ELM327 BT / Torque Android


Participant hyperactif
26 Juil. 2009
Score de réaction
P3 MC Lounge TOS
MaJ Torque
Avec un menu graph temps réel à tester et prise en charge du multiframe.

Right, there’s an update out which supports the multiframe CAN (which I posted to the market this morning). Variables start at A, go to Z, then start at AA, AB, AC, etc and the multiframe messages should now be handled
I’m a bit paranoid as I’ve had to make a change in the core OBD2 code to facilitate this, I’ve done 2 days of testing with all I can here and it seems to work ok, but let me know if anything stops working (if you could send debugging information that would be great) and I’ll see what is up and try to fix asap
guy, tu vas me faire passer à Torque, surtout si HSD Can monitor a du mal à avancer.
Peux-tu ouvrir un nouveau sujet et décrire toutes tes trouvailles ?
il y a un malentendu, je me mets Torque sous la dent bleu en attendant l'appli maison et si ça peut donner des idées à nos développeurs maison (encore que je me demande si c'est pas l'inverse).
Dernière édition:
De mon coté, je teste torque, mais je reprends cet été, à partir du 15 juillet, le dev sous android (avec B4A).

Le dongle BT permet, normalement, l'usage des 2 dev.

En ce qui concerne torque, je suis effectivement le premier à avoir demandé à Ian HAWKINS, lors d'échanges par mail, d'intégrer le multiframe; sa réponse date du 26/11/2010:
You want to look at the ‘extended PID’ part of Torque’s main app Settings page (it’s hidden away at the bottom)

That’ll let you enter your own PIDs into the app which then get passed through to the vehicle.

At the moment it only handles single frame responses – however it doesn’t rule out multiline in the future.

The main problem is that I don’t have a prius to test with J so app development would be slow to begin with until the basics were storted out.

I’ll have a look at the xls you linked at the weekend and *might* stick something into torque in the next couple of releases to handle multiframe responses /nicely/ if I can see a straightforward way to handle it in a user-interface kind of way (I assume this is all mostly using CAN protocols on the prius?)


qqs infos à propos de Torque:
il s'agit d'un échange de mails datant d'hier au soir:
bonjour IAN
I learned today about a new release of torque managing multiframe
just one question:
regarding A, B , C, AA BB & so on
where do you start for A,
on the second frame ,where do you start?

is there a difference for you regarding single & multiframe?
I will do a lot of tests, but I prefer to have a basic idea of the way you affect the bytes to variables...
best regards
ps: are you planning to developp calculated variables
for instance : Power(kW) = RPM * Torque(N.M) *k
et sa réponse:
Salut! :)

A is the start of the first frame, and basically treat it as if all the frames had been 'glued' together, so A, B, C, D, (and so on until you get to AA, AB, AC, AD, etc)

The app basically stitches together multiframe and single frames so they look the same (just a really long reply) to the equation parser

Funny you should mention calculated variables :) The equation parser can now reference these variable in the latest version released (though I have not documented it yet) basically for internally calculated variables, you would reference them inside square brackets, eg: [FF123A] (which is the number of GPS satellites locked) - I have not yet tested this function as I ran out of time, so it may not work! though if it does there is 'limited' documentation in: http://torque-bhp.com/wiki/PluginDocumentation (standard PIDs, remove the MODE from it, so it *should* just be [0D] for speed (etc) instead of [010D], but I must stress I haven't tested this part of the equation parser yet! :)

Hope this helps!


je n'ai rien compris à son affectation des variables,
il faudra donc que je créée une vingtaine de variables nommées
TA correspondant à A
TB correspondant à B etc....
et que je loggue leur valeur tout en regardant passer les trames,

@ suivre...
ps: j'ai créé cette discussion à partir des derniers messages de celle sur elm327 & android
Sais-tu si il fait la version payante (sachant que c'est acceptable), ou la version gratuite suffit ?

Il semble en lisant sa réponse que cela ne fonctionne que pour les variables internes de les utiliser dans les équations. Mais tu peux essayer avec une variable crée par toi pour voir.
il me semble qu'il faut la version payante, mais à 3.27€ :grin:.

j'ai procédé ce matin à un test qui m'a fait lui écrire ceci:
Bonjour Ian

there is a bug in the affectation of variables....

there is a glitch of one (missing) byte
I created tests vars from A to P named TA, tb ,... TP
and I observe that there is a missing byte and by the way, all the previous custom pids are wrong.

see attached xls documents
tracklog.xls is generated by torque;
colums VWX should each contain 64
prius_multiframe_example.xls is what we are waiting for , vars A, B & C are in col H, I, & J

another important point:
could you remove the 3 bytes limit in OBD2 Mode and PID (even if you just use internally only the first 3)
because in case of toyota hybrid, we can use "CAN STUFFING" like this : 21 61 62 67 68 74 8A

best regards

bon, il est plutôt réactif le Ian:

I will send you a log, but it will wait until this evening because sometimes I work for a living....
the car is at home, and I am in paris during the day...

OK, no worries, thanks for going to send the log.

I will look at it when I get the log to check what is going on with the predefined PIDs and the multiframe message(if you include that as well)


Ian (also at work ;-) - torque is a hobby)
et on continue:
Hi again

I'm sorry I couldn't wait for the debugs - releasing something broken has bugged me so I've taken some time off and fixed it - It was an oversight on testing, basically the multiline can responses are one byte longer and I hadn't noticed in testing, hence the off-by-one on the original PID handling. This should now be correct.

I've run some more testing on the multiline stuff, and it should be working for you - when you send some debugs I'll have a look to see what's going on

In the meantime, I've just posted a new version to the android market which fixes the mistake I allowed to get in - it may take up to an hour for a refresh (you can usually help things along by rebooting) for the phone to see the update

Regards, and thanks

Pages vues depuis le 20 Oct 2005: 307,755,389
Haut Bas