Need a stock, or near stock 3-bar compatible .BIN file....

Darth Fiero

New member
Need a stock, or near stock 3-bar compatible .BIN file for my GN Turbo speed density conversion project. -OR- Please explain what is all involved in the VE tables when it comes to converting the 2-bar code over to be compatible with a 3-bar MAP sensor. Any help is appreciated. Email files to my email address on the website in my sig please. Thanks.
 

JSM

Active member
stock 3bar doesn't exist. What computer are you using?

With any, it is WAY more than adjusting a few VE tables. It requires you become a full fledged computer geek, and understand computer code, which isn't that easy since GM doesn't publish anything to help you.
 

Darth Fiero

New member
I know the Sy/Ty's don't come with a 3-bar sensor stock. What I need is a 3-bar compatible bin to start on my GN Turbo speed density conversion with. We are running 20 lbs of boost and the stock 2-bar will only read up to 15lbs.

I am using tuner cat and datamaster and have done may chips myself. Just need a little advice on what needs to be changed to get a near stock sy/ty bin to accept a 3-bar sensor so I can start out from the beginnning. Thanks for the info in advance.
 

PPI Typhoon

DIY Madman
Well, a lot of things in the definition files out there are a bit hard to find out how the 3-bar conversion happens. You can't take a 2-bar bin and convert it because I know it involves the actual table constants later on in the code, which is not defined. Now, if you have the binary files of each, you can actually just do a bit compare, and find out WHERE in the code this happens. For something like the 749 ECM, 2 to 3 bar conversions use all the same tables except for F1, F3, and F77. The actual table is different, not only the variables within. So if you go into something like Promgrammer, you can change everything, and compare the file within there to where the "compare" function says nothing is different. You do a bit compare with a hex editor, and you'll see that the files are still not the same. But later in the code (still using 749 as the example), you'll see that the MAP tables are changed to go up to 285 kPa where the stock only goes up to around 190 kPa. This is done around........0x3572......

The bit goes 1000 (8 in hex)

to 1100 (C in hex)


Here is an example (there are a few of tables that change) of what a stock 2 bar table looks like in Promgrammer for one of the spark advance tables:


Deg. vs MAP & RPM
| 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190
---------------------------------------------------------------------------------------------------------------




Now the math I crunch comes out to this.......

190 kPa x .14514 = 27.4 psi

Of course, with 1 atmosphere of vacuum and one of boost, that comes out to roughly 28 psi. Makes sense.

But if you increase that MAP table to extend up to 285 kPa, which is done at around 0x3572......you can change the values of the table itself. So when you change the bits as above, it comes out as:




Deg. vs MAP & RPM
| 45 60 75 90 105 120 135 150 165 180 196 210 225 240 255 270 285
---------------------------------------------------------------------------------------------------------------



285 kPa x .14514 = 41.4 psi

You'll see that it adds roughly one more bar (atmosphere) to the readability of the tables.

Now this isn't the ONLY thing that changes. There's something else at 0x3576 that changes, but I haven't figured out what that does yet. Perhaps someone can fill in this gap.

The bit goes 1001 (9 in hex)

to 0101 (5 in hex)

I'm not sure what this translates into yet.

But that bit change registers as a 3-bar chip in Promgrammer, and other than what you can compare in Promgrammer, those are the only "hidden" values not able to be altered through Tunercat or Programmer.

But remember, this isn't the ONLY thing that changes......not even close. But it is the only two things that I can see change that aren't able to be modified by Tunercat and Promgrammer.


Hope this helps explain a little.
 

Darth Fiero

New member
I see what you are saying. Not only will I loose resolution in the VE and Spark tables, but other things, such as TCC load limits, Accel Enrich tables, and DFCO values will also need to be changed because the ECM doesn't know you have put in a 3-bar sensor; we are just recalibrating the 2-bar tables to be compatible with a 3-bar sensor which will operate in the same voltage ranges as the stock 2-bar.

Sounds like a lot of work, but I believe mathmatical equations can be applied to the affected values to make them compatible with the 3-bar.

Thanks for the info. And if anyone has a 3-bar compatible .BIN file out there for the 7749 ECM and can email it to me so I know where I need to start looking, I would really appreciate it. Thanks.
 

smeagol

Active member
Well what is done is kind of hard to explain unless you have disassembled the chip a bit.

The code has values for manifold pressure, vacuum, boost, barometric etc. They are all separate. It just doesn't use one value for manifold pressure. Also has raw input value for the sensor as well. So what was done was converting that raw input to vacuum, so that all the NA type of calculations using vacuum readings were accurate with a 3 bar, then rescale all the tables/constants that used absolute manifold pressure or boost values. The code already has provisions for 1 and 2 bar, so I just followed the factory modification to allow for a 2 bar, and made it fit the 3 bar.

The code change was simple, but it took me lots of time & work to figure it out since I wasn't familiar with code (and I still suck at it). The calibration changes were numerous, but were pretty much plug n play math work.
 

Darth Fiero

New member
I see said the blind man. Looks like I have a lot of work ahead of me. Think since I am starting out with a MAF to SD conversion anyway, I should just start out using a 2-bar until I know for sure all of this stuff will work on the Buick before I waste a lot of my time. Thanks.
 

PPI Typhoon

DIY Madman
Yeah, if anyone knows about this stuff, it's Brian. Any words that he can offer, go with.

I'm just in the beginning stages of figuring this stuff out. A lot of it leaves me like :eek: after the day. But it's fun figuring it out.
 

BillC

New member
PPI Typhoon said:
Now this isn't the ONLY thing that changes. There's something else at 0x3576 that changes, but I haven't figured out what that does yet. Perhaps someone can fill in this gap.

The bit goes 1001 (9 in hex)

to 0101 (5 in hex)

I'm not sure what this translates into yet.

But that bit change registers as a 3-bar chip in Promgrammer,
Be careful here. There are a lot of other code changes for 3-bar besides that one byte. Promgrammer simply looks for the presence of the code mod for displaying 2-bar vs 3-bar, and you just happened to find the one byte it checks.

Changing that byte will NOT convert your bin from 2-bar to 3-bar, even though it will show up in Promgrammer as 3-bar. However, changing just that byte will (most likely) give you unsolvable tuning problems. :eek: :x
 

PPI Typhoon

DIY Madman
BillC said:
PPI Typhoon said:
Now this isn't the ONLY thing that changes. There's something else at 0x3576 that changes, but I haven't figured out what that does yet. Perhaps someone can fill in this gap.

The bit goes 1001 (9 in hex)

to 0101 (5 in hex)

I'm not sure what this translates into yet.

But that bit change registers as a 3-bar chip in Promgrammer,
Be careful here. There are a lot of other code changes for 3-bar besides that one byte. Promgrammer simply looks for the presence of the code mod for displaying 2-bar vs 3-bar, and you just happened to find the one byte it checks.

Changing that byte will NOT convert your bin from 2-bar to 3-bar, even though it will show up in Promgrammer as 3-bar. However, changing just that byte will (most likely) give you unsolvable tuning problems. :eek: :x

I wasn't saying that it converts it to 3 bar, but that bit change at 3572 changes the table values which helps when dispersing the timing values (and others) to adapt to the 3-bar map sensor. Promgrammer recognizes this bit change and reflects on the tables. As I stated later in the post that this is "not even close" to all that you have to change. But if you look at a 3-bar chip vs the 2 bar stock, those are the only bit changes that you can't modify inside of programs like Promgrammer or Tunercat. Everything else can be. Because if you change a stock 2-bar bin to to same settings as a 3 bar......it does seem to work (at least so far from what I'm testing). But if you do a bit compare, you notice that there are some differences. Then I noticed that there are some table changes later on in the code. The first does a table change......the second I don't know of yet. You're certainly the man behind Promgrammer, so do you know what goes on after that?? I like the fact that you made your program check that byte. It helps when you have to readjust your tables.

So yeah.....in short, you're right, and I already said that this isn't the only thing that changes..........just the only "hidden" thing between the 3-bar chips that are out there (people working with the 749 ECM on various cars) and the stock 2 bar. The rest of the conversion can take place inside an editor like Promgrammer, Tunercat, GMEPRO, etc.

The point is that for a while now, people ask questions like the original post and no one really answers them. There's talk about you have to change the assembly code......which if I'm not mistaken, would change the overall HEX/Binary of the file. So there are two things in the code that change........one is the table values...........the other I have yet to find out.

I'm still trying to find out how to disassemble the code. But I have yet to find the disassembler that works. Anyone have any links??

Brian explains the other stufff pretty good on his pages. Some of the deeper code changes just aren't explained. :)
 

Darth Fiero

New member
Have you guys tried posting on either the topica tunercat lists or the diy-efi.org mailing lists? There might be some people on there that know where to find a good dissassembly of the $58 code.
 

Darth Fiero

New member
PPI Typhoon said:
Actually, someone pointed that there was a AYBN hack on DIY-EFI, and I'm looking at it now.

Hey, if you could either forward me the hack or point me to the appropriate URL, I would appreciate a look at it as well.
 

turbodig

Active member
PPI Typhoon said:
ftp://ftp.diy-efi.org./incoming/$58t.zip

Not sure if it's accurate or not, but it's a start

Looks accurate, although it's not very complete.

I've thought about this for a while... seems to me the best way to
approach it would be to snarf the DIS code from a TGP. There is a lot
of code similarity in the TGP code and the Sunbird code (heck, in lots
of GM ECMs for that matter), and I think it would be possible to graft
that code into the $58 by replacing the old spark routine with the TGPs.

The '730/727 and the '749 are very similar beasts... using the whole
TGP code might be a better way to get to where you want to go.

The big problems would occur in the areas of spark fault codes, and
where subsystems crossed (like boost reduction based on timing retard).

It's not really a trivial project, but it would be a cool thing when you had
it done.

Later,
Dig
11.94@112 | '749 with custom code
 
Top