Why are the easy answers the hardest? Using the OEM fuse box and stock harness for a G3 swap

[Map63Vette]

I'm not sure it is a proprietary architecture. The bus is sometimes referenced to as ISO 11898-3 compliant, and if anything is just older. But not really sure. I am pretty sure it is speed limited and is close to topped out while the more current stuff has a much higher speed but can be slowed down to 125kbps easily.

I've seen discussions that the CAN-B/CAN-IHS is used for comfort stuff, like heat/AC, while the CAN-C is used for critical stuff like PCM stuff. I could see the argument that headlights are critical, but seems they left that on the IHS bus.

I did tried and send the messages for the headlights on the CAN-C bus just to see if it would work but didn't seem to have any changes. But I don't have confirmation that the messages are even right yet. The ID's I have are for a 2008 Durango, so there is a good chance I will still have to try and figure out how to sniff the correct ones.

http://[URL]https://github.com/BiggRanger/CANBUS-Vehicle-Reverse-Engineering

[/URL]

Ah, okay. The ISO 11898 sounds a little familiar from some of the bus stuff I was looking at before. I got into it with my Viper because the TPMS is the one thing that you have to have a factory scan tool for, and not just a factory scan tool, but the real deal handheld DRB III one. They later released an emulator that ran on the newer scan tools (the WiTech VCI pods / Micropods), but evidently the one module those couldn't talk to for some reason was the TPMS, so I started trying to find out why and learned how many different busses can go to the diagnostic connector. I want to say the 11898 stuff was maybe more favored by GM, but I could be wrong. Don't know if you might have any better luck finding stuff looking in GM places. I thought some of the nicer OBD II interfaces like the OBDLink stuff might support it, but I could be wrong.

 

[DionR]

I thought some of the nicer OBD II interfaces like the OBDLink stuff might support it, but I could be wrong.

I do have an OBDLink MX module and AlfaOBD will connect to the TIPM and I can do tests where it turns on the headlights and such. And that all works. It is wired into the OBD port like stock, which only has the CAN-D bus wired into it. I tried to watch the CAN-D network and see what messages the AlfaOBD was sending but came up blank. Not sure why.

I have not tried to wire the OBD port to the CAN-B network and see if the MX module can still connect, so no idea if it has support for the FT bus.

 

[Map63Vette]

I do have an OBDLink MX module and AlfaOBD will connect to the TIPM and I can do tests where it turns on the headlights and such. And that all works. It is wired into the OBD port like stock, which only has the CAN-D bus wired into it. I tried to watch the CAN-D network and see what messages the AlfaOBD was sending but came up blank. Not sure why.

I have not tried to wire the OBD port to the CAN-B network and see if the MX module can still connect, so no idea if it has support for the FT bus.

Interesting. Have you got any other modules on the network at the same time? I wonder if something is acting like a gateway and taking the message in on one bus and relaying it over another.

 

[DionR]

Interesting. Have you got any other modules on the network at the same time? I wonder if something is acting like a gateway and taking the message in on one bus and relaying it over another.

Yes, the TIPM is the gateway between all the buses.

I currently have the PCM, TIPM, WIN, IP, SCCM and headlight switch all wired together along with an OBD2 plug. The PCM, TIPM and WIN are on the CAN-C bus and the OBD2 plug is on the CAN-D bus from when I was sniffing the start/run and start messages. I just left them alone when added the IP, steering column control module and headlight switch to try and sniff the headlight on/high messages. The IP, SCCM and TIPM are all connected through the CAN-B/IHS bus while the headlight switch is wired direct to the IP.

 

[Map63Vette]

Did you have everything hooked up when you did the ODBLink? Might try unplugging modules one at a time while sending the headlight message to see if one of those is the culprit.

 

[DionR]

Did you have everything hooked up when you did the ODBLink? Might try unplugging modules one at a time while sending the headlight message to see if one of those is the culprit.

Not following. Culprit for what?

 

[Map63Vette]

My thought process was that if the OBDLink can turn the headlights on and off while only being connected to CAN D, was it actually routing that message through some other module that's "translating" the message to the other bus. So hook up everything and connect with AlfaOBD and send the headlight message, then start disconnecting everything but the TIPM one module at a time to see if one of them is acting as the "translator", if that makes any sense.

 

[DionR]

Long ways from getting this working, but if I can get the headlights to work, I think I might run most of the exterior lights off the TIPM.

The TIPM has the following exterior light outputs:

Left low headlight
Right low headlight
Left high headlight
Right high headlight
Left front turn signal
Right front turn signal
Left rear turn signal
Right rear turn signal
Park lights

The headlights are easy. Just four wires, two to each headlight and delete the 2 coming through the bulkhead connector. Run the wires from the dimmer switch to the IPR (Instrument Panel Replicator) to tell it the head lights are on and high or low.

Tie the park light output to the park light circuit coming out of the bulkhead connector and it should light up the whole car. Remove the connection at the headlight switch as it isn't needed anymore and use that pin on the switch to tell the IPR that the park lights should be on.

Left front and right front turn signal outputs get tied into the corresponding front turn signal circuits which are separated from the turn signal circuits coming through the firewall.

Tie the left rear and right rear turn sign outputs to the corresponding turn signal wires coming from the bulkhead connector. Note that these outputs are also the brake lights and it appears that that brake signal might be a power wire rather than a CAN signal, so that potentially dodges the complexity of putting it on the bus. The stop and turn signal circuits are actually separate under the dash, so it will take some changes to get the wires at the firewall connected to the tail light wires running to the back of the car. Best idea I can come up with there is to pull the wires from the inside of the bulkhead connector and plug in wires that run direct from there to the taillight harness plug. This will isolate the turn signal circuits so the indicators function as before rather than both being on when the brake lights are on.

The turn signal flasher should be removed entirely and the two wires tied together as the turn signal switch still needs power. The brake light output should be disconnected from the turn signal switch and run to the TIPM, which means adding that wire to the bulkhead connector. The hazard flasher should also be removed.

So the IPR would have 2 headlight wires, a park light signal and 2 turn signal wire inputs and then a CAN-IHS output of some type. Plus key on power wire, a constant power wire and a ground.

The hazard flasher setup is a ground on a wire coming from the TIPM to turn them on. Should be simple, but it will add a wire to the bulkhead connector. That wire could run to the hazard flasher switch and connect to the hazard flasher output that was removed. Then connect the output from the switch that used to have power when the hazard flasher switch was pushed to a ground.

Advantages:

No headlight wires through the firewall.
No brake light power passing though the turn signal switch (or plug).
Turn signal switch only sees low amp power to indicate turn signals.
Turn signal flasher no longer required.
Hazard flasher no longer required.
Headlight switch doesn't see parking light or headlight amps anymore.
Dimmer switch sees only low amp signal power.
Lane change feature now available.

There is also a backup light output, but that one feels overly tedious to add. But time will tell.

And either way, I have to be able to make the CAN-IHS bus work before any of this is possible.

If I do make it work, the bulkhead connector would lose the two headlight wires but would gain the brake and hazard signal wires so that would be a wash. And there are still the 6 potential CAN bus wires to route somewhere, or 4 if I wire the OBD2 plug into the CAN-C bus and ignore the CAN-D one. Plus 3 for the CC wires, but minus one for the starter signal wire I won't need. So a net gain of 6-8 wires to pass through the firewall. Plus a couple of power wires direct from the battery I will need to figure out as well.

 

[DionR]

My thought process was that if the OBDLink can turn the headlights on and off while only being connected to CAN D, was it actually routing that message through some other module that's "translating" the message to the other bus. So hook up everything and connect with AlfaOBD and send the headlight message, then start disconnecting everything but the TIPM one module at a time to see if one of them is acting as the "translator", if that makes any sense.

Ah, gotcha. Pretty sure the TIPM is the only one doing any translating since it is the one turn on the headlights in the test. I will probably disconnect the TIPM from the CAN-IHS bus and see if the tests still work and if not go through those steps and see what happens.

If it does work with the TIPM disconnected and AlfaOBD able to turn on the headlights, give me hope that I can find a way to get it working.