did.

careful george, that may be an admission that the current repair allowances are not enough

I can't help but think it would take us a big step closer to Production if we open up ECUs and wiring harnesses.[/B]

I am not advocating the opening up of ecu's if there was a way to enforce it I would go back to stock, but in reality I do not know how to solve this.

Speaking of Production and what to do with a car with NLA wiring harnesses... I don't know the rules for Production, but isn't that a possible channel for them?
[/B]

well yes it is but is that not the equivilant of go away.

George
thank you for participating in conversation like these. by doing so you add value and foster understanding of different sides of the issues.

merry christmas
dick patullo

 

[Geo]

Originally posted by dickita15:
well yes it is but is that not the equivilant of go away.

Not quite. It is for IT, but it doesn't render someone's investment worthless. There are alternatives. Sometimes we may not like the choices we are given, but if there is a choice it's not a dead end.

I'm not trying to say I don't understand or care. It's just that these things must be considered. If we allowed a lot of things that we discuss here all at once we'd have Production cars anyway.

In the end, sometimes you do have to say "if that's what you want, you need to go to production." That's never meant to be rude, but some lines must be drawn with respect to what is allowed. I don't have all the answers by any stretch. And opinions will vary. So with 9 of us on the ITAC we stand a fighting chance of getting things right or pretty close to it.

Originally posted by dickita15:
George
thank you for participating in conversation like these. by doing so you add value and foster understanding of different sides of the issues.

merry christmas
dick patullo

It's my pleasure. At least this has been a conversation that, while occasionally passionate, has remained thoughtful and civil. That is what this forum (and most forums) should be all about.

Merry Christmas!


------------------
George Roffe
Houston, TX
84 944 ITS car under construction
92 ITS Sentra SE-R occasionally borrowed
http://www.nissport.com

 

[Ron Earp]

Doesn't really seem to be a good answer here since George doesn't decide this issue by himself and neither do the rest of us. All we can do is interpret the rules and do the best we can. I suppose I'll race my car until successfully protested since in my unique situation the rules don't cover what I face. Moving any car out of IT into production due to a wiring harness, is, IMHO, senseless. A harness change offers no competitive advantage at all for cars that are as old as mine.

I think I'll check around for whom to write on the issue and state my case, it'll probably come back sufficent as written. Nonetheless, I'll learn something about how the process works. Sorry to break my promise and post again on the subject.

Merry Christmas to all you folks, take care,

Ron

------------------
Ron
http://www.gt40s.com
Lotus Turbo Esprit
Ford Lightning
RF GT40 Replica
Jensen-Healey ITS
My electrons don't care if they flow through OEM wires, do yours?

[This message has been edited by rlearp (edited December 24, 2004).]

 

[bldn10]

As far as I know, the change in the ECU rule was not benefit anyone or make our cars faster but was to deal w/ the fact that ECU mods could not be effectively prohibited. Now here we are debating what will logically lead to an open ECU rule. If stock now can indeed be enforced, that is where I would like to see us return; if not, is there not some middle ground that can be enforced? If chips were restricted to those available to anyone from 3rd party vendors (i.e. no in-house or one-off custom jobs) would that alleviate Steve's concern? Cannot ECU's be plugged into a diagnostic tool and determined to be stock or not? I don't know much about them (and prefer not to have to learn) and really don't know so please forgive my naivete on this subject.

BTW I am one of the lawyers on this forum and if I was told exactly what the intent was and what the pitfalls were, I could draft a rule.

------------------
Bill Denton
87/89 ITS RX-7
02 Audi TT225QC
95 Tahoe
Memphis

 

[Renaultfool]

While I am not for tons of rules creep, there have been major changes in automotive design over the past 35 years, the age range of the IT rules at this point. Most of the discussions on this forum are discussions of keeping it equal between the "old" carb/distributor guys and the newer "electronic/injection guys. Guys, it never will be equal and it never was. Find a 30 year old Volvo and drive it and then drive a 5 year old BMW. They are not in the same world performance and handling wise.

If equal is the goal, then why not rewrite the rules to allow logical modifications for ease of building, safety, and equalization of speeds.

As examples,let everyone run a vented disk on their brakes using the factory diameter and type of caliper. If you can beat someone because their older car overheats their unvented brakes and your vented ones still work did your beat them or their old car design? Let them run the computer and wiring that they choose to do, even the old cars. The Renault throttle body system would work on the older cars and I am sure that there are others. If you get beat by a wiring harness, shame on you. If you cannot police the computer or the older cars don't have them, free it up. Computers are here on all the new cars and no one will ever be able to police them all. Do you really think a computer and throttle body will make an old 8 valve engine equal to your computerized, port injected, 16 valve, dual overhead cam, engine anyway? Try and find a computer tester for my Renault to check that one out.
Get rid of the side marker lights and the glass headlights for safety. And the useless washer bottle for those who still cary them. And on and on. Make it simpler, easy to police, fairer, more competitive. They are race cars.

Every body gets vented disks, everybody gets computers, everybody gets whatever, as long as everybody can have it.

That does not get them even close to Production cars. The suspension mounting points would remain the same as now. The suspension arms would remain the same as now. We would not have alternate windows and body panels. We would not have hand grenade engines. Big differences. Major.

Yes your BMW might still beat my old Volvo, but not by as much, and you may have to work harder to beat me. More competitive cars would be cheaper to build for those to do so. Those who want to spend $50k on their IT car will still do so, you cannot change that. But at least they will beat me with their $50k, if they are equal drivers, not 30 years of technology alone.

 

[Knestis]

The question then becomes, "if you are going to allow all of those changes, why stop before alternate body panels and polycarbonate windows?" All of the same logic can be applied - safety, for one - so why not go all the way? Why the artificial limit at relocating suspension pick-ups?

Because that's where your particular construct of what IT should be bumps up against the characteristics of what you see as "a different category."

Where that conceptual line defining "IT-ness" happens is completely arbitrary and regardless of where it gets put, someone will always find attributes that they think should - or should not - be held by IT cars.

K

 

[Renaultfool]

Agreed.

But it may be time for some total revision of what IT is suppose to be with the goal of making it safer, cheaper, and more competitive for everyone.

Hey, it's Christmas, we can dream the impossible dream!
Carl

 

[ShelbyRacer]

Originally posted by Renaultfool:
But it may be time for some total revision of what IT is suppose to be with the goal of making it safer, cheaper, and more competitive for everyone.

Wow, I think I heard that somewhere before!

Nice to know I'm not the only one who sees that as The Right Thing To Do™.


------------------
Matt Green
"Ain't nothin' improved about Improved Touring..."

 

[gsbaker]

Originally posted by Geo:
Sorry Gregg, but what gizmo would that be? They don't exist...

Build it and they will come.

It's not rocket surgery. In fact, it sounds like a good semester project for a EE on the SAE formula team.

You heard it here first!

G

 

[ShelbyRacer]

Originally posted by gsbaker:

It's not rocket surgery. In fact, it sounds like a good semester project for a EE on the SAE formula team.

I like your thinking. The concept itself is VERY simple. Download the code from the chip, compare it to known program for that car using hex editor/comparison program.

Potential issues-

Connecting with the box. Before 1996, most companies had their own style connector for diagnostics. Some even changed pin location within the same connector. You could just use probes and skip the whole connector thing, but that can cause miscommunications, and if someone jumps the wrong pins, POOF, magic smoke released.

Interfacing with the box. There are more data stream rates than you can shake a stick at. Many computers (Dodge for one) simply won't even talk to something unless it speaks almost exactly their rate (which is proprietary of course).

Obtaining equipment. Every club (or series or whatever) would need its own laptop (granted, a REALLY cheap laptop would do the job) or other controller box.

Obtaining stock, unmolested .bin files for the factory code. This would be the most difficult thing due to lack of information on many cars. Also, a file for each year/model/engine/etc. would need to be kept.

The only way to be sure of doing it right would be the bin-compare method. I personally could go in and change other parameters to make something *look* like an unmolested map. I can program your ECU so that if you hold down the accelerator pedal while turning on the car (not starting, just turning the key to the ON position), it would have the program query other fuel and spark tables at previously unused locations, but if turned on regularly, it would seem perfectly stock. This would show on a bin-compare, but not on a regular scan-tool. The checksums would be correct, the PROM IDs would match, and no one would be the wiser. Imagine what can be done by my friends who actually WORK FOR a car manufacturer...

Ever wonder why those magazine test cars seem faster than the showroom model you bought? When I dropped the "factory tweaked" (read- magazine road test program) .bin file into my Firebird, I picked up .2 in the quarter with no other changes...

Again, it would show up in a bin-compare, but probably not anywhere else... Neat, huh?




------------------
Matt Green
"Ain't nothin' improved about Improved Touring..."

 

[gsbaker]

Originally posted by ShelbyRacer:
...Download the code from the chip, compare it to known program for that car using hex editor/comparison program....

Again, it would show up in a bin-compare, but probably not anywhere else... Neat, huh?

A "core dump" as today's grey-haired programmers used to call it.

At the opposite end of the spectrum, one could empirically map the chip functions. If they are not OEM, the chip is declared cooked. That gets messy.

G

 

[Tom Donnelly]

The binary files are in hex? I would have thought octal. Most "mini" systems use an octal based binary coding. Although it would readily convert to hex. Aren't the ECU systems similar to PC assembler? Or am I way off? I haven't worked with process-control systems for a number of years.

Tom

 

[ShelbyRacer]

Originally posted by Tom Donnelly:
The binary files are in hex? I would have thought octal. Most "mini" systems use an octal based binary coding. Although it would readily convert to hex. Aren't the ECU systems similar to PC assembler? Or am I way off? I haven't worked with process-control systems for a number of years.

Tom

Nope, hex code makes sense if you think about it. Two digit hex converts straight to 8-bit binary (one hex digit take the place of 4 binary digits). Raw binary files are assembled for whatever processor is used, which in the case of many of these systems is a Motorola HC11 or variant I believe. It's been a while since I messed with the hardware side of things, so that may not be accurate...

------------------
Matt Green
"Ain't nothin' improved about Improved Touring..."

 

[ShelbyRacer]

Originally posted by gsbaker:
A "core dump" as today's grey-haired programmers used to call it.

At the opposite end of the spectrum, one could empirically map the chip functions. If they are not OEM, the chip is declared cooked. That gets messy.

G

Only problem is that nothing is ever located in a "standard" spot. You can go by table maps and start locations, but again, you have nothing really to compare it to as to see if it is "stock"...



------------------
Matt Green
"Ain't nothin' improved about Improved Touring..."

 

[Geo]

And what do you propose to do about ECUs where the code has not yet been cracked?


------------------
George Roffe
Houston, TX
84 944 ITS car under construction
92 ITS Sentra SE-R occasionally borrowed
http://www.nissport.com

 

[Ron Earp]

I don't think this comapring loaded progams in ECUs will work. Not only are the previous posts dead on with baud rates etc. but from previous Ford experience working on Mustangs there are too many possible ECU programs for a just a given year. They would be no real way to have all possible programs available and on file for comparison.

I remember that 1995 Mustang GTs would knock quite easily with a timing increase unless you had a certain CPU program that would avoid it. During that model year it was one of three as I recall.

And, while tuning my Lightning this year I've learned that there were four different ECU programs the processor could have been loaded with during the 2004 model year. Model years 1999-2003, while fundamentally the same truck, also had differences during the year. Each one is slightly different, and while not appreciably changing much of anything, they would all appear slightly different when their hexadecimal codes were compared to a "master" on file. Unless you had four masters for 2004.

Maybe Ford is the oddball manufacturer in that they do this but I seriously doubt it. And, as mentioned while you might get programs for Mustangs and Lightnings easily because of all the forums/tuners/rodders I bet getting programs for something like a Olbsmobile Calais would be next to impossible (ECUs, at the core of this discussion I know). This would be a tough enforcement to pull off in this fashion.

------------------
Ron
http://www.gt40s.com
Lotus Turbo Esprit
Ford Lightning
RF GT40 Replica
Jensen-Healey ITS
My electrons don't care if they flow through OEM wires, do yours?

 

[ShelbyRacer]

Originally posted by Geo:
And what do you propose to do about ECUs where the code has not yet been cracked?

Kind of my point above, but the code doesn't have to be "cracked" for you to compare it. Hex is like a language. If I have two paragraphs written in, let's say, Korean, you don't have to know Korean to compare the two to see if they are EXACTLY the same. However, this would not take into account any minute differences, so it would have to be an EXACT match.

This was the whole point of my post. It's not impossible, in fac the premise itself is simple. Implementing it is another thing entirely...



------------------
Matt Green
"Ain't nothin' improved about Improved Touring..."

 

[Tom Donnelly]

Here's an example of a hex dump:

000046448 0001 0000 2004/11 020722602
4FFFFFFFFF4FFFF4FFFF4FFFF6FF4FFFFFFFFF
00000464480000100000020041110020722602

This is from an IBM mainframe (Jurassic code)
dump just for illustration purposes.

A 40 is a space, an F0 is a 0 etc.
You don't have to know what the representation is to compare the data.

However, a program dump can compare as being
different just based upon assembly date or
compile date, as those dates are usually
loaded into the executable.

What is being proposed here is sort of
reverse-engineering or de-assembly of
the program in the ECU. You can do it,
its just difficult. And you have to have
an idea of the language its written in. I
doubt its COBOL. (Thats a programming joke)

Tom

 

[ShelbyRacer]

Originally posted by Tom Donnelly:
And you have to have
an idea of the language its written in. I
doubt its COBOL. (Thats a programming joke)

Tom

Maybe that's why our cars didn't all crash on 01/01/00?

------------------
Matt Green
"Ain't nothin' improved about Improved Touring..."

 

[Joe Harlan]

The point is that all tuning should be done to the stock ECU....( much like tuning a stock carb)because it is not easy to police. As far as an ecu that has not been cracked I would suggest that's part of the development of any car that is an IT candidate. I would also suggest that people need to look outside the normal sources for cracked ECU's. If there is a desire to get it done it can be done for any car out there.