In the last issue of GEARS, we examined how to diagnose a single module that wouldn’t communicate. We covered the three requirements for a module to operate: power, ground, and its ability to communicate.
This month, we’ll examine a vehicle that has problems with multiple modules. The diagnostic approach will be slightly different, because the vehicle’s fault is different. As always, a networking issue should start with a scan tool.
The vehicle is a 2007 Silverado that won’t crank or start intermittently. When the vehicle is acting up, there are additional problems: the power windows and the HVAC won’t work, to name just a few.
As with most modern vehicles, the first diagnostic step is to connect a scan tool and check for any trouble codes. In this case, many of the modules won’t respond. That indicates a network issue, forcing us to shift gears, so to speak.
When multiple modules won’t communicate, your first step should be to try to communicate with all of the modules, and then use wiring diagrams to choose which area to focus on. On this particular vehicle, communication is possible with the Engine Control Module (ECM), Body Control Module (BCM), Transmission Control Module (TCM) and the Electronic Brake Control Module (EBCM.)
These modules won’t respond: Radio, Passenger Door Switch (PDS), Instrument Panel Cluster (IPC), Driver Door Switch (DDS), HVAC Control Module (HVAC), Theft Deterrent Control Module (TDM), Body Control Module (BCM), Memory Seat Module (MSM), Sensing and Diagnostic Module (SDM), and Remote Control Door Lock Receiver (RCDLR.)
Armed with that knowledge, the next step is to consult a wiring diagram and examine the system. The object is to step back and get a feel of how the vehicle’s networks function.
This Silverado has two networks: high speed CAN and low speed CAN. Based on which modules respond, and those that don’t, the wiring diagram shows that all of the modules on the high speed CAN bus are communicating; the modules on the low speed CAN bus aren’t.
The next step is to break out a scope and observe communication signals. As daunting as this may sound, remember that you don’t need to decipher what the modules are saying; you just need to confirm they’re saying something.
We’ll start with the high speed CAN bus to see what a working signal looks like (figure 1). The red trace is CAN Hi probed at DLC pin 6; the green trace is CAN Lo probed at DLC pin 14.
Here’s what you’re looking for:
- The signal’s turning on and off properly.
- The voltages are good.
With most two-wire CAN networks, CAN Hi should rest at 2.5 volts and rise to 3.5 volts when communicating. Conversely, CAN Lo should rest at 2.5 volts and drop to 1.5 volts when communicating. The two traces should mirror each other. As you can see, that’s exactly what they’re doing.
The next step is to check the low speed CAN network to see if it’s trying to communicate. To do that, we’ll connect the scope positive lead to DLC pin 1.
Obviously, something’s trying to talk (figure 2). We have what appears to be communication, but look at the voltages: This vehicle’s low speed CAN network is starting below zero volts and rising to about 0.7 volts to communicate.
Does that seem right? Hard to say for sure: First we need to know what a good GM low speed CAN should be before we can make the call. Knowing what good voltages are for a particular communication protocol is critical for diagnosis.
For reference, a GM low speed CAN should rest at 0 volts and rise to about 4.5 volts.
This vehicle’s signal is nowhere close.
So the low speed CAN network has a fault. Now we should start asking questions:
- Is a wire cut? No. If a network wire were cut, we’d lose some modules but the voltages would still be good for the remaining network.
- Is there a short to ground? No. If there were a network wire shorted to ground, the scope would read zero volts all the time.
- Is there a short to power? No. If there were, the scope would show system voltage.
- Could we have a module corrupting network communication? Yeah, that could give us this problem.
It’s common for one module to affect the communication of all modules on its respective network. So how can we isolate the problem module when there are ten modules on the network?
One way is to disconnect modules, one at a time, until communication is restored. This technique may be our only option on some vehicles, but it can be time consuming, depending on the locations of the modules.
In other cases it’s easier to isolate. In all cases, wiring diagrams and network topology are the key. We’ve simplified the Silverado low speed CAN network diagram to make it easier to follow (figure 3).
As you can see, there are several modules on the faulty network. We could choose to find and unplug each module, one by one. But a better option may be to divide and conquer: JX221 and JX339 are splices. If we disconnect JX339, we could split the network.
If communication returns with splice JX339 disconnected, the problem is with the Memory Seat Module (MSM), Sensing and Diagnostic Module (SDM), or the Remote Control Door Lock Receiver (RCDLR.) If communication doesn’t return, we can focus on the upper half of the diagram. So the next step is to find splice JX339.
The JX339 splice pack is located under the passenger’s seat (figure 4). To isolate the circuits, we remove the connector, or comb. The comb is basically a jumper that connects the networks. Removing the comb isolates every module that connects through it.
So we remove the comb and, lo and behold, communication returns to all the modules on the top half of the diagram (figure 3). By process of elimination, we can conclude that the problem lies in the Memory Seat Module (MSM), Sensing and Diagnostic Module (SDM), or the Remote Control Door Lock Receiver (RCDLR.) Where do we go next?
One option would be to find and disconnect each of these modules independently, which is a perfectly acceptable technique. But, for the sake of efficiency, we could approach this issue backward: We could use a jumper wire at JX339 to add one module at a time, and see which module brings the network down.
Using the jumper wire technique, we connect one module at a time, using the scan tool to check for communication. When we reconnect the Sensing and Diagnostic Module (SDM) to the network, all modules on the network go down.
So now we need to find the Sensing and Diagnostic Module (SDM), or the air bag module, and disconnect it. Service information shows the module is underneath the driver’s seat. We pull back the carpeting and there it is (figure 5). Apparently, road salt collected under the carpet.
As a final check, we disconnect the Sensing and Diagnostic Module and reconnect everything else. The vehicle starts and runs, and all of the other modules communicate with the scan tool.
Checking the communication signal from the low speed CAN bus with the Sensing and Diagnostic Module disconnected (figure 6) shows the communication circuit is functioning correctly. It rests at about 0 volts and rises to about 4.5 volts to communicate.
In addition, the HVAC works and the power windows come back on line. Everything looks good. The Sensing and Diagnostic Module (SDM) in this Silverado was causing all of the customer complaints and more. Replacing and programming the SDM resolved all of the issues.
To diagnose the communication problems on this vehicle, we followed a structured plan: First we attempted communication with all of the modules. Next, we consulted network wiring diagrams to narrow the search.
Then we checked for communication with a scope, and confirmed it with a scan tool, while isolating portions of the network until the process of elimination led to the faulty module. Finally, disconnecting the faulty module and restoring the rest of the system confirmed the diagnosis.
You can follow these same steps with most multiple module communication problems. If you stick to this procedure, you’ll greatly enhance your success rate and decrease your diagnostic time.
