First, the good stuff...
Center seat connector resistances [brackets include tested ranges]:
The longer story...
This has been beat to death on this forum, but as far as I can tell, there are no reported successes without purchasing another seatbelt buckle assembly ($280 MOPAR), or in rare cases, the "ebay dongle". Took me awhile to notice that. With countless hours dumped into this, I was at my wits end about to cave and purchase the MOPAR part or maybe hit up a junkyard for late-model FCA belt buckles.
The primary gotchya is that the seat belt switch is a Hall Effect sensor. *** Chrysler/FCA.
The Occupational Restraint Controller (OCR) applies a voltage to it every 100ms and reads the output current. If it is outside the expected range, your SRS light pops on after 2.5 seconds.
Notably, when you measure the switch resistance with a multimeter, you will find an open circuit whether the belt is buckled or unbuckled.
This is because the sensor requires a voltage across its leads to operate.
Couldn't find any spec on the Promaster Hall-Effect sensor but I found a datasheet for another hall-effect seat belt switch (attached).
Wondering if the switch could be spoofed with a resistor for a particular state, I used Ohms Law to calculate a substitution resistance. I assumed 12V "operating voltage" and 5mA output current:
12V/0.005A = 2400 Ohm
I was ready to test with a bunch of resistances and had set up a potentiometer to do so (pic attached). Miraculously, the first attempt nailed it.
Cycled ignition 3 times and the SRS light was off for the first time (10,000miles, 6 months) since removing the bench seat . Did not have to start engine. Dunno if I actually had to cycle ignition.
EDIT: There are some reports of needing to start the vehicle for the change to take effect.
I did not unplug the battery for this. I used to, but came to the conclusion that the only thing we need to avoid is applying a voltage to an existing squib. Latent charge in the OCR could possibly do that so I would still recommend disconnection of battery before unplugging connectors from any seat. We are working only with resistors.
The occupant detection thing (white plug) measures 400-Ohm empty, begins registering the seat belt warning around 350 Ohm and goes down to 20 Ohm with a heavy person in the seat. I went with 390 Ohm as I did not want to risk a fault by going over 400 Ohm.
I have no code reader that can read SRS codes.
Other notes pertaining to misconceptions seen elsewhere in the forum:
Big thanks to all the others who paved the way in other forum posts.
Good luck, hope this helps!
Edit: @smdub reverse engineered the whole system, worth taking a look at his thread if on the same topic you have trouble.
Center seat connector resistances [brackets include tested ranges]:
- Occupancy Detection (white plug): 390 Ohm [360-400]
- Seat Belt Pretensioner (half of yellow plug): 2.5 Ohm [2.3 - 3.2]
- Seat Belt Buckle (half of yellow plug): 2300 Ohm [1900-4500] ***
***Edit: newer model PMs (~2017+?) seem to do better with a ~460 Ohm resistor in place of the 2200 Ohm I used for the Seat Belt Buckle
***Edit 2: it seems the wiring for the 2.5Ohm and 2200 Ohm pairs is sometimes swapped from what is shown.
***Edit 3: AlfaOBD has a BCM configuration option that appears to allow reconfiguring the seating layout to not include the center seat, potentially making this thread obsolete.
The longer story...
This has been beat to death on this forum, but as far as I can tell, there are no reported successes without purchasing another seatbelt buckle assembly ($280 MOPAR), or in rare cases, the "ebay dongle". Took me awhile to notice that. With countless hours dumped into this, I was at my wits end about to cave and purchase the MOPAR part or maybe hit up a junkyard for late-model FCA belt buckles.
The primary gotchya is that the seat belt switch is a Hall Effect sensor. *** Chrysler/FCA.
The Occupational Restraint Controller (OCR) applies a voltage to it every 100ms and reads the output current. If it is outside the expected range, your SRS light pops on after 2.5 seconds.
Notably, when you measure the switch resistance with a multimeter, you will find an open circuit whether the belt is buckled or unbuckled.
This is because the sensor requires a voltage across its leads to operate.
Couldn't find any spec on the Promaster Hall-Effect sensor but I found a datasheet for another hall-effect seat belt switch (attached).
Wondering if the switch could be spoofed with a resistor for a particular state, I used Ohms Law to calculate a substitution resistance. I assumed 12V "operating voltage" and 5mA output current:
12V/0.005A = 2400 Ohm
I was ready to test with a bunch of resistances and had set up a potentiometer to do so (pic attached). Miraculously, the first attempt nailed it.
Cycled ignition 3 times and the SRS light was off for the first time (10,000miles, 6 months) since removing the bench seat . Did not have to start engine. Dunno if I actually had to cycle ignition.
EDIT: There are some reports of needing to start the vehicle for the change to take effect.
I did not unplug the battery for this. I used to, but came to the conclusion that the only thing we need to avoid is applying a voltage to an existing squib. Latent charge in the OCR could possibly do that so I would still recommend disconnection of battery before unplugging connectors from any seat. We are working only with resistors.
The occupant detection thing (white plug) measures 400-Ohm empty, begins registering the seat belt warning around 350 Ohm and goes down to 20 Ohm with a heavy person in the seat. I went with 390 Ohm as I did not want to risk a fault by going over 400 Ohm.
I have no code reader that can read SRS codes.
Other notes pertaining to misconceptions seen elsewhere in the forum:
- The OCR system is "floating" as to prevent accidental discharge, so the ground is isolated from the chassis ground.
- The short circuiting shunt is only applied on the seat-side of the yellow connector when it is disconnected. Not our problem here.
- The Airbag system is almost certainly entirely disabled if the SRS light is on. If OCR is getting funky measurements, it does not want to risk accidental deployment.
Big thanks to all the others who paved the way in other forum posts.
Good luck, hope this helps!
Edit: @smdub reverse engineered the whole system, worth taking a look at his thread if on the same topic you have trouble.