I have recently reached what could be considered a milestone in the completion of the OBD2 LCD.  I have (after days of troubleshooting) succesfully initiated a connection with a vehicle and been able to send and recieve data.  I spent an immense amount of time troubleshooting this problem, only to find out i had installed a transistor backwards.  After this two minute fix, i plugged the interface into the test vehicle, and it initiated a connection with no problem.  I am still working on the data reception bugs, but by then end of the week i surely will have a video of the LCD outputting data from the engine!

It feels SO good to make progress... Finally!

UPDATE NOV. 12 - 2007 - 11:30am - ELM Communication is working!  It was as simple as sending the data inverted to the ELM.  The ELM323 is intended to be hooked directly to a PC's serial port.  serial data from a pc is sent inverted, and thus the ELM is set internally to deal with this data.  The MAX232 converts the data back to non-inverted, so i simply needed to r-invert data in teh PIC.

The board is SO close to actual in-car testing.  I am still having issues with the MAX233 interface chip.  I have spent hours trying to track down the gremlin in this chip, and i have come to the conclusion that the chip may just be bad.  i have a few of them laying around, so i will switch it this week before in-car testing.  The LCD is functioning.  This was an easy fix.  At one point, while trying to get the PIC working, i cut all the traces to the LCD.  Once i discovered this, i simply resoldered all the cut sections, and blamo, it works.  The initial code is certinly nothing fancy.  Just a few test items to test the LCD and get output from the ELM.  I have an e-mail out to elm electronics at the moment because i am having trouble getting coherent messages back from the ELM323.  The only message i can recieve that makes sense is the startup message, which is "ELM323 V2.0 >."  In theory, i should be able to send "ATI," followed by a carriage return, to the ELM and recieve its identification message back.  I recieve data back, and for any given command it is always the same data, but it just makes no sense.  Hopefully this is an easy fix.  I feel it may be something in the transmission to the ELM, although i have checked this wiring plenty of times, and even cut out the trace and attached it with a jumper wire.  I get the same results when using a direct connection to the PC via the max232.  I REALLY hope this is not a problem with the ELM chip.  This is a relatively expensive chip, and it has been sitting around for some time now.  We shall see what they have to see.

Test Code

INITIAL TESTING CODE (PICBASIC PRO) (NOTE: must save with extension .pbp for use in PICBASIC Pro)

If interested, here is the assembly language code generated after compilation, before compilation into HEX:

INITIAL TEST CODE ASM FILE (NOTE: File extension changed from .asm to .txt)

The initial test code is done in PICBASIC Pro.  It is very simple to w ork with, and is a language i am farmiliar with.  I am using this language for testing purposes only, as i know certain things work.  The intention is to write the actual software for the OBD2 LCD in C.  However, time constraints may force me to write the code in picbasic.  Also, the more low-level nature of PICBASIC PRO is more fun then the higher level C.  For more info on the PICBASIC Pro compiler, click here.

Click READ MORE for a short video of the LCD starting up, and me talking about it!

LCD BOARD VIDEO

The lcd board has grown by leaps and bounds in the past few days.  I currently have the PIC16F877A working and the LCD is all wired up.  I haev yet to fix the MAX233 interface, but am currently using a MAX232 interface on a different board for the time being.  I have succesfully communicated with the ELM323 chip directly from the PC, but am still running into some small issues communicating between the PIC and the ELM323 (also pic based, a 12c505).  I have included a video of board power on.  The ELM chip flashes 4 LEDs in sequence upon being powered on, as a "lamp test."  The LED's represent serial TX serial RX and OBD2TX and RX.  I can succesfully send commands from the PIC to the ELM, but am having trouble recieving uncorrupted data back.  This issue will be corrected very soon.

Click read more to view video!!

Update

Well, i've made some progress.  The main brain (pic16f877a) is working.  It is running my small light flashing program.  The problem turned out to be with the MAX233 circuit.  After disconnecting the +5v trace to the MAX233, the brain started working.  I have since setup the LCD circuit and attached the LCD with a computer IDE cable.  The backlight appears to be working, as well as contrast adjustment, but have yet to test the LCD with actual code.  By the end of this week, the LCD should be functioning, and hopefully i'll get some in car testing time!