Who wants a custom ECM calibration?

[lookin]

Maybe some of the various providers, like MS and TU and FWDP, could all team up also. They could work together, to reach a legally binding profit sharing schedule, apportioned by the outlay of time and money each company contributes etc; Or each company could agree to contribute different parts of the unit, a la Skunkworks in a DOD style blind prototype race in which the DOD then choses from among the various offerings, the best components from each provider, to then be assembled into a single device.

Now it just may be also, that one or all of them, are already secretly working on the unit even now. Such may be publicly stating, "they wont touch it" and "there is no money in it" and or "the market is too small" etc, when all the while they are really working to be the first to get one on the market.

If any of these various companies, can really trust that the others are not already doing so, then they can trust each other enough to team up and get this baby spanked too.

[irocelectric93]

Honestly the best way to do it would be money up front. Id be down to do that if i knew that their would be some quality CALS made. Im talking a couple set ups. Some for Boosted setups and some for N/A.

[lookin]

Same here. I would sell my flux capacitor in a heartbeat to raise the dough.

[Shel-Game]

Quote:

Originally Posted by lookin This has thus far been discussion which contains, yes, gasp , differences of opinion. It did not "become a huge mess". While I disagree with your opinion of what Shel-Game "was just trying to offer", ....

Well, OK. What did you think I was trying to offer? While I do cals for people on the side, I certainly don't do it to make a living. Everything I do, I post for free so that anyone who wants to do thier own cal can. Your assertion that the custom cal arena was dead is wrong. There has been a lot of progress on the V6 dis-assemblies. I was trying to inform you of that, and you went sarcastic on me. You stated that the memory locations need to be figured out, and I was trying to point out that it's not as big a task as you originally assumed. I'm 100% positive that I could dis-assemble the '91 code in 2 weeks, given the time to do it. And I could probably get the '92+ done in another month. But, then, as I also pointed out, it's already been done. You just need to pursuade Derek to share it. I really just don't understand the reason for the sarcasm to begin with. As for a cal, if someone would socket thier V6 SMEC, I could send you a modified chip in a week. With a couple of iterations, I think we could build a pretty good cal. Certainly not as good as TU or FWD could do on the dyno in a day, but you can only do so much thru the mail.

[lookin]

Quote:

Originally Posted by Shel-Game Well, OK. What did you think I was trying to offer? While I do cals for people on the side, I certainly don't do it to make a living. Everything I do, I post for free so that anyone who wants to do thier own cal can. Your assertion that the custom cal arena was dead is wrong. There has been a lot of progress on the V6 dis-assemblies. I was trying to inform you of that, and you went sarcastic on me. You stated that the memory locations need to be figured out, and I was trying to point out that it's not as big a task as you originally assumed. I'm 100% positive that I could dis-assemble the '91 code in 2 weeks, given the time to do it. And I could probably get the '92+ done in another month. But, then, as I also pointed out, it's already been done. You just need to pursuade Derek to share it. I really just don't understand the reason for the sarcasm to begin with. As for a cal, if someone would socket thier V6 SMEC, I could send you a modified chip in a week. With a couple of iterations, I think we could build a pretty good cal. Certainly not as good as TU or FWD could do on the dyno in a day, but you can only do so much thru the mail.

I have gone back and looked at the thread. You are right. I was wrong, and I apologize.

The sarcasm happened because I got defensive when people got all negative about it. I just could not understand what all the doom and gloom was about , because it is so doable. So I began to think, as did Dyan also by the way, that the nay saying was purposeful being based upon economics. I am very sorry for the misunderstanding.

We want to help get this done. Are you willing to work with us? It sounds like you are, and I sure hope I read you right this time.

Dyan has the tech skill to do a helluva job on socketing the board and more. We will need some help to figure out how to socket a 3.0 SBEC-II though. She doesn't know our engine at all, and I know relatively little. I just know what things are called, where they are and basically what they do. I know where the sensors are, and how to change them out or r&r them, and that is about it electronic wise. I do have the 1994 AS Caravan T&C Voyager Electronic Service Manual though , and I have basic tools, like sockets torque wrenches and an 02 sensor socket. woop-te-do

[lookin]

I know nothing about electronics, that I haven't learned only recently, but I just got an idea. Maybe it would work.

An exogenous socketed board, run switchable in parallel, with the OE board on a running engine on the dyno. You have the OE code on a PC, which you plug into the exogenous socketed board, and you set up these two boards so you can switch back and forth between them while the engine is running on the dyno.

You run the engine on the dyno, switching back and forth between the two boards, as you try different code configurations on the PC connected to the exogenous socketed board. When you finally get the desired results, you save the program code on the PC, and then later transfer that onto the chip to be plugged into the socketed board.

I think that is one way you could do it.

[90TurboVan]

In these computers, data about how to run the engine is stored in tables. For example, you'd have a table of injector pulsewidth (output) versus MAP reading (input). It gets more complex than that with other tables coming in to play during warmup or acceleration, but that's the basic operation. All the microcontroller does is look up what it's supposed to be doing based on the sensor inputs and controls its output. When you pull a cal off a chip and open it with D-Cal, you can see all the different tables, the problem is that you don't know what all the tables mean and what their input and output units are. For the old cars (SMEC), most of the tables have been figured out and you can get files from the Yahoo D-Cal group that have pretty much everything marked, which makes changing a cal really easy. I'd suggest at least joining the D-Cal group (you need a yahoo id) and getting D-Cal and some calibration files to play with. Source (assembly/commented) files are also on there. With the newer (90+) computers, not much has been done that's as publicly available as the old stuff is.

Running it on the dyno with two different computers wouldn't work out that well because it'd be hard to figure out tables that only have control during warmup, or during transient conditions. If I assume correctly, you have to look at the disassembly and see where the different tables are "looked up" in order to find out exactly what they do.

Haven't opened a V6 SBEC-II, but I'm sure they're just like all the other ones. Since they're enclosed, you should be able to see through the potting after the top cover is off. There should be ~3 square chips with pins on all sides. The 52 pin ones are the MCU (Motorola 68HC11, should have the M Motorola symbol on it) and a peripheral expansion chip. There should also be one large 28 or 32 pin chip with pins on two sides. This is the EPROM that holds the calibration, and is the chip you want to socket. Best way to learn about it is to get a computer that you can sacrifice and remove all the potting and the large connector, then you can see where all the pins go and figure out what the chips do. Just be careful with that potting compound, its nasty stuff and the fumes during soldering aren't fun.

[lookin]

Thank you 90TurboVan. That is very helpful information.

Dyan said she could get a development kit for the Motorola 68HC11, and it would take her four months, maybe less, to learn how to program it. Then she would need the SBEC-II 3.0 assembly/commented calibration files, to begin the development work to reprogram it, and she is not willing to do that part because she lacks the background to make it time feasible.

So is there any chance Derek will share the codes with the rest of us? Will he sell them to us if we all, pardon the pun, chip in? I suppose he has been asked before about this. Can anyone tell me what his response has been?

If Derek wont do it, then we gotta find a way, to get Shel-Game to decipher the 92+ code. Certainly much substantial appreciation would need to shown by the entire 3.0 community, in advance even, so he would want to do it. Though I have already apologized to Shel-Game, I would also be willing to publicly grovel and literally kiss his, I am certain stunningly gorgeous butt, before all creation on live TV. Now, seriously, .... I really will if necessary.

[lookin]

Quote:

Originally Posted by Chris W Still waiting for a 3.0 test vehicle in MI. Chris-TU

I may be able to drive my van out there for this purpose.

I would be willing to live in my van, since I have super insulated it, for the duration of time you would need it. I can't afford a motel these days.

So how long would you need my van Chris?

PS

Will I need a machine gun to be safe while staying in the van out in the parking lot? I hear Detroit is a scary place.

[Shel-Game]

Quote:

Originally Posted by lookin I know nothing about electronics, that I haven't learned only recently, but I just got an idea. Maybe it would work. An exogenous socketed board, run switchable in parallel, with the OE board on a running engine on the dyno. You have the OE code on a PC, which you plug into the exogenous socketed board, and you set up these two boards so you can switch back and forth between them while the engine is running on the dyno. You run the engine on the dyno, switching back and forth between the two boards, as you try different code configurations on the PC connected to the exogenous socketed board. When you finally get the desired results, you save the program code on the PC, and then later transfer that onto the chip to be plugged into the socketed board. I think that is one way you could do it.

Yeah, you can definitely do it that way. There is a comercial product called the "Romulator" that will do that - for sure it will work on a SMEC. But, the chip the SBEC uses has a built-in address latch. I'm not sure if it will work or not. I think someone was sucessful in using a Romulator on a Turbo SBEC, so it may be possible. But, I don't remember who it was. There's also the "Ostrich" available on Moates.net that works similarly (but is much cheaper) and it may work as well.

[Shel-Game]

Quote:

Originally Posted by lookin If Derek wont do it, then we gotta find a way, to get Shel-Game to decipher the 92+ code. Certainly much substantial appreciation would need to shown by the entire 3.0 community, in advance even, so he would want to do it. Though I have already apologized to Shel-Game, I would also be willing to publicly grovel and literally kiss his, I am certain stunningly gorgeous butt, before all creation on live TV. Now, seriously, .... I really will if necessary.

Hahaha!!! Yeah, think Shrek and you'll get a picture of my butt. Thanks for the offer, but I don't think that'll be necessary. And, thanks for the apology, BTW. I understand your frustration, it was the same with the Turbo guys before we really got people to sharing info. Now, there are a lot of guys doing thier own cals and contributing "to the greater good" so to speak.

If Derek isn't willing to release his work on the 3.0 stuff, then I'll take a look at it. I'd hate to duplicate that effort, though.

[lookin]

Quote:

Originally Posted by Shel-Game Hahaha!!! Yeah, think Shrek and you'll get a picture of my butt. Thanks for the offer, but I don't think that'll be necessary. And, thanks for the apology, BTW. I understand your frustration, it was the same with the Turbo guys before we really got people to sharing info. Now, there are a lot of guys doing thier own cals and contributing "to the greater good" so to speak. If Derek isn't willing to release his work on the 3.0 stuff, then I'll take a look at it. I'd hate to duplicate that effort, though.

Thanks for being such a good sport. I can be a real, rhymes with witch, sometimes. You seem like a really good egg Shel-Game. So I am very glad we are not at odds any longer, and I am really glad that you don't require the requisite butt kissing.

Chris at TU is interested in helping also, and I think considering that all you smart type guys are looking into this, we have a real good chance of getting this done soon.

Thanks again.

[Ondonti]

The romulator setups are really cool! The guys running that really have a great tuning instrument over burning chips.

[lookin]

Quote:

Originally Posted by Ondonti The romulator setups are really cool! The guys running that really have a great tuning instrument over burning chips.

Over, or instead of, burning chips? You mean you save the altered output code signals, in their device, and then you leave their device on the car? Have you ever used one? Have you ever seen it in use in real life? Are you now or have you ever been a member of the .... .

Sorry to grill you like that.

[Shel-Game]

You can leave them in, but they're not usually very weather-proof. Once you get a good cal, you should burn a chip and put it in permanently.

That said, it looks like we've figured out how to convert a SBEC to flashable (just like the SMEC's can be built now). Much cheaper than the Romualtor, but you still have to turn the car off and wait for the new cal to upload to the SMEC/SBEC (you can alter the cal on the fly with a Romulator).