IHADAV8.com - Turbo Buick Tech, and Nonsense
Tech Area => General Buick Tech => Topic started by: daveismissing on September 05 2012, 12:36:03 PM
-
I'm sure Bruce Plecan, Rob Rausch, Kent Chu and Bob Bailey have the architecture down pat but
has any documentation for the internal bus ever been released to the wild?
-
The only internal bus for the ecm is for memory address (16 bits) and data (8 bits) lines. I did a lot of research in the late 80s and most of the 90s on the processor and support chips. During the same period I studied, no hacked all the chips I could get my hands on for the purpose of understanding the programming. Eventually, I figured out all the prom data and code. Later, I tackled the rom program.
I was more interested in the programming than the hardware so I didn't document much on that. I still have my old notes though so I might be able to help you. I did lots of documentation on the software, some of which Eric (TurboTweak) has on his website. Over time, these docs were updated but not on the web site because nobody cares anymore about what goes on inside the ecm.
The processor is a Motorola 68HC11. The Motorola user manual for it is 4 inches thick! The processor chip started as a M6800 then was customized by GM. The other chips used were also made for GM and don't have normal markings/numbers.
Let me know what you're looking for, I probably have it or can figure it out. I've attached a memory map as a sample of what I have.
-
You mentioned Kent Chu, well back in the day, he and I swapped emails when I was trying to get him to make improvements and fix minor bugs in Direct Scan. When it became obvious that he had no desire to make changes, I offered to make the changes to his code and he could sell it as a new version. He wasn't interested. Oh well.
I taught Eric Marshal assembly language programming before he started TurboTweak. I consider him a friend although he never did "pass the final exam" in my class. Don't get me wrong, all the engineering ideas were his and he makes great chips. :rock: Its only his programming techniques that I tease him about. :icon_lol:
Dennis
-
You mentioned Kent Chu, well back in the day, he and I swapped emails when I was trying to get him to make improvements and fix minor bugs in Direct Scan. When it became obvious that he had no desire to make changes, I offered to make the changes to his code and he could sell it as a new version. He wasn't interested. Oh well.
I taught Eric Marshal assembly language programming before he started TurboTweak. I consider him a friend although he never did "pass the final exam" in my class. Don't get me wrong, all the engineering ideas were his and he makes great chips. :rock: Its only his programming techniques that I tease him about. :icon_lol:
Dennis
I was getting that impression from a few of your posts.
Keep it up and hats off to you.
AJ___
-
It is sad that many guys put time and shared effort into research and somehow it ends up as proprietary product yet built on "open source" efforts. The Lockers board was out there and disappeared. B&G seem to be the only makers of megasquirt hardware.
I was interested in the bus pinouts. If Kent wasn't going to do anything further with his product he could have open sourced it.
And yes- we appreciate having your knowledge here.
:)
-
I was interested in the bus pinouts.
[/size]Are you referring to the ecm edge connector that PL (and DS) adapters plug onto? If so, let me see what I do. It might take a few days.
-
I was interested in the bus pinouts.
Are you referring to the ecm edge connector that PL (and DS) adapters plug onto? If so, let me see what I do. It might take a few days.
Yes, I would have expected that to be available by now. Not a rush, it will be something to engage my brain over the winter.
Thanks
(Tho its already getting dark way too early :( )
-
Adding to the list there was another dude who modified WinALDL to be Buick specific and that project was
abandoned when he got rid of the car. Another source lost. He was on mp3car.com as well.
-
I have most of the ecm bus figured out but I'm having trouble locating some of the pin definitions for the processor. From the little information I have, the processor appears to be a 6805. I know the instruction set matches the MC6805 (little brother to the MC68HC11) but I don't have a pin diagram for a basic 6805. I'll keep digging.
On another note, I taught myself Visual Basic a few years ago and in the process wrote a WINALDL replacement program. After that I created a replacement/improvement to a program called TurboMeter. Fun projects but since they were written in VB, the executables need VB support files to run. Became too difficult to manage with all the Windows variations so I never released the programs. :icon_eyes:
-
Is the pinout the same as the 6800?
http://en.wikipedia.org/wiki/Motorola_6800 (http://en.wikipedia.org/wiki/Motorola_6800)
-
I have most of the ecm bus figured out but I'm having trouble locating some of the pin definitions for the processor. From the little information I have, the processor appears to be a 6805. I know the instruction set matches the MC6805 (little brother to the MC68HC11) but I don't have a pin diagram for a basic 6805. I'll keep digging.
I have a guy.....
On another note, I taught myself Visual Basic a few years ago and in the process wrote a WINALDL replacement program. After that I created a replacement/improvement to a program called TurboMeter. Fun projects but since they were written in VB, the executables need VB support files to run. Became too difficult to manage with all the Windows variations so I never released the programs. :icon_eyes:
Turbometer- that was it!
I kinda stalled around QB45... :O
Can't compile VB into a standalone .exe or that requires the big bucks package?
-
Not a rush, it will be something to engage my brain over the winter.
Since we won't have hockey to distract us...
-
Is the pinout the same as the 6800?
http://en.wikipedia.org/wiki/Motorola_6800 (http://en.wikipedia.org/wiki/Motorola_6800)
I suspect its close but not exactly the same. Not really sure. (?)
Can't compile VB into a standalone .exe or that requires the big bucks package?
[/size]That's right. I happen to have the complete package for VB6 so that's what I used. I don't want to spend the money for a current Microsoft Developers package.
[/size]
[/size]Both of my programs run if the appropriate support files are there. And they're available, hell I'd even give out the source code if it would help someone, but who uses the 160 baud serial data these days?
[/size]What's that you say ... Steve probably does because he's "computer challenged".
[/size](Steve, I respect you but the other guys made me say that.)
[/size]Dennis
-
Great Thread, Maybe someone here can explain my loss's I've had 3 different motor failures all due to trying to get a Stock type ECM to perform. I am now also as of 9-14-12 banned from TurboBuick.com for 30days because of the same issue above i tried to express.LOL What i was told by a person (Heavly into Engine Management) in the 7's in the 1/4mile with Buick an company's still current an running, That there's a Factory Code in the all Stock ECM's that unchangeable an is design to make the motor go Lean on the Top End. It's a Government/insurance/GM code My 3 builds all had factory ECM's then a QuadAir mod. factory type, both Same effect. Eric burnt the chips, all different an specially designed for each the builds with translator plus it was professionally dynoed. (I'll save the place of the dynoing but there a "Buick Specialist" in Orlando,FL, easylie Googled) Dynoed at 610hp at the motor 533rwhp, ( i have 9" an full, larger exhuast) Real World an Muiltipul Dyno Results where the car at just over 100mph fell over an leaned out.No Fuel issues!,Reds pump hotwired an voltboosted,inj. only at 50%,duty,it's on 93/Alky,22-23psi boost,No trans issue, All brand new billet job with artcarr3500,3:50 gear currie 9", an 28x10 et street tire. It act's like the old speed/fuel pump cut off,But!!! It Lay's it over slower, It doesn't cut the pump an nose dive the car,It's a Fast gradual cutoff/lean out some how. This is a video of the first motor loss, headgasket,1/8 mile mark,Ran 11.4 with 1.80 short,The Supra left on .006 light,When i hooked I smoked that Supra so hard he trapped 133mph to catch me! i hit the 100mph ECM fail like i felt on the street. This happen 2 more times,video's on my youtube channel,(watch me hang with mid 9 second prostreet mustang for 2 seconds till the 1/8...heh) Buick GN vs Turbo Supra (http://www.youtube.com/watch?v=5KGQs8Pt68I#) I have a scantool log i tried to post here of first motor,It was to large the first time,If anyone would like it i can Email it,each file is 8mb,,I'll try an shirk it down later, It's 3 views of the same pass of a street 1/4 mile run,everything i thought was fine Then Zap, Motor's tunes Done...... Your going like bat-outa hell, then all of a sudden you can kick back an drink a cup of coffee..... After a lawsuit ($30,000) i won from the "Specialist" for wasting 3 years me an my car's life i went to an XFI unit an the ECM fail Problem above Vanished!!!!! 2010 Buick Grandnational Dyno Runs!!! (http://www.youtube.com/watch?v=baEVD-s_uWM#) I wish i had a dyno sheet, but my luck, Titain's computer was down that day,I didn't care, I had Cal Hartline tune it anyway, just 17psi boost. Notice, no over 100mph Cut off!! we Even made a mistake of leaving the parking brake on in the Car,on the last run, No Motor Damage!!! Thanks, Bozo
-
50% of a 50lb inj is 25lb hr x6lnj is 150 lb/hr which will only support 300hp
your combo trapping what it did was closer to 500+hp
50% on your build that you listed
how is this an ecm issue ..its a tuning issue
for 500+ when i was on 50s and alky it took 86% inj duyt at 25psi to go 10.90s (99 1/8 121 1/4) on a moderate motor build.. pte FMIC , stock shortblock , flat tappet 206 cam, home ported irons, stock intake , 3" maf pipe and lt1 3" /translator , stock headers ,3" thdp, tt street 93 chip oh and with a ta49
to give you an idea of 50% of 50s
i ran 50s at 55% when i was pushing the stock turbo to 12.0s with a bone stock motor 22psi 110mph 1/4 , 89 1/8th
-
I think i have the scantool log's small enough now for load..(Nope,LOL,ok try number 3, the charm?) the joy's of conversions... lol
I'm no expert Tuner,Auto Detailer ,Yes, I set out to pay for a tune job done,an then Learn it All from there out.I Never got the chance! I was just geting the lap top running with the car an scanned this log,the combination to a T i can't recall exactly,but it dyno 610hp an 533rwhp If my Tuner screwed me so be it, i got my cash back, I just want to know the Exact Reason it'd go from Rocketship to Snail...3 different tunes,combo's,inj,chips, but all factory type ECM. My fresh build's virgin fire up's in week or two!!
My issues are solved now with XFI unit. But other's could hit the wall i did an benefit before hand,I'm all about the Car,not the people owning them...they die off but the Car Lives!
Enjoy!
-
I'd like to meet the attorney that got you $30,000
-
I'd like to meet the attorney that got you $30,000
She cute with large Mammery's,heh,Thats what i settled for....of coruse owning a classic car helps..8^)
-
All I can say is that there is nothing programmed into the stock ecm that would make it lay down on the top end. If the stock ecm was the cause of your problem it would be because it has an 8-bit processor with a tiny amount of memory and is slow by today's standard. There is no precision on the math functions. The processor doesn't understand fractions so it rounds down and this becomes more critical as the rpm goes up. If you compared this ecm to a PC it would be the equivalent of a Radio Shack TRS-80. (You younger folks can Google it if necessary.)
I'm not sure what happened with your car. Perhaps you did discover the limits of an engine management system that absolutely was not designed for a 500 - 600 hp engine.
-
Hopefully Bruce will visit you tonight and give you an update on the 8 bit ecm as he demonstrated several years ago that the factory ecm could keep up with a lot more than 500-600 hp :)
I believe a couple of guys in the past went into the 8's on the stock ecm without any problems.
Currently Walt Judy,running in TSM very successfully, is running a stock ecm with an ME chip. Steve Y is using fuel curves checked by Bruce to provide the fueling. Not only has Walt been quite successful, he is running 9.3-9.5 in the 140-150 mph range which is a lot more hp than 500-600.
As Paul stated in one of these threads, the modern systems make it very easy to tweak parameters run to run, or after component changes without going thru the hassle of burning new chips as well as offer many minor tweaks that may offer more consistency during the day as ambient conditions change when used by a savvy tuner.
Way too many of us are making plenty of hp to blame failure on the stock ecm. I suspect one could statistically demonstrate that there are more failures incurred on aftermarket fuel management systems than by the stock ecm, but, that would be less than the whole story, of course :D
-
I've read some fast car's on stock ECM,yes,most in the 9's now are on Aftermarket system though if read Sig's. I'm sure a computer engineer could do what ever they like to factroy Ecm. Me i don't have those resources or room to work on car. So i "thought" i did the next best thing, Find the closet to me professional Buick tuner. (KDK automotive,you can look him up,or might know him already) get chip from a respective maker, (Eric's which KDK use's anyway). I had wanted an DFI system or Bigstuff3 before any this, But KDK said "You can use the the stock ECM with Translator plus', So i listened an went that route.... When i picked up my car the first time, Was a RocketShip,Dynoed 533rwhp all numbers matching motor... (Dyno sheet attached below) Pull of 10 second car easy. But 1/8 mile it was done. I told KDK the problem an him an Eric an Cotton Perf. (which was having the same issue on there car,he even sent the tranny out his car an it was put in mine to test,Same Issue though) I tried 2 other full builds to try have the Pro's fix the issue. They couldn't.....(or Wouldn't ) But don't think people want bad name.......onl y read a duty cycle issue i guess so far,does the chip or the ECM control that an how's it adjusted (short version's fine) the Scantool run post above still show's 530hp though. Is what happened was the tuners going for HP on dyno but in Real World driving with wind an rolling resitance the car needed more then what a dyno run could tune??? Even them testing the car didn't like it over 100mph. Here's the dyno sheet so you can the rest the info...I going to learn my car i just need base line to go from first,I learned Astrophotraphy in 3 years (while my car was in the shop), Tuning be easy if i could get a starting base that'd atleast run one 1/4 mile...8^)
-
As I stated, your experience is different from what the vast majority of us have experienced over the past 25 years. There are certain vendors that tend to tell people the stock ecm is no good, often on 12-14 second cars, because they like to sell aftermarket systems at a good profit and then tie up the tuning for years to come at a further good profit when compared to selling a programmable chip for a few bucks.
There is absolutely nothing in the stock ecm that shuts the fuel off at 100 mph or 600 hp. You can tell yourself anything you wish, but reality is different.
As Paul stated earlier, it is much easier to tweak an aftermarket system to adjust for ambient conditions and that may be more important on a Nine second car than on a run of the mill 600 hp car if the user understands tuner. If not, you can always show up at the tract with a hired gun to do it for you.
-
those dyno sheets show me typical dyno mentality
get the highest hp number and to do that they run it lean
you had 23.4psi with 12.1:1 AF !!
news flash no fng way that tune would hold up at the track in a 3600+lb buick
i dont dyno because i could give a crap less about the number
work the tune up on the 1320, and that doesnt mean run 1/4 first tme out it means dial it down and work it up while looking at your data and watching everything you can while your doing it and lift at any sign of trouble
and typical with tt chips he brings the fuel up over 5000 as you can see you didnt go lean, your AF dropped to 11.0 ..thats richer but on 93 alky real world 1320 i would be targeting even richer than that
-
As I stated, your experience is different from what the vast majority of us have experienced over the past 25 years. There are certain vendors that tend to tell people the stock ecm is no good, often on 12-14 second cars, because they like to sell aftermarket systems at a good profit and then tie up the tuning for years to come at a further good profit when compared to selling a programmable chip for a few bucks.
There is absolutely nothing in the stock ecm that shuts the fuel off at 100 mph or 600 hp. You can tell yourself anything you wish, but reality is different.
As Paul stated earlier, it is much easier to tweak an aftermarket system to adjust for ambient conditions and that may be more important on a Nine second car than on a run of the mill 600 hp car if the user understands tuner. If not, you can always show up at the tract with a hired gun to do it for you.
That's where i'm at now,I see Cal Hartline this Tues. So... my last "hired gun's" ,KDK an Eric failed there task? KDK is die hard stock ECM user. (plus he said him an one his tune guy's said they couldn't figure out XFI programing an was to difficult) So... i went there route.the stock ECM way, total opposite of what you stated, got me 3 wasted motors,4 if count the one that went with me an Cal on another dyno an shop. I have hooked up to the XFI myself, i've read the table chart's,I see exactly how it works so from now on, I won't be having issues, an haven't since the XFI install. I just should went the aftermarket way at the start,no conflicts between tuner,chip maker an factory equipment issues
-
There are no factory issues....that is a created reality to explain the problems....Yo u may be the unlucky type that could possibly draw three bad ecms in a row, but, no one else will ever believe that.
I do hope you have better luck this time :)