Wideband Logging; was Tunerpro RT datalog aldl def

[ryan.hess]

Alright, it's done... just have to find time to test it. Came down with the flu though, so I'll be stuck indoors for a while. yuck.

[Mars]

Too bad you got the flu.

Glad to hear you got the thing done though... I wish I could test it out, but that won't happen anytime soon. We are getting BURIED up here this year...

If you can make a little write up on which pin, and how to connect the wire into the ECM connector, that would be cool.

ALSO, does the 0-5V number come in as a 0 - 255 variable (8 bit)? Just getting my factors right before I do anything else. My LC1 is programmed for 7.35AFR (0V) to 22.39 AFR (5V)... So the correct formula should be :

(X * 0.0589 + 7.35

...should display the proper AFR for datalogging, I think...

Later. Get well soon.

[ryan.hess]

Well, I made the assumption that the 8f code would use mostly the same wiring scheme as that that I'm already using. I guess that's not the case, because I can't communicate with the ecm via aldl, and my coolant fan kicks on.

Bare minimum to test it, I would need to know which pin is the aldl serial data. So, I guess I will need those wiring diagrams.

[Fierobsessed]

Should be as easy as hooking up the WB02 to the open port, then look for the ram address that it modifies the contents of, and put a call to that address in the ALDL stream callout section in the data section. You can put a simple add on program into the algorithm section to utilize the filtering subroutine, just assign it a coefficient address and a coefficient, and you are well on your way to datalogging with a wideband. Theoretically it should be a breeze.


If the dissassembly of the 8F code I have is correct, then change the value at 0F80 from "EB" to "1C" And hook the wideband to pin "A3" on a 7730 or C28 on a 7727.

Then look for fuel pump volts on your scantool, and that will be your wideband output. 0.0 -25.5 and depending on the scaling on the wideband, mine in particular would be 0 is 10:1, 25.5 would be 20:1. 12.0 would be shoich for me. This uses the TPS2 input, and sends the information to location 001C on the ram, witch is called out in the ALDL stream definition and displayed as the Fuel pump voltage, witch is called out at location 0F7F (or 8F7F if you want to be technical). If you are using a computer based datalogger, you could possibly rescale (factor and offset) the output into something slightly more useful, such as an actual AFR.

This sounds reasonable, no?
Spares For 1227727's:
C28 (TPS2), A14 (MAT2), C21 (MAP2), A5 (MAT3)

Try it out.

[Fierobsessed]

Oh, and BTW, on the 7727:

B3 is the aldl mode, and B5 is the serial out.

Figure I will point my sources, You need these if you are going to go down this road.

Raw pinout non specific:

http://www.cruzers.com/~ludis/quadplugecm.html

Great website for the down and dirty ECM info,

http://www.cruzers.com/~ludis/

Good stuff in general, it has "as per application" pinouts for a lot of good ECM's: (thanks DarthFiero!!!)

http://www.gmtuners.com/

on the left towards the bottom download wiring diagrams and pinouts.

[ryan.hess]

If the dissassembly of the 8F code I have is correct, then change the value at 0F80 from "EB" to "1C" And hook the wideband to pin "A3" on a 7730 or C28 on a 7727.

Then look for fuel pump volts on your scantool

well yeah, that would be the easy way... :P However, I think fuel pump volts can be an important diagnostic, so I don't think that would be the ideal way to go.

I think I forgot to recalculate the checksum after I applied the patch... I'll do that and try again.

You know what, upon reviewing your suggestion, I didn't even know TPS2 was being read in the code. In fact, it's being referenced multiple times, and subtracted/compared to the actual TPS reading. I have no idea why this is being done, but it may be a bad idea to use this pin. It may provide some kind of ground reference. I'll have to look into it further.

[burn]

Does a complete WBO2 unit have to be used with the reader, or is this programming so that you can just get the sensor and hook it into the stock ECM?

[burn]

also, I know this isnt for the ECM to trim the engine with, its just for fluidity in the datalog...

[ryan.hess]

Does a complete WBO2 unit have to be used with the reader, or is this programming so that you can just get the sensor and hook it into the stock ECM?

mmm... yeah. This will read the output of a wb sensor. It won't be able to control it. There is a cheap WBo2 sensor/controller you can put together yourself at wbo2.com, costs you a couple hundred bucks and a few hours of soldering that will work with this.

I just tested it with the new checksum, and it works perfect now. So that's what the problem was. It reads out perfectly, but I'm still nervous about the stock program actually reading the "TPS2" input, and apparently doing work with it.

[ryan.hess]

Okay, figured out what was going on... The TPS2 input on the 7727 is actually on pin C18. It is used as an EGR position feedback. So in this case, it is being used. Now the question is, if you remove the EGR function, will it be okay to use this pin? Probably. This pin is the ideal input to use, because it's not filtered or pulled to +5 volts.

The second best would be the "MAP2" input, but this input is filtered over about 8mS by the hardware. This would be pin C21 on the 7727, or F14 on the 7730. Now 8mS is small enough that it won't be noticeable, so I think I'll switch the input to there.

[Fierobsessed]

maybe TPS2 isn't really TPS2, but a value derived from TPS used for comparison? I don't know, didn't look into it that far yet. I am kinda curious though, How does an input become a ram address? Is it assigned by hardware, or is it something in the program that assigns it to a ram location? Because if it is assigned by hardware, then all 7730's and all 7727's will have the same values in ram no matter the program. But if its assigned somewhere in software, then the address would be different for each program. any Ideas?

[Mars]

Cool, looks like you just about got her figured out then.

I can't see you having trouble communicating with the 7730 vs. 7727, I didn't change any wiring around when I ran the 8F chip in my 7730.... Should I have changed any?

Also, I have no EGR, and I don't really care for the fuel pump voltage in my logs, so if you have to go that route, I have no issues.

[ryan.hess]

maybe TPS2 isn't really TPS2, but a value derived from TPS used for comparison? I don't know, didn't look into it that far yet. I am kinda curious though, How does an input become a ram address? Is it assigned by hardware, or is it something in the program that assigns it to a ram location? Because if it is assigned by hardware, then all 7730's and all 7727's will have the same values in ram no matter the program. But if its assigned somewhere in software, then the address would be different for each program. any Ideas?

Read above, TPS2 is being used for the EGR position.

There's limited amount of RAM available, and most of it is allocated to crap that's being used in the code. I'm sure you could find a dozen things that have no use, and "borrow" their address, and delete any references to them.. But what I do is borrow the last byte from the stack. The stack is like a temporary RAM that gets used when you're doing calculations, or meaningless things that will get deleted in a few cycles. I don't recall how big it is at the moment, but lets say it's 30 bytes. At most, you might see 15 of those bytes being used by various processes at one time. It would probably be rare, but that's about the most it uses. Using the last byte guarantees free space.

So in short, it's assigned by the software (programmer). The location I picked was the last byte of the stack.

[ryan.hess]

Works... I'll send you an email with all the files you need...

[Mars]

Cool, I just X-fer'd everything to the laptop...

Thanks!

How did you guys go about adding a new wire to the existing connector?

Maybe I should goto the junk yard and pick up a connector, so I can use it to "re-pin" my connector with the one missing wire...?