Software On ELM Street - OBD2 Software Development

[TazzI]

Well havent had much time to do much more on this. But.. Iv gone ELM crazy and now have 5elm cables to hook up at once to do different tasks.

Currently I only have 2 cables implemented. Cable 1 is dedicated to Monitoring traffic. And Cable 2 is dedicated to answering requests. Iv tried this new config on the tech2 to see if I cant get the DPID section going... but.. seems the tech2 is still unhappy and is looking for something else!

So two options left. Add yet another elm cable to the situation and dedicate that to spamming back the DPID requests leaving the two other cables free to continue their tasks.

Or second option being grabbing an LS1 pcm to mess with. But ebays current prices of 300+ isnt eaxctly what I was after. I mean common! Most of these buggers are tuned locked at some point or really are useless without the correct knowledge.. shouldnt be so expensive with absolute no support for fitting/reprogramming!

[TazzI]

Woooohoooooooo!

Got it working with three elms. Really I need four as the elm cant "spam" the requests as quick as Id like but still working well at the moment in the "data display" screen in the tech2.

Least I know with enough elms attached.. Will be able to reverse anything the tech2 throws at me.

Tazzi 1 : Tech2 0

Heres the current setup.. excuse the mass of wires!

[TazzI]

Figured id update on the apps current status.

Sooo, VPW is pretty much 100% sorted. So all LS1 vehicles can have engine data, fault codes, VIN, OSID ect read out without any hickups. There is also a Seed/Key finder int here for some added fun
As for speed, requesting holden specific PIDs gets up to 35-55PID's... depending on how much is going on the main user interface (can speed up with some more threading), But its pretty damn quick!
EVen got some custom controls programmed up...(excuse the messiness)

appcontrols.PNG

Am currently working on the CAN protocol.. more specifically, 11bit 500kb/s. This is the protocol that the VZ V6 ecus utilize, so I will be attempting to listen in on the Tech2 and try figure out some of the GM specific pid requests, faults, vin ect.
If the ELM is too slow.. I imagine those CAN shields will come in handy very soon
(yes that does say 200,000+ logs.. I forgot I had it running )

logcan11bit.PNG

[TazzI]

Has anyone had any experience or luck with sending multiline responses?

I can receive them just fine... but it seems sending on the elm is.. not possible?

[TazzI]

Have contacted pretty much anyone and everyone that knows something about elm327 development.. sadly.. its simply not possible to do a multiline frame write from the elm.

Sooooo.. a good old can shield implementation is a must to be able to do anything interesting on the elm.

Also looking through some CAN documentation.. I should be able to request pids using mode 2C.. but it simply doesnt work on CAN.. it could just be the e55 ecu thats not a happy camper as I dont have any other ecu to test on, but either way, that limits the ELM327 to only being able to request the standard 15 PIDs.. nothing more!.

I also checked through some other available software... they dont seem to do anything fancy at all... in fact.. they set the cable up wrong for the VZ, get a whole heap of CAN errors. Only a couple software's work correctly for these ecu's.

[jezzab]

I can send multiline CAN frames and that was suppose to be impossible. There is a trick to it

[TazzI]

I can send multiline CAN frames and that was suppose to be impossible. There is a trick to it

hmmm.....

Iv contacted pretty much all developers/forums with the same "cant be done" answer. And neither is there any examples on the net! *sigh*

So for example, this sets up revs ect on the DPID "FE".. this is exactly what is received from the tech2
7E0 10 0C 2C FE 00 0C 11 C9
7E8 30 20 02 2C 7F 7F 7F 7F
7E0 21 00 05 00 0F 00 10 00

Now..setting the header to 7E0 (obviously). Then simply sending off "10 0C 2C FE 00 0C 11 C9" gets back a "?" from the elm.
So, I then tried 08 10 0C 2C FE 00 0C 11 C9 and that also failed.

As well as 01 10 0C 2C FE 00 0C 11 C9.

Failed all that, Iv attempted to set the "10" as the extended address, and then send off 0C 2C FE 00 0C 11 C9 with no luck either.
Iv modified the timeout time to ensure Im not missing anything and have another elm constantly logging at the same time to check for results.

Seems that the 08 10 0C 2C FE 00 0C 11 C9 sort of did something.. caused the ECU to completely stop communicating. haha

[TazzI]

Taken from scantool.net... an interesting bug..

--- Code: --->atz


ELM327 v1.3a

>atsp6
OK

>atcaf0
OK

>1122334455667788
?

>020100
06 41 00 BE 1B 30 13 00
06 41 00 88 18 00 10 00
06 41 00 00 08 00 10 00

>1122334455667788
NO DATA
--- End code ---

When ATCAF0 is initially sent, sending 1122334455667788 doesnt work.. a '?' is received. But.. after getting the initial '?' out of the way, it seems to work afterwards?

...Will have to test this out...

[TazzI]

And we have lift off! It works... ohhhh it workkkkkssssssssssss!

[silversv6]

Well done Tazzi. Do you sleep?