Red91_Z

Hey guys,

As a Computer Engineer who's about to embark on the custom PROM programming journey for the first time, I read some DIY info and had an idea...

Why should I have to keep pulling chips in and out and flash externally? Why can't there be a memcal adapter that just has a USB port on it and I can just flash it via my laptop?

I've already come up with a couple of rough schematics where this would work out. There is still a lot of hardware and software development on my part, but what do you think... Is it a good idea?

Here are key features that I'm thinking about:

• 2 Meg of memory so you could hae up to 8 - 256KB programs on there at one time!
• Some type of switch that will allow you to switch what program you want.
• The ability to download you program to any of the 8 slots.
• The ability to read out any of the 8 and save it as a bin.

Are there any other cool features that would make this worth while?


I would test drive this on my '91 Z28 first and if I really like it, I'd sell them so that I could recoupe the engineering costs.

Any ideas or suggestions are welcome.

Jeff

MonteCarSlow

Been there, done that. LOL!

Last year I put together an emulator sort of like what you describe. Instead of USB, it uses the ALDL port for "flashing" the "prom". Maybe you've come across it already if you've searched the board a fair bit...

Red91_Z

Yep, I did run accross it. So is it 256k or can you fit more than one program on it? I was thinking that 8 programs would be nice so you could quickly flip a switch to see how different programs behave to let you know if you've done something right (or wrong).

I thought about programming it through the ALDL port, but I wasn't sure what type of communication it was (Class 2, J1850, etc) How did you reverse engineer it? That must have been tough!

mayh3m

are you talking about craig's ostrich? it's a realtime emulator that lets you tune on the fly. look into it... http://www.moates.net/product_info.p...products_id=57

junkcltr

iTrader: (1)

Do a search for Kalmaker, Winbin, or here for emulators in the threads from years ago. Others have made emulators years ago. Craig was the first one to push them commercially. When it came time to show what he built.......he blacked out all of the chip part numbers which was really funny when he basically used the same architecture he learned to build from this board. The chips are dead give-aways for the architecture but he tried hiding it.

The best thing out there right now is MonteCarSlow's NVSRAM module. Take a look at it before you build anything. Also, the so called Moates and Xtronics emulators are not real emulators. They F'up the ECM sometimes because the timing that the have is only quick enough sometimes. Try repeatly loading a large table and they will through on the check engine light. They are fine for fooling around on a test bench but down right scary in a vehicle.

Red91_Z,
Since you do code stuff, then you know what happens when a bus is corrupted with a bad instruction and how things go bad. That is what the emulators out there do sometimes. They corrupt the bus with a data value and the ECM uses it as an instruction. So if you build one and do it right and sell it commercially, it will probably sell good. Don't do like others and release it half-tested and keep telling everyone to update the firmware every two weeks. Funny, though because most around here call that "good support" when it is actually bad design release practice.

Anyway, you are trying to build stuff that already has been done.


EDIT: Hopefully if you build something with info you learned here at DIY PROM (yes DIY) then you don't end up like some others that learned how to build stuff here at DIY and then sell it commercially for a profit and stop posting because they trolled enough to get the info they wanted. Instead of keeping things DIY and teaching others (like they were taught) they moved on to targeting imports to cash in even more......sad.

Grim Reaper

Yes, we usually "blacklist" those types. And if they become "nasty" once they discover they are blacklisted, a banning may occur.

The good news now, TGO now has a "small sponsor" type Board for people trying to sell their goods too. Speak to the Admins for more information about becoming a "small sponsor" when your are ready.

If it is well designed, well supported and you remain a contributing poster; you can generate business on TGO if you chose to become a "small sponsor".

However, please do not put links in your signature telling people to go to your website to buy goods on TGO if you are not a sponsor. No freebies allowed.

Red91_Z

Grim,

No I'm not about to become some greasy salesman that just wants to win a sale with half-a** engineering. Those dudes irritate me. I get frustrated when trying to get free info on the web and all I can find is people tryin' to sell me the stuff I want to build.

If I did venture into this project it would be for the sole purpose of learning about ECM's and designing a solution that is useful. If it turns out that this works well and could help out the f-body community at large then I've achieved a bigger goal than I set out for (Just to get my ride running well).

I will look into the reference material in this tread before considering what to do next. But in the meantime if anyone has ideas of what I can do to make a tuner's life easier, let me know!

Jeff

Grim Reaper

No probemo Jeff. As a bit of background, the DIY Prom Board itself is primarily a Do-It-Yourself oriented Board where people come to learn about the ECM, eprom burning and EFI tuning.

This board does not permit "eprom writers" to offer their services, allow people requiring an eprom to get someone to burn one for them, or provide a referral service to where a person may get an eprom burned. Some of the other Boards at TGO may allow it (such as a Regional Board or possibly the new "Small Sponsor" Board)...but not this board.

However, a clever person, such as yourself, may see an opportunity for some entreprenuership. TGO now has a means for these entreprenuers to advertise their products and/or services (for a small fee). You must talk to the Admins if you wish to persue this matter.

This Board also has a unique following with many people with different vehicles (even non-GM) come here to help expand the knowledge base behind the ECMs. While this was NOT the first place on the Internet to help reverse engineer the GM ECMs and programming; DIY Prom is lucky enough to have a few of those originators ... (RBob & Grumpy immediately come to mind, though I am sure there are others I have forgotten to mention).

RBob & Grumpy came here in the inital days of DIY Prom from one of the very first Internet sites when some "individuals" (such as junkcltr is referring too), let a bunch of hardworking enthusiasts such as RBob & Grumpy do a LOT of R&D work.

And when that R&D work was basically done, they took that work and made a business for themselves without recognizing all the hardwork from others; either monetarily or even as "intellectual contributors".

That is why we are a "little touchy" here. But all newcomers are welcomed, and hopefully they will be able to be "contributors and teachers" one day.

Red91_Z

Grim,

I hear you loud and clear. If things progress to such a point where I would remotely think of producing and selling these to others, I will definitely look into the ad info you listed below. Don't worry, it is not my intention to clutter up this board.

For now I am on a information gathering quest to see if the ideas I have in my head are worthwhile or already done. I kinda want to hear from guys who have been doing this for a while and say "Man, if I could only program my car with ... or it'd be great if I could have a ... "

That being said, I promise that if this EVER develops into anything bigger than just my car, cudos will be properly distributed. (monetarily even if the contribution is large enough)

----

So here's my next question... Does anyone have a pinout for the memcal unit? (I.E. what pins do I have at my disposal for designing an adapter.) I have the pinout for the ECM, but can't seem to find anything about the internals...

Thanks,
Jeff

MonteCarSlow

Actually, it's only 4kb of battery backed ram that contains the calibration section of the bin file. What I built isn't really an emulator, but more of a hack to ECM software and hardware such that the ECM has the ability to update it's own tables via commands on the ALDL port. It works real time with the engine running. If you are handy with surface mount soldering, you can implement my ECM hack for about $20-$25. I only made it into a through-hole component module such that other folks could enjoy this hack too. Craig Moates can sell you a bag of parts to make yourself one...


EDIT: Here is a challenge for you. Take my DIY emulator and give it the ability to contain banks of calibrations. Only 4kb of 128kb is being used in the nvsram module... the rest of the address lines are tied to GND right now. If you take this challenge up, please post here documenting the modification so that the rest of us can benefit from it. My hardware/software hack is not 100% my own, it is built on a foundation of documentation provided by other board members and other folks on the 'net.

Grim Reaper

Here's a thought for you electrical engineering types (I'm software and it's been awhile). Is there anyway you could reset the Flash Prom while the vehicle is running? Basically make a Flash Prom behave like volatile RAM?

junkcltr

iTrader: (1)

Do a google search for "Ludis ECM". He has the memcal pinouts.

The bank select can be done pretty straightforward with a very small CPLD, some input addresses, and the data bus. My second design emulator uses this type of architecture and a complete design fits on a 2.5" x 4" PCB. There is plenty of room on the NVSRAM module to have just the CPLD. That is using through hole parts so a DIY could build it.
----------
That was my first attempt of on the fly tuning before I built emulators. With the commonly used FLASH device, no. Part of the device needs to be erased before it is programmed and the ECM will have a fit when it accesses the part that just got erased. NVSRAM does not have this problem. If you used banked FLASH memory then it is possible.

Red91_Z

Grim,
I'm not 100% sure I understand the question. What would you want reset? Do you mean erase the entire part? Flash is sticky memory so on power reset it will retain the same data from before power down. If you want to reset it to 00's then you need to perform a page erase which should execute very fast.

Jeff

Red91_Z

Cool, thanks for the tip. For my design I was thinking that I'd have a PIC micro onboard that would do the address bank generation as well as the USB communication to the outside world. After reading you're earlier post, I'm thinking to prevent ECM hiccups that I'd buffer the data bus. Then during programing the PIC would force the outputs to read as continuous NOP's.

I'm assuming this would work from my limited knowledge. The PROM only conatins program code and lookup tables, correct? Stuffing in NOP's in the execution stream should harm anything except that outputs will stay constant during this time so things like the injectors will not change during this time. Obviously the goal is to design a SUPER-FAST program time so that this time is kept to an absolute minimum.

Again, these are just ideas that I have.

Jeff

junkcltr

iTrader: (1)

On the real time operation of the microcontroller the page erase is very slow. UC operates at the MHZ range and the FLASH erase is in the ms range. It isn't going to happen.

junkcltr

iTrader: (1)

I have an emulator that works kind of like that. It uses a Serial FLASH for the program and loads up a normal RAM at boot time. It holds off the MCU with NOPs until the RAM is ready. It also fits on a 2.5" x 4" PCB but required a surface mount CPLD and processor. The benefit was not having a battery to go dead. Yes, you can hold off the MCU in real time operation just not too long. That is how the slow emulators out there should be doing but corrupt the bus sometimes instead.

Yes, PROM contains code & data. I think this is about as far as I can go with stuff about designing emulators. I told Craig how to do it years ago and then he went commercial with it along with a decent price mark up instead of a DIY kit. The best I can say is to search 2003 and 2004 posts where I was telling Craig what signals to use and how to manipulate them. The sad part is in the end his design loaded the bus with resistors to work around the bus contention instead of doing it right with digital logic. The bus loading will burn out the MCU over time. Not like most cars people will understand. Or why the CAL change emulated doesn't work right sometimes.

Sorry for not being able to help more. I learned my lesson years ago. I will say this, it can be done with a 68 pin 128 Macro CPLD, a 28 Pin MCU, a single RAM with battery backup, RS-232 or USB chip, Voltage regulator, and some caps and a couple of resistors. All on a PCB that is 2.5" x 4" with all through hole parts that a DIYer can build. That is on a double sided PCB. A 4 layer is required if you don't understand electronics and hack something together.....yes there are large 4 layer PCB emulators out there because they couldn't get the signals clean with 2 layers (keep in mind this is low speed signals with slowish edge rates).

Good luck and have fun researching it. It is a good project. The NVSRAM module that MonteCarSlow is far superior to any emulator out their in terms of design and reliability. It will fit in a 730 or 749 ECM. It appears to fit fine in a 427 (95 Chevy P/U ECM) also. It requires some code changes, but you can pick up a PROM burner cheap.

TRAXION

Didn't the PROMinator do this?

http://www.speedtronics.net/aboutprominator.asp

t

MonteCarSlow

This is where things get more complicated then they need to be..... buffering crap and adding another CPU to take care of the bank swapping, etc.

I used the ECM's CPU and simply added some nvsram into it's memory map (to hold the writeable calibration so that the ECM can make it's own changes). The rest of the hack is simply in the ECM's software.
1/ modify the fetching calibration data from the nvsram instead of prom
2/ modify the ALDL mode 4 command to write to memory locations

junkcltr

iTrader: (1)

Maybe? I never looked into it. In terms of emulators just about everything has been done at this point. He would be better off searching than trying to do any traditional methods like DPRAM (Winbin guy), NVSRAM (Craig & Monte), Banked FLASH (Prominator??). On top of that there is my Serial FLASH emulator. Well, I have one more idea but it requires a high speed chip and only one chip. Turn to an FPGA FLASH based or RAM based and it is a two chip solution (very small). Ideas are endless....it just depend how much time he wants to play around with this stuff. Only a few stick with the DIY stuff and most just come and go.

Mangus

shoot... real reply coming...

edit... see next post. Sorry, the "submit" button had input focus while I was typing in the edit control. Hitting the spacebar caused the post to be prematurely submitted.

Mangus

Junk - can you qualify this last statement? Where exactly did Craig "learn" this architecture from the board. If I recall correctly, he sent you the entire layout, and you said it wouldn't work. I was around during the inception of Craig's designs, and can attest that he did the majority of the work himself. No, he didn't go out and design the Atmega from scratch, or build the AVR compiler. No, he didn't invent electronics theory, or even the underlying concepts used to design the products. The stuff isn't rocket science (as a rocket scientist by education, I should know!)

I have yet to have a hiccup with the Ostrich. I've even uploaded entire binaries with the engine running. And what's with the "so called?" Your post started out innocent enough, but here your tone is getting dark, almost as if you had something against these people. I'd like to make sure that Craig doesn't get put on the same level as the DIY-WB debacle. He doesn't deserve that any more than he deserves your snide implications...

...like that one. Others? Again, qualify statements like these, otherwise you make enemies of those who are actually very friendly to these boards. I'm really starting to believe you have some aggression (even if it is passive) towards these "others."

Where is this emulator, Junk? You've mentioned it a couple of times, but as far as I know no one has ever seen this. "In theory there is no difference between theory and practice. But, in practice, there is." —Jan L.A. van de Snepscheut

Not the same Craig Moates that Junk seems to be saying bad things about? No, can't be.

Now that's just it, isn't it? No one really reinvents the wheel. Monte knows that. He's been a very gracious, constructive contributing member here. Even RBob's product isn't a result of information that only he originally gleaned by practice. Rbob, in my view, is probably the single most important member of this board, and in my view he has contributed more than anyone. He can probably lay claim to more information on these boards than anyone else. So why does the majority of content in this thread seem to include so much smack talk? Junk has contributed more in the last year than he has previously, and he is gaining respect in my book, but then he loses it in posts like these. I don't understand it, really.

Why not offer constructive help to Red, rather than tear down other heavily contributing long time members of these boards. Again, Craig doesn't deserve any of that. I'm sure it's not just Craig you're implicating here, but I feel Craig deserves a defense.

Red - go on with your bad self. Develop a cool gizmo. Use the information that's available here. Make it available to people. Give back. Give credit where credit is due. Ignore the peanut gallery.

Mangus

I should add that I personally take offense to this comment. As a software engineer by both profession and as a hobbyist, this is just the way of the world. Your iPod has firmware updates, nearly every piece of software on your computer, be it Linux, Mac, or Windows, does this. My Soundbridge does this. My ReplayTV does this. TunerPro does this. CATS does this. EFILive does this. Hell, even my computer monitor does this.

You're just trying to be an a**, and it's not appreciated. There is *nothing* to be gained by anyone from such a comment.

Red91_Z

junk,
You've already helped me out a lot, so don't worry if you can't carry on anymore. These are all very interesting ideas. I still feel life there is a simpler way to custom program ECM's that doesn't require me cracking open the ECM everytime *and* can use standard PC equipment like USB cables. This is really the design goal I'm striving to. I don't have time to mess around with the normal EPROM unplug-burn-pluggin process because I have kids and if I get a hour or two to work uninterrupted I want to make it useful!

For me this idea is two-fold. 1) I can speed up my tuning process 2) I love the act of breathing life into something that I built ground-up. That's why I'm an engineer, it's what I love to do. So I may end up doing this for the selfish reason of creating a memcal adapter that is new and unique. (not to mention learning about my car from literally the lowest part outwards)

I'm still on the fence on this one. Monte's design looks like it would work for me, but the temptation is still there to make something from scratch. Monte's challenge to improve on his NVSRAM is also quite tempting.

For the meantime I'll keep digging into different chip architectures, and you'll probably see the occasional post from me asking more about the 3rd gen ECM's. Also, I hope you guys don't mind if I keep bouncing ideas off you for another memcal adapter design.

Jeff

Red91_Z

Mangus, thanks for the inspiration! I would love to build a cool gizmo. In fact, I have to exercise much restraint in keeping myself from buring my head in this idea for the next several months. I'd also like to contribute more to the f-body collective in general. I've been looking for a way to do this for a while now, and this might be a good way to do it.

No secrets here, if I do anything with this the information will always be free. If a board materializes out of it, then I'd sell that just to recoupe the engineering development costs (anything past that would be nice but not my main goal). The internet is for sharing knowledge, not hacking your wares.

Jeff

MonteCarSlow

I use a USB cable with an ALDL converter. My ECM remains under the hood of the car (or behind the dash), and I use the same cable for data logging as I do for reprogramming. It is just a matter of switching between applications on the laptop...

The gotcha I have with my design is when I want to modify the firmware of the ECM, then I need to crack it open and change the eprom. But for calibration changes, it works beautifully.


EDIT: What would be neat if you decide to pursue your PIC idea anyway, is to make it a bus sniffer as well so that the P4 ECM gains the feature that Rob gave the C3 ECMs. Without it - you still require the ALDL interface for data logging. And then you have 1 standard USB cable, and the ALDL box and another USB cable...

Grim Reaper

Please, let's keep the emotions down. Or I will edit out remarks as I would like to keep this post open and on topic.

But since this point was raised, I will add my and hopefully we can get back to the original topic.

As a former computer programmer & systems analyst in my early life, I agree with you Mangus that no matter how much you test; unless the software is EXTREMELY SIMPLE, there is apt to be some "gotcha". There will always be some "user" who does something that no one ever thought was possible. And that was in the days before there were malicious hackers trying to exploit some hole in the software.

But, I (as well as others) have also been the unfortunate purchasers of software where I doubt the developers did the simplest "Alpha" Test. Not only would their software not perform as expected, it would crash my system to the point I had to go to an older backup/checkpoint before I installed the offending piece of junk. If the developer had been sitting beside me, I would have rammed his head into the monitor so he could have real good look at his crumby code.

junkcltr

iTrader: (1)

Mangus,
I was not trying to offend you so I don't see why you wrote all this. The strange thing is that Craig was logged in viewing this thread and did not respond at the same time you responded. I don't think I have said anything that is not true and I think that is why Craig agreed with it and did not respond.

I respect you for expressing yourself. You are honest and say what you mean. You are not sneaky like him. For instance, back in 2002/2003 when I stopped posting for a while and did not have an email or PM setup on here (on purpose), Craig got the MODS to give out my PRIVATE email address to him. We emailed back and forth as I helped him with his design. Did he ever tell you about that? He was making the AutoProm back then and the power stuff was marginal, the signal integrity was marginal and a few other things. If you search the 2002/2003 posts you will see I told him how to use the OE/CS stuff and MUXes and I think I even gave him part numbers. After he had been told the architecture, he made a schematic and sent it to me for review. The thing is that he knows just enough of electronics to be dangerous and build hobby electronics. He had no understanding about designing for min. and max. tolerances (devices do vary from part to part) and a reliable design for all device specs/datasheet info. He would just build something and call it good if it worked. That doesn't mean it would work for all ATMEL AVRs, LS logic, or ECMs. That is why it works for some and not for others.

I will do my best to answer your questions.

I still have the emails from him if that is what you are asking.
Craig learned about the CS/OE switch time and bus arbitration on the DIY board.
I never said the entire layout wouldn't work. I said to watch the bus arb. timing. I said the power supply stuff was marginal and over complicated and to remove the ALDL stuff. Funny, a year or so later the Ostrich came out......no ALDL stuff on it.
Yes, you were around then but haven't been told the whole story.
Craig always said he doesn't understand C so he wouldn't be using the AVR Compiler. He still does it the old way with AVR Assembly.
I have never seen a rocket science course being offered anywhere. Where can someone be educated about rocket science? What does a rocket scientist actually do?


Maybe he finally figured it out after 3 years. It is impossible to load two different bins with considerably different code and have it run without tripping the bus. This can only be done with same code bins or very similar code. That is a jump to address xxxx will be different between the two. As a rocket scientist, you should understand that.
My post is innocent. I have nothing against anyone. I justed wanted to see if the original poster (Red91_Z) was a sneak. No, Craig has nothing to do with the DIY-WB. There were no public schematics that I know of when he built his thing. He seems to be OK with what I posted since he didn't respond when you did.

Others are the people before my time here.
Mr. Bill, RBob, MonteCarSlow are all brilliant with there own original ideas. I have a great amount of respect for all of them. They don't do sneaky things and claim as their own.



The first wire-wrapped proto-board one is in a box.
The Serial EEPROM one is in a box.
The CPLD one is on a bench.
The Bank switch one was never completely built for action.
Are you looking for pictures or something like that?
I can tell you that TunerPro doesn't use the emulator internal EEPROM like TunerCats does. The whole thing about the "hard/soft pul-up" which is actually a design flaw for the Xtronics emulator, but they try to say the GM ECM is flawed. That is funny because the emulator was designed for the ECM. So by definition the emulator is flawed.
Interesting quote. I always liked this one "Stupid is as stupid does"--- Forrest Gump


You are referring to MonteCarSlow's NVSRAM module that is a DIY kit. Since all the Craig built stuff is non-DIY then I guess you proved my point instead of yours. Not trying to offend you here. Think about it for a second and you will see it is true.


No one reinvents much anymore. It is the matter in which things are done. RBob is a truly smart person with an excellent product. He is one of the most brilliant DIY PROM members and has done a great deal for the DIY EFI people and continues to do so. MonteCarSlow's product is great and he released it as a DIY KIT and his AUJP code patch is open source. I don't see how you could compare them to the what Craig did.
I can't help you from losing respect for me. I debug and send you the info for TunerPro bugs. I respect you for building a stable product and keeping it DIY based/low-cost/free. I don't understand why you associate yourself with the way the CM emulator was designed.


I believe I did. I told him what is out there already and some of what could be done. He said in his last post that I have already helped him a lot. Notice the commercial emulator designer that is a member didn't say a peep (just lurked). Only the DIY guys helped him out. I am guessing that CM is taking a hard look at Serial EEPROM datasheets as I type....he probably didn't think of this before. Yes, I saw him logged in lurking this post.

I am not trying to upset you but you are the peanut gallery in this case. You didn't not post anything about memcal adapters. Instead, you tried to turn this into a post about CM stuff. You don't appear to do electronics and do software stuff. TunerPro is a great tool and seems stable and you support it and keep it DIY oriented. You continue to contribute to the DIY community. You do a lot of good things for all the people here and I respect that. This thread was amount memcal adapters.

I hope more people contribute ideas to Red91_Z for what could be done with the adapter. I was trying to give ideas of the technology side and a little background about the DIY PROM stuff and keeping things open so others can learn. It is the lurkers that ruin it and cause people to keep things closed so that they don't grab it and sell it commercially.

Craig Moates

Junk,

The reason I didn't post to this thread is that I didn't want to engage at the level of personal attacks. And frankly, the moderators should not have tolerated your comments.

Your comments were indeed and continue to be out of line and I do not agree with them at all, particularly in reference to me.

Furthermore, I believe that you have overstated the degree to which you have contributed to any advances which I or anyone else might have made. You've also misrepresented my role on this board. I am not a lurker, profiteer, or thief.

What you HAVE done here is to foster an adverse environment where folks are afraid to make any developments or contributions. Think about it, if someone comes forward with an offering that may have somehow been referenced even in passing in even an obscure thread, they can look forward to being inappropriately fingered as I have.

When I feel that I have something worth contributing, I do. If you can somehow quantify the degree to which I've contributed by making tools available, both hardware and software, and bringing new people in to the hobby, then you'd be more on the right track.

It was kind of you to review the APS1 design several years back, however, I had no idea that you would begrudge what little you did offer so harshly. If I had, then I certainly would have never approached you. I was excited to be so close to a functional prototype, and you seemed like someone who knew what they were talking about, and I thought you might share some level of interest on the subject.

I have nothing to hide, and would be glad to share the details (as I have in previous threads) of any of the devices which I make available as well as many I have in mind. The whole 'blacking the chips out' thing was just to let people have a little fun with the guesswork, not to conceal chipsets. And of course I've been working with serial EEPROMs for some time now, they are present on at least 5 different prototypes for various things I'm working on.

I enjoy working with other folks toward a common goal. However, there is a certain point where if you want to truly contribute, you need to give people something they can use. Sitting around and talking about it just doesn't cut it. You've got to provide something which helps folks apply what they learn without having to cut their tools from copper and stone. As an added benefit, I try to offer these tools in a way that forces the 'newbie' to read, research methodologies, and basically 'learn to learn'. Then they can teach other folks, and therein lies the real benefit of all this. Bringing new people in to the hobby, forcing them to learn, and letting them experience the joy of the results. Let others share in that experience by passing the teaching on.

The fact that you're calling me out personally to defend myself is just rude. I'm a PhD Engineer that has worked with practical, applied, industrial research and development for the last 10+ years. I've been well respected by my peers in the profession throughout, and have always been sought out in order to reduce things to practice in a timely and effective fashion. My work has been published in several high-level professional journals. My background isn't even in electronics, but I've always had to make use of them for data acquisition and control purposes, and I enjoy the challenge of learning new things. I'm really not trying to honk my own horn here, but it is important to put things in perspective.

I enjoy working on cars, especially my own stuff or helping friends. That's how I got into this whole thing, making a chip adapter for my own vehicle when the hacked memcal and short ribbon cable just didn't cut it anymore. And then I got tired of digging behind the dash and erasing UV chips. And then I got tired of having cables everywhere. So I built the tool which I wanted for myself: An all-in-one chip programmer/emulator/datalogger. But, unlike others, I wanted to make my own tools available for folks to benefit from.

As far as profit, it really isn't what you characterize. Unlike others, I have fully dedicated myself to the EFI tuning hobby by making enabling tools available to everyone. I feel like the pricing is probably lower than it should be on many things, and I'm not sitting back collecting a check. How much is time worth? I talk with folks for hours over the phone, walking them through everything from troubleshooting Microsoft Windows to putting their chips in right-side up, along with at least what I know of tuning strategy. I answer dozens of emails every day, people call me at home, they come by the house. I get to meet and talk with a lot of good folks. I spend every waking moment available on this stuff, in hopes that I can help someone along that really is trying to learn. After it's all said and done on ANY given day, there's nothing left of me. There are certainly more profitable ways that I could spend my time, but that's not what it's about for me.

In terms of DIY kit offerings, I did that for a good while with the simpler devices like chip adapters. After seeing how much frustration folks ended up with, and how many mutilated bits were coming back, I just decided that it didn't make much sense for me or the users. The NVSRAM kit makes sense, because folks messing with that are a bit more hardcore than someone who is trying to disable VATS or change their fuel injectors.

Can we just put this behind us and get along like adults here? I'd love to contribute to the discussion, but if it is going to be met with back-handed sarcasm and sour grapes at every turn, then it is just going to get my blood pressure up. I'm not trying to build on your conflict here, but making some effort at peacemaking.

Red91:

With respect to the chip adapter, I think it's a great idea. In the simplest incarnation, you could have a switching adapter that is able to mux the address and data away from the target device during reflash while the vehicle is powered down. Then you could safely reprogram the resident parallel memory as much as you like via a small USB/Micro combo.

If you like, I could do at least the basic PCB schematic and board layout and you could use that as a starting point reference. Could even make the firmware compatible with TunerPro for some seamless integration. At least give you a starting point or something to think about.

Best regards,
-Craig

MonteCarSlow

Alrighty. Now that we've all had a turn, can we get back to having fun??


Red91_Z: Have you decided what you want to build? I have another suggestion for improvement on the nvsram that I'd like to do this winter. I'd like to utilize the ECM's standby Vcc to power the nvsram, to minimize the drain on the lithium cell that is inside the chip. Do you want to collaborate?

kaylin4u

Craig, since you were on this topic, I wanted to ask you something. Is there such a cable to allow you put your chip in a ZIP socket remotely for ease of chip replacement instead of pulling the ECM?

I really like the Ostrich idea, but being near Christmas, I can't buy a toy like that.

Craig Moates

Monte,

If you were to do something along those lines to use the standby Vcc, you could go so far as to utilize an SRAM. Then you could cut down on size, cost, and have a much larger memory area. That would give the capacity for the switcher stuff that Red91 visualizes. Something maybe to consider.

Kaylin4u,

I think it's possible. I'd tried something like that before with just the cable and it worked sort of, but not at any great lengths. I think it could be made to work probably just by buffering the data outputs back to the ECM though. You could put together a small daughterboard at the EPROM side with a 74hct541 or something like that controlled off of the target device CEorOE lines. You could even use ATA66 type cable and take every other conductor as a ground. But I think just having the address info going right to the EPROM and then the data coming back across the cable driven by the 541 would be adequate.

scooter

iTrader: (11)

I have an autoprom which I know how to use but have always thought there must be a way to hook it up without having to pull the ECM out of the dash every time. Forgive my ignorance but would your idea sort of incorporate something like this?

Craig Moates

Scooter,

Sounds like it, yes. What has been discussed here would give you at least the ability to reprogram your chip without having to swap stuff out all the time.

Kaylin4u,

I've set up a (Cadsoft Eagle www.cadsoft.de) layout for a remote chip socket that should work. If you're up for it, take a roll and put something together. It's pretty simple and compact, with components on both sides, but it should work:

http://www.moates.net/projects/