Interesting CAN BUS fault on 2018 Landrover discovery

RYM6746 · Post by **RYM6746** » Thu Mar 14, 2024 8:27 pm

Hi guys need a bit of help with this one. 2018 Landrover discovery sport. Customer has just purchased vehicle. Since owning it they have had this intermittent fault, it is fine when first started but after 10 -20 mins of driving it starts with a traction light and warning messages about the 4WD and EAC having a malfunction it ends with the cluster blanking out and then the engine will cut out and you basically loose everything at that point wont start and no cluster, its dead. full scan of the car just gives you a million can bus faults in every module effected. So the scope goes on through the DLC break out box (thinking I will find an intermittent short to ground) but it would appear looking at the data I have captured that the network voltages stay well with in their tolerances but my best guess after decoding them (I'm certainly not expert) is that one of the modules is sending out messages that are not valid and this becomes more and more frequent to the point where I believe all the other modules are pulling them selves off the network (best guess) it would appear that the ID of this message that is at fault is 0 E0 along with other ID's (pretty sure one module has multiple ID's correct me if I'm wrong) but it would appear that ID 0 E0 is common. I'm guessing the best plan of attack is to unplug each module off the network one by one and see if I can stop ID 0 E0 from being present (if fault is not present) or if fault is present see if the network returns to normal. If you see what i mean? Either way it will end up with a long road test with a module unplugged to confirm the fix (and then replace as necessary). I am asking if anyone has any knowledge on this type of fault or a diagnostic procedure I can follow to speed up the process and make it more accurate, Thank you for your help in advance I will upload what the forum lets me as the file is too big to upload all in one go. Thanks again

KimAndersen · Post by **KimAndersen** » Sat Mar 16, 2024 11:35 am

Hi

First let me say that I have no experience with Landrovers and therefore do not have an easy solution to your problem. However, I spent a lot of time decoding other canbus systems and since almost all canbus systems are built the same, it must also apply here.

However, I will start by quoting a very well known Youtuber Ivan from Pine Hollow Auto Diagnostics who says that 80% of troubleshooting on various systems is based on research. So the key to all troubleshooting is research and now on to your problem.

One thing I find strange about your Can bus communication at your DLC break out box is that there is active communication on Pin 6 (HIGH) and Pin14 (LOW). Normally there will be no can bus communication at this point unless you have the EOBD tester connected to the DLC box.

Since it is a relatively new car (2018) I would assume there is a gateway module before your DLC box. This module prevents direct access to the canbus. If you do the same on a Volkswagen or Audi, you just get the message ID 0X17F00010 from the gateway.

If you've been following IVAN from Pine Hollow Auto Diagnostics, you'll see that research involves a lot of things to look into before you start troubleshooting. But in my opinion the most important factor is understanding how the systems are built and work.

A good wiring diagram is really important when it comes to Canbus troubleshooting.

Hope you find a solution to this problem.

Regards
Kim

wiki38andy · Post by **wiki38andy** » Mon Mar 18, 2024 12:31 am

Engine should not cut out for a CAN buss issue. It is a mater of powers/ grounds rather. So, scope them at ECM when engine stalls.

RYM6746 · Post by **RYM6746** » Sat Mar 23, 2024 9:54 am

Hi guys thanks for your help but there is no gateway module between scan tool and can bus wires. Pins 6 and 14 are directly onto the bus. And the engine definitely does cut out due to the fault as I have since unplugged the diff module and other than a traction control light and a couple of warning messages there has been no further symptoms. The car has been given back to the customer to drive for a week or two to absolutely confirm that no further can bus fault/symptoms appear. Checking lives and grounds at engine ECU would be pointless as its extremely unlikely to cause that particular fault on a can bus system. We needed to eliminate the node from the bus that was creating long unreadable messages that was messing with the clock timing causing the nodes to stop transmitting. A power and ground issue would just stop 1 node from transmitting/receiving on the bus. This particular fault was effecting the whole bus including all nodes on it. Its definitely a rare fault

Rfmotors1 · Post by **Rfmotors1** » Sun Mar 24, 2024 2:00 pm

Hi RYM6746,
Just by looking on your waveforms, the good one has high load compared the problematic one.
I mean there is so many more frames sent on the good one, it is probably due to engine not running when you have captured the waveform of faulty vehicle.

Also by looking at the affected frames, it looks like the CM is unable to send the frame or another CM is talking to frames that are in progress, causing CAN collision.

This is probably due faulty CM or bad power supply to some CM which is constantly resetting, starting up and sending own frames directly to data stream, to other frames causing collisions.

See the table where the faulty frames are highlighted with the percentage of faults. (frame 0E0 is most frequent)

NOK Land Rover_Discovery_2018_Diesel_can for forum.xlsx: (17.83 KiB) Downloaded 15 times

So what to do with such info?
If we are looking for a faulty CM, then the best practice would be to set filter 0E0 to Picoscope decoding and keep the ignition ON or engine running, then unplug some CM and watch if the 0E0 disappears.
With this technique you will find the "owner" of this ID. (one CM has usually more ID)
Now, if the engine is the offender, then you need to monitor the power supply and grounds and see if they are OK at the time when the problems are present. When engine running, there are vibrations, current surges and other conditions that does not happen at ignition ON only.

Also, as wiki38andy wrote above, on most cars if you totally cut off the CAN wires form engine CM, the engine will not stall. May not accelerate but usually missing communication does not stall engine, it is basic safety feature, to keep engine running at any failure to support charging, braking etc. So it is good to keep this in mind.
From my experience, if you have corrupted CAN then engine stalls, one of the problematis ID is 0E0 then I would look at engine. The ID that are lower value have higher priority on front higher value ID, it means they will be sent firs in case two ID wants to transmit at the same time. Logically the priority is given to important CM and most important are Airbag, ABS and Engine/Gearbox, I just don't see why would engine stall if the ABS or Airbag are off the bus or something kills the bus.

So it is not a solution but maybe some info that can help you, I may be also totally wrong.
Anyway, please update us here with your findings and progress as your case is very interesting one.

Regards,
Roman

Rfmotors1 · Post by **Rfmotors1** » Mon Mar 25, 2024 8:50 am

Just small update, I have added the CAN ID in decimal numbers and sorted from small to high.

PicoScopes
& kits

For vehicle manufacturers

EV kits and
accessories

Noise, Vibration and Balancing

Milliohm tester

Pressure sensors

Current clamps

Breakout
leads/boxes

Accessories

Automotive Forum