That will get Craig's attention! Craig - imagine writing firmware in something close to quickbasic!

 

Check out the ATMega Development Board

It comes with an AtMega163, (a step up from the Mega8535, can be upgraded to a Mega32) a prototyping area, serial i/o chip, and a paralell programming cable and program all for $33. I have used them and they work great.

Bascom AVR is another good basic compiler if you want to go that way.

Greg

 

Those are nice boards Greg, I ususaly dont prototype with anything that fancy, just a 3300 pin breadboard, If it'll run stable on a breadboard it should run okay on a real circuit board.

As far as Bascom goes, I tried it, it was okay, but seemed to produce bloated code compared to FASTAVR, which seems to be realy well optimized.

A wideband controller with a serial interface, two voltage outputs and a display driver should all fit on an 8k AVR, but the cool thing about these chips is that you could upgrade to the 32k mega32 without any real problems as feature creep sets in later.

As far as the choice of assembly or a higher level language goes, when done right assembly is obviously more compact and a bit faster (depending on the architecture and quality of the compiler and assemblers being used) . I like the higher level stuff so that the code is really fast to develop and easy to read by other amatures.

I'd have already done this myself a while back if I had all the equipment needed to develop a wideband driver, but unfortunatly I dont have those resources here.

The Punisher

 

Way back when........
Or once upon a time.

I was running an ecm bench in closed loop.
To generate the feedback of the O2 I just pulsed a simple resistor divider circuit with a reference pulse. And then feed the *O2* pulse into the O2 sensor's input.

That was for a TBI, on a port set up, I think I had to add a 555.
Maybe it was a 556.

Anyway, I think things *may* be over thought about response times, etc..

The thing to do IMO, is mock something up, and actually run it.

The closed loop stuff is just for running at Stoich., so it's not *that* critical. The AFR is constantly toggled over stoich it not that the ecm is looking to maintain a constant 14.7 anyway.

 

Yeah, the BasicAVR stuff had me intrigued early on, but I opted for the ASM route on my recent projects for the code efficiency value.

That Mega163 is a nice piece as well. I had my eye on it for another project but it only has the one UART ;^). For A/D along with dual UARTs I guess I'll be stuck with the Mega128 as far as I can tell.

For this project, there's also the Mega168 which could be cool. Only 28-pin DIP. Nice.

I like that little dev board, pretty slick! Might not be entirely necessary though. Once we decide on what AVR to use and get a couple of the primary circuits decided on, then I can throw together a board in no time with ISP. I've already got a solid little bootloader routine, and Mark's software can be used for the firmware updates as is.

 

yeah the input value on the ecm itself floats at 450 mv or so. with no sensor switching itll stick at 450 mv. so obviously the sensor is pulling the line up and down depending on which way the AFR swings. ill see if i can figure out some sort of buffer circut.to do the job.

 

Probably a day late and a dollar short but...
To keep it "simple" and keep the original DIY devlopers "happy", could you just post the board layout and parts list? And be specific on the sensor ie exact P/N?

With just a board layout that can be printed out you can get a product called "press & peel" and print your own boards and etch them. Thus taking some of the "burden" off of Craig. Then if someone simply can't solider their own they could at least etch a board, buy the parts and contact someone here that does solider about an acceptable price to build it for them.

BTW, a display would be a real nice addition.

 

anyone have any Updates ?

 

I've got 32 prototypes built and ready to go. There's an active GaAs proximity screen (320x240) that projects a heads-up display on your windshield. Also, it's wireless, so no connection is needed between the sensor, control box, display, and ECM.

Anyone interested in beta testing?

 

Sure thing, just tell me what I need.

 

The circuit draws about 40 amps intermittently due to the rapid rise time on the heater and the high-intensity GaAs screen projection. Might need to check your alternator's capacity and make sure you can spare the extra draw.

I tried to keep the size down, but it ended up being about the size of an old-school lunchbox.

 

Wow, that's big.
I have a 108 amp alt, should do ok, no big amps or "extra" gadgets in the car. The wiring is something to look at though, to handle that kind of amp flow. Might need to run directly off the battery with an on/off switch for the "box".

 

Hey Craig, is this thing going to make me fab up a small trailer and put a trailer hitch on my car? I guess if I used the gold plated 0 gage amp wire all the way back to the trailer it would be ok. Or better yet just put a gas powered generator on the trailer itself.

Oh never mind, I just solved my own problem.

Steve

 

if you have a box this big run a extra alternator. the trailer idea is good. load the car down really heavy makes it easie to tune(slowsit down) . opppss did i just give away a tunning secret.

 

Could hook up a load cell to the trailer and do dyno pulls on the interstate...

 

"YES DIY TRIALER LOAD CELL DYNO AWESOME !!!!!!!"

in all seriousness actually something that ive been toying with for a while. need some good wide sticky tires however to make it all work though. then some big eddy current crane brakes and a old unused dana 80 or something with a diesel generator for a arc welder.

 

Craig (and anyone else interested),
I just read about this service in the latest PC Mag. Have you heard of them? It says they will do assembly also and do price quotes based on your designs and components used...sounded interesting.
http://www.pad2pad.com/

 

pick me! pick me!! i was about to order one next week.
im about to put the finishing touches on my new motor and have a friend thats tuning an turbo ls1..

if you pick me what sensor do i need to get?


and i just picked up a 7749 ecu, hey grumpy still looking for betas too??

 

I'm beginning to tune my 406/Stealth ram combo. I'd love to beta-test for you. I'm assuming you're joking about the wireless stuff and the HUD, sounds like the opposite of the goal of the project, but I'd still love to test an AFFORDABLE DIY Wide Band kit.

 

The beta testing offer was a joke guys.

 

see what happens when i post past my bedtime..

 

Have any of you guys seen this: http://www.vems-group.org/index.php?page=AfreshBoard It looks quite old, but it may be a good resource for anyone working on getting a DIY-WB LSU4.2 kit built.

Also just found this, may be old news: http://www.megasquirt.info/PWC/

{edit: link updated,mod}

 

Craig, you gotten to do anything on this lately?

 

Heh. Look what happened last time this was asked ^^^

=)

I'm reasonably certain that updates will be posted here when they're had.

 

Last I talked to Craig he was working on the active GaAs proximity screen . Trying to get it to display at 640 x 480 as the 320 x 240 was a little hard to read

RBob.

 

Sorry, I didn't mean it like a noob question, I just wanted to make sure the idea hadn't been dropped, since there hadn't been any more convo on it.

 

Craig, I was wondering while you were working on a wideband display, could you also add readouts for your ECM software. What I'd like to do is permanently mount a display so I can always see the engine readouts.

I figure I can mount a display on the passenger side dash. I don't want my laptop sitting on my seat for the rest of my life.

 

The whole thing about the fancy display on the WB was in fact a jest.

Haven't had time lately to pursue WB work, but I'll be looking forward to the collective contributions of others on this one...

 

The display was something I had been wanting to ask about ever since hooking up your software. Would it be possible to make a small display that showed just screen #4?

I stopped reading all the replies after it got too technical for me. I don't even know what a Gaas screen is.

 

The whole display thing will be a separate topic here before too long, particularly with respect to datalogging ;^).

 

=)

 

Excellent!

 

I just got 3 free atmega8535s in the mail today. Thanks, ATMEL!!!

Let me know if you want me to do anything. With them. (They look pretty nice on my desk.)

 

Things look a bit slow right now, but I myself have some plans to work on this project shortly after I get finished with a temporary out of town job I have for the next two months.
If somebody with good negotiating skills could talk the people working on the megasquirt-AVR into releasing their wideband driver routines and circuit that would be a tremendous help on this project.

 

Progress?

Regards,
Cobra289

 

I thought the megasquirt stuff is open source?

http://www.squirrelpf.com/msavr/inde...e=OnlineCourse

 

Well I chatted with Craig for a bit...I'm willing to take on a new "DIY WBO2" project...once SpeedReader is done...BUT, I'm not willing to get into "feature creep". If I do it, the thing will simply give a 0-5V output...nothing more nothing less, so please don't ask for extra features. There are too many systems out there already that will input an analog channel and do all kinds of fancy crap with it...such as AutoProm (and SpeedReader will too). We can do something like an overkill processor, and leave some "proto" room on the board for extra circuitry...and anyone that wants to develop features for themselves...well knock yourself out.

Now with that said...I'm willing to do everything myself if need be...but I'm not really looking forward to firmware development. I will do everything hardware related, no problem there.

So...firmware volunteers (that are familiar with WBO2 operation)? Please Email or IM me. I'm also on Yahoo IM often as redcorvette_84

I'd prefer to keep the language in either assembly or some form of BASIC...just because more people can understand these than C/C++, that way more people might be able to design in their own "features".

Bill

 

Actually if you want to handle the hardware stuff I'd be glad to contribute on the firmware side if it's AVR-based. When it comes to fabrication, distribution, and what-not, I can chime in there too if need be. I just don't have the spare time to shoulder the hardware design and protocycling.

There's some solid stuff out there in terms of existing designs. MegaSquirt, DIY-WB, etc are all right there. You'll need to negotiate how to come to terms with all that with respect to IP and design rights.

Didn't mean to drop the ball on this one, I was more trying to stir the pot a bit and see where people's heads were at. Good to see Bill picking up the stick.

 

MrBill & Craig I would like to thank your for keeping this alive.

Now with that said I have a Question. Please don't concieve this as feature creep but would the 0-5V be able to work a 3 diget LCD readout, My point is that when I am in my car I sometimes have passangers and Stuff in car going some place and the lap top isn't always runing. I think that a on board Header of some kind should be on it so that it can be made to work with a readout or Guage, that way a single Glance I know what is going on with out Turning on the laptop and datalogging.

 

Hi MTPFI-MAF,

The answer is YES.
On the first mail from Craig, he mention:
So you can use a digital voltmeter with you desired LCD, you can convert the volts 0-5v to a real AFR digits.

Craig and Bill,
Do you have an agreement and a path forward?
Do you need more people involvement?

I would like to hear that this project have a GO!

Craig,
I think, reading all the responses, you have the blessing of the DIY-WB people. So lets GO man!

Regards,
Cobra289

 

Well MrBill here are the Technical Files I have on the LSU4 they are located http://home.ripway.com/2004-7/149764/ It is Labled LSU4Tech.ZIP

 

Anything new on this? I haven't been around the boards lately because school is killin', so I wouldn't have seen any of ther threads, but I keep an eye on this one. I'm still interested!

 

Well now I agree that No Guage out put is needed.

When this does happen I would like to Get One of PLX's 2 1/16th WB O2 Guages.

The guage is 0-5v Linear, would it Mess up the readings for dataLogging if The guage was spliced into the same 0-5v output that will be used for data logging?

Here is the Scale that they show for the guage



and finally here is a link to the page http://www.plxdevices.com/2InchGauges.htm

 

They should really swap the axes on that.

 

As long as both devices have high input impedance it should be fine. If either one one does not, you could just put an op-amp follower at the output of the WB-O2. It will adjust current at the output to maintain the output voltage regardless of the current drawn by the gauge (within the op-amp's limits).

 

MrBill
I mean No Disrespect, But are you still going to be Taking this on or is there still other obligations.

 

I'm betting the Speedreader has MrBill's hands pretty full right now.

Teeleton

 

I thought Craig was going to have a heavy hand in this. I'm very disappointed that this hasn't seemed to have gone anywhere, I figured a lot of people would be interested. I can't place any blame, though, I don't have any skills that could help. If I were an EE student, I'd be inclined to make this my design practicum project next year, but I'm ME.

 

Craig agreed to do the firmware, and I agreed to do the hardware...the problem of course is time. Right now I am concentrating more on Speedreader....whenever my "regular" job isn't so busy (which it is right now).

 

I've got a little something in the works. However, its grown to have so much feature creep and use all SMT and non-hobbiest micros, it could no longer be considered a DIY project.