High Speed GMLAN Diagnosis
This is a demonstration of a High Speed GMLAN failure diagnosis. A 2020 GMC Terrain was used to identify the process and convenient access point to the network for testing.
The vehicle has a no crank, no start condition with many IPC telltales illuminated, exterior lamps on and more conditions indicating a major failure has occurred.
Connecting the MDI 2 and opening GDS 2 as shown in Figure 1, results in the VIN not being decoded, requiring manual vehicle and component/systems entries. This is a useful clue to keep in mind, when diagnosing network concerns.
With the vehicle built manually and a Vehicle DTC Check completed, we quickly discover that we have a network issue and GDS 2 has identified no communication with all modules on the HS GMLAN, as shown in Figure 2. When modules are communicating normally, the Status column is filled with green check marks.
Clearly, something is amiss, that has resulted in a total loss of HS GMLAN communication.
While LS GMLAN modules are communicating as shown in Figure 3, with the module list re-organized to show a mix of HS and LS GMLAN communication status, it appears that all expected LS GMLAN modules are communicating normally.
An understanding of the physical structure of vehicle networks and shared elements, eliminates DLC ground terminal 5 as being an issue, since both buses share terminal 5 as a common ground and the LS GMLAN is communicating with DTCs set in the modules that are listed.
Note: Not shown, but if viewing the Details it reveals that no HS GMLAN DTCs are set, simply because there are no HS modules communicating to flag the DTCs. The Radio Controls showing Lost Communication is a Terrain anomaly.
First, we should consider what type of network we are dealing with. This HS GMLAN may have a primary network connection direct through the SDGM to the DLC that can be tested using traditional DMM, DBDT and DSO methods. However, beyond the SDGM, there are also networks such as Gateway Isolated HS GMLAN and Gateway Isolated LS GMLAN, to name just two networks.
To diagnose the loss of communication will require gaining access to the affected network on the "vehicle side" of the SDGM. There is nothing different about the construction of these networks, other than the SDGM provides a "firewall" function that closes the "front door" path via the DLC. Testing at DLC terminals 6-14 with a Digital Multi-Meter would find a lower than specified resistance and Bus voltage at terminals 6-5 and 14-5 pulled low, as will be identified by the Data Bus Diagnostic Tool.
So, armed with a wiring schematic, finding an easy to access location to enter an isolated network without major disassembly or removal of fragile and expensive trim panels, will provide diagnostic opportunities for DMM or other desired test equipment.
Figure 4 is a typical representation of … GMC Terrain vehicle network topology.
In the case of the Terrain, let's consider the scenario where there is no communication with any modules on the primary HS GMLAN using at DMM at terminals 6 to 14 of the DLC. Since no high speed GM LAN modules are communicating, there are no relevant DTCs to follow.
Since the MDI 2 is already connected, switching to the Data Bus Diagnostic Tool (DBDT) is quick and will net some useful information as shown in Figure 5. The Measured Voltage tab identifies that Bus+ and Bus - are pulled low to 0.07V. This test is done prior to confirming with a DMM, simply because the MDI was still connected following the Vehicle DTC Check.
Utilizing the GM Data Bus Diagnostic Tool (DBDT) shows that no HS GMLAN modules are communicating, since the module field for the Detected State is blank. If the DBDT was connected and turned on before the failure, it would show the modules.
The message at the top of the DBDT screen in Figure 6 identifies that the Bus is pulled low to ground. The Measured Voltage tab also shows Bus + and Bus - pulled low.
With the MDI 2 disconnected, let's verify the measurements at the DLC using the Digital Multi-Meter. Bus resistance measurement with the failure present identifies resistance lower than 60Ω specification, consistent with the HS GM LAN being shorted to ground in Figure 7.
DMM Measurement at DLC 6-5 and 14-5, confirms that the bus voltage is pulled low as shown in Figure 8.
At this point, intrusive testing is necessary and reviewing the network wiring schematic, accessing the HS GMLAN can be made at the SDGM, which is effectively the point where all networks culminate.
On the Terrain, this is quite easily accomplished as shown, by removing the Electronic Transmission Range Shift (ETRS) selector bezel for access to remove the ETRS module. Use a plastic hook tool in the slots underneath the bezel or fingers to pull the bottom of the bezel out first and then work around carefully to remove the bezel as shown in Figure 9.
Remove 4 Torx T15 pan head screws from the ETRS module and remove it from the console as in Figure 10.
This will expose the Serial Data Gateway Module, which can be removed once the retention latch is depressed. Note: Connections have been removed for an unobstructed view of the module location and retention latch in Figure 11.
Once removed from the dash, the circuit measurements can be performed as per service information, to verify bus integrity or issues. If no issues are discovered, the SDGM requires replacement. Follow SI replacement instructions and Control Module References at the end of the repair procedure for programming and set up requirements.
So, the only real difference in diagnosing gateway isolated networks, as opposed to primary networks that are or were previously and conveniently accessible via the DLC, is that access on the vehicle side of the SDGM becomes necessary. A suitable access point into the network can be determined using the wiring schematic, that may allow testing without significant trim panel removal for access and the associated risk of damage of these often fragile pieces.
With the SDGM disconnected and circuits tested, the SDGM was found to be shorted internally. Replacement of the SDGM and programming selections as shown in Figure 12 restored normal operation.
With the vehicle systems back up and running and all functionality returned, verification of the affected bus circuits identifies that the network fault was resolved. A visual confirmation using the DBDT "Detected State" is okay and all HS GMLAN modules are connected as in Figure 13. Switching to the "Measured Voltage" tab shows normal graphing activity, with Bus + and Bus - voltages responding correctly as shown in Figure 14.
Using the DMM, terminals 6-5 Bus + voltage is normal with the SDGM replaced in Figure 15.
Terminals 14-5 Bus - voltage is also back to normal as in Figure 16.
Normal expected HS GMLAN ~60 Ω resistance is present at terminals 6-14 in Figure 17.
It is always important to perform a system verification following diagnosis and repairs, to ensure that the root cause of the concern has been correctly identified and fully addressed. No further diagnosis is required at this point.
Thanks for that Martin. Excellent as always.
Thanks Todd. I've had this sitting on my desktop unfinished for a while. The weather is back to rain here in BC for the long weekend, so getting this finished feels like I've at least done something useful today!
At least you don't have to shovel it.
Great write up as usual Martin. Thanks for that!
You're welcome Andrew. The objective is sharing to make this stuff a little easier for all travelling a similar path.
Hey Martin: Again THANK YOU!! VERY informative and in depth posting. VERY interesting and helpful. I am saving this one and your previous one to my "saved/archive" folder. Agreed. At least you do not have to shovel the rain. Enjoy your day. Take Care. Best Regards. Harley
You're welcome Harley. BTW, I finally got that other reply to you finished!
Thanks for a great write up Martin. Just curious was the SDGM bad or did you just short the bus to ground to simulate a bad SDGM?
Thanks Allan. This was a bad SDGM
Push button transmission?!?! LOL - Wasn't that tried in the 1950's? I can't wait to see how people bypass that in 20 years.
LOL Geoff. Should be right up your alley then! VBG. There must be Saratogas and other Dodge, Plymouth Chryslers still trundling around your neck of the woods withe button shifters! LOL Just wait until you run into one! If you get a chance, read up on "Latent" and "Critical" faults with ETRS. With Latent faults the operator gets up to 50 key cycles with warnings that must be recognized and then…
Is the ETRS an option on some vehicles or do all cars after a certain point have them? In areas where vehicle owners are conditioned to ignore all warning lights, this system failure will not go over well. As for cars from the 50's and 60's they were never here to begin with. Any classic muscle cars I have seen (even my GTO) were all shipped here more recently.
Hi Geoff: I assume that you're speaking of your island. Since I can't think of any Chrysler product from the late 50s-early 60s which didn't have them, I'd be very surprised they weren't at Pearl. Same goes with the Rambler Classics, likewise Packard. (We had an Imperial and a Classic throughout the years. We didn't have a Packard but, boy, were they a beast.) Guido
Yes, Guido, by "here" I meant where I am. :-)
Hi Geoff. Electronic Transmission Range Shift (ETRS) has been around for a few years now, with 4 different versions in GM products. The control in the Terrain was introduced at the model introduction in 2017. The shifter controls solenoids inside the transaxle and there isn't really any intrusive diagnosis. The other style of shifter looks like a traditional shift lever, but also has no physical…
"From recall, there are currently 4 versions of ETRS, with 2 being classed as "external" and 2 as "internal." " - Whichever proves more robust in the long run, will determine which used cars stay on the road, and which I eventually see. If I even bother looking at cars in 15 years...LOL
Hi Martin, You spoke some truth with what you wrote. I'm thinking we're in for some interesting times and challenges with the newer vehicles. I'm getting nervous about what I'm going to be able to fix in the future and I only work with one manufacturer. Some of what gets me nervous is I realize the value in what I do, and charge accordingly. I believe we're approaching a point of where car…
Hi Scott. I agree. Not only the vehicle owners, but many service providers really are in for a ruder awakening than even Jim Wilson predicted many years ago! I am quite happy keeping my '03 Cavalier Z24 commuter on the road until retirement along with my wife's '15 Terrain. Our backup vehicles are our '99 Venture and my '71 GMC 2500. Fortunately, I have fix most anything on these vehicles with…