Motor Age Garage: Broken door locks an issue on Dodge Caravan

Lot cars also tend to kill their batteries if the salespeople don't remove the jumper or fuse that feeds memory and lighting circuits. (Chrysler calls that an I.O.D fuse for Ignition Off Draw.) And let's be honest, a salesperson usually doesn't want to be bothered with a trip to the underhood PDC every time he or she leaves and comes back from a test drive.

I once worked for several hours on what at the time was a new 1995 Ford Windstar that was blowing door lock fuses because a salesman had clandestinely ordered and plugged in a door lock button (the one in the back by the lift gate) on a Remote Keyless Entry (RKE)-equipped van that was never supposed to have a switch plugged into that connector. The problem was that the switch and the wiring looked factory, but adding it caused a dead short to ground when the front lock switches were activated.

Deadlocked on the Driver Switch

This Grand Caravan was a lot car with in intermittent door lock problem that had been sitting out there for a couple of weeks. Adam, one of my graduates, is the resident tech at a local used car dealership, and I found him dealing with this concern on a Friday afternoon. Two heads are better than one, so we attacked it in tandem and with gusto.

First and foremost, it's important to understand power door locks in general, and most techs with any time in the field have a pretty firm grasp on the basic principles. First, the switches on older power door lock systems simply apply current to the door lock actuators like power window switches, except the door lock actuator doesn't move as far or for as long.

This Grand Caravan, like most newer vehicles, has inserted a solid state middleman between the switches and the locks, i.e., the Body Computer. The Body Computer has its colorful copper tentacles in a lot of pies, and among other things, it locks the doors above 15 mph and has some neat little customer programmable features, like the option to flash the lights and/or toot the horn when the locks are activated with the RKE remote.

On this model, there's only one wire running from the Body Computer to each door lock switch; the other door lock switch wire is hard-wired to ground. Inside each switch we find a two-position switch and three resistors. It's a stepped arrangement that works this way:

1. With the switch disconnected, the voltage on the line leading to the switch is 5 volts.

2. With the switch connected and current flowing through all three resistors to ground, the voltage drops to about 4 volts.

3. With the switch in the Lock position, the Body Computer is reading voltage drop across only one resistor, thus the voltage is 1.5 volts, and the Body Computer drives the door lock actuator to unlock the doors (no relay involved).

4. With the switch in the Unlock position, the Body Computer is measuring voltage drop across two of the three resistors for a reading of 2.75 volts and the BCM responds accordingly.

There are other factors to consider besides the voltages, and it's always good to read the description and operation of the system, which can clear up a lot of confusion. For example, on this particular system, part of the description and operation section reads this way:

With that little blurb in mind, any and all troubleshooting must be done without the key in the ignition and with the possibility in mind that the customer's concern might simply be a misunderstanding of the way the system works. For a brief digression, that strategy isn't new. Early 1990s Jeep Grand Cherokees had an output from the key warning indicator that only provided a ground to the driver door lock relay when the key was removed, and Ford introduced a similar strategy on the Taurus in 1996.

It really helps to have a scan tool that will talk to the Body Computer, but Adam's scan tool would talk to everything on this van except the very box he needed to see into. That's always a revolting development and calls for additional work and research.

Old Fashioned Diagnostics

Position, temperature and pressure sensors have had a current-limited 5-volt feed for a very long time, and before Fords had datastream, I'd reach for my voltmeter in a heartbeat when there was some kind of problem I needed to figure out. Knowing good numbers and comparing them to those on the vehicle in question seems to work best for me, even when I have a scan tool in hand.

Our locks were operational from the passenger door switch and the fob, but were as dead as a hammer from the driver side. Checking the voltages on the passenger side door lock switch revealed a nice healthy 5 volts with the switch unplugged, 4 volts with the switch plugged in and at rest, and normal voltages in Lock, for this unit in the lock (1.5 volts) and unlock (2.75 volts) positions.

The inoperative driver switch, however, had no voltage at all, not even with the switch unplugged. Did that mean the wire was shorted to ground? Without a scan tool to even check codes, it wouldn't have been evident, but I personally checked the wire for a short to ground with a logic probe AND the digital multimeter.

Without the benefit of a suitable scan tool, we obviously were at a disadvantage — in many cases, not knowing the state of a particular input can be a deal-breaker. How could we be sure the BCM wasn't getting a false key-in-ignition input? Well, there was nothing dinging in the way of a key alarm, and if that input had been the problem, the passenger door lock switch would have been inoperative, too.

We measured no short, so it was time to check the source of that wire, which, as it turned out, was occupying one of 34 pins in the Body Computer's C3 connector. Six of the 34 cavities were vacant, but that still left 28 wires. Pin 21 was our cavity in the C3 connector shell, and we found a good 5 volts at the BCM. Rechecking at the switch connector we found a big fat zero. It wasn't a short, it was an open — the voltage was getting lost somewhere in between. So what was going on between C3 and the driver side door lock switch?

Well, there was a connector — C300 to be exact (just above the driver door lower hinge) — but the legend describing the pins didn't mention anything about the door locks. The schematic, however, pinpointed cavity 5 as the channel for our circuit on its way through C300, and the wire color actually matched. Interestingly enough, there was no voltage at pin 5; we had an open circuit between pin 21 on C3 and pin 5 on C300.

Conclusions

We had reached the point on this diagnostic trail where we could categorically say we had found the problem. So what was the best solution? The concern was intermittent enough to where it would go for days at a time without manifesting itself. We had, quite interestingly, caught the problem at its worst.

When we reconnected all the wires, the locks began to work properly. And while it was plain to note than an overlay of the circuit between C3 pin 21 and C300 pin 5 was in order, it was difficult to convince the sales manager that such a repair was necessary, especially because the locks were now working perfectly.

There was an outside chance that the problem was at a connector crimp, and in that case we might have disturbed it enough to let that whisper of voltage make its way past the problem spot back to the switch.

The van was reassembled and parked back on the lot. Sooner or later, the overlay would need to be done. But for the present, the sales manager was satisfied, and because the vehicle still belonged to the dealership, that was all that mattered. Having finished the troubleshooting process with Adam, I realized that just knowing what the problem was gave Adam peace of mind.

Richard McCuistian is an ASE-certified Master Auto Technician and was a professional mechanic for more than 25 years. He is an auto mechanics instructor at LBW Community College/MacArthur Campus in Opp, Ala. E-mail him at [email protected].