ISO9141-2 K-Line Network bus communication diagnosis
Part 3 of a 6 part series. Go here to access the full set.
WHO USES IT:
ISO9141-2, also known as K-Line, is used by almost all the OEMs at some point in their network designs. It is primarily used by Asian and Euro brands from model years 1996 to 2006 but is also implemented by GM, FCA and Ford when the system is a joint development with their European engineering teams (think shared platforms or Euro design). Case in point: most pre-CAN Ford systems are J1850PWM but many of their airbag and ABS systems are wired for K-Line, as they were designed by a European supplier and/or Ford of Europe.
HOW IT IS NETWORKED:
ISO9141-2 is wired to DLC pin 7. It is a 12V DC digital signal, but will trick the unaware as the network rests at 12V DC rather than 0VDC. (Aftermarket accessory installers not knowing the electrical signature often tap into it, seeing 12V DC with their test light, and ground the circuit, shutting the bus down. VW/Audi published a TSB about aftermarket radio installs and the related OBD emission test failures).
The bus is active when the ignition is in the ON/RUN position BUT you have to have scan tool connected to the DLC and be requesting something (DTCs, PIDs, etc) to watch any communication activity. It may be active when the BCM is alive but don't count on that. The challenge with ISO9141-2 K-Line is the bus is not a true networked bus. ECUs do not communicate with each other and share data like the J1850 & CAN busses. In other words, they only speak when spoken to. You will NOT get UXXXX DTCs between ECUs as they do not share data. Each is a standalone, orphan, daisy-chained together on a single comm line.
Keep in mind the state change on K-Line. Its either 12V DC when at rest or 0V DC when transmitting. A scope is very helpful here. Set your scope to 5V DC/div and 1ms/div to get you going. Channel 1 lead goes to DLC pin 7. Ground channel goes to DLC pin 4 or 5. DLC breakout box is preferred so you don't compromise (read: spread) the DLC pins
Several OEMs also wire up L-Line to DLC pin 15. The electrical scheme is the same, and testing is identical. Same goes for KWP2000. (bonus material!)
Thanks for posting this series on communication!
Does the scan tool supply the 12V bias voltage or is a module like the BCM doing it? I just glanced at a Mercedes from the early 2000s and it states at ignition on pin 7 should have 12 volts. This isn't always strictly correct information though. I think when K-line is used for OBD2 diagnostics the engine control module must be connected to terminal 7, but other modules may or may not use that circuit for their own K-line diagnosis. In your example the ABS control unit is also on the ECM K-line.
An important note is that many Euro vehicles had their own diagnostic communication connectors with K-lines to each module prior to the ~1996 OBD2 standardization with the 16 pin DLC (SAEJ1979?). Mercedes used a 38 pin round connector from the early 90s to the early 2000s. Sometimes those connectors were kept even after the 16 pin DLC was added to a vehicle and you end up with a car that you can do some engine module scanning with but no communication with the body systems unless you connect to the 38 pin underhood connector.
An example is this 170 SLK made from 1998 to 2004. Terminal 7 K-line connection of the DLC is actually spliced into the existing 38 pin connector. N3/10 at the top is the engine control module. (See attached diagrams).
Going forward into the 2000s Mercedes got rid of the 38 pin connector but jammed the DLC with K-lines on the discretionary pins. There was a fuzzy migration from K-line to CAN based diagnosis. The 170 had a drivetrain CAN but no diagnosis over CAN. Then on later cars like the 203 C, there was CAN at 6 and 14 but also K-lines for many modules. The shifter for example has both a K-line to the DLC and a CAN connection to the DLC (via a gateway).
This gets important to understand when there are communication issues or variations between tools and the vehicle. Some tools will communicate to the K-line modules even if the CAN gateway is down. Others may try to get communication to the gateway to better identify the car and detect which modules there are, how they are wired, and how to talk to them. So in some cases when the CAN is down you may be able to scan the car for a quick test done via K-line, but unable to program anything. Bit of a mixed bag.
It's good to look at both the way modules communicate to each other and the way they communicate to the scan tool when diagnosing communication problems.
Good addendums. The bias voltage comes from any of the ECUs, not the scan tool. The tool grounds the line to initiate communication.
An important thing to know is most OEMs that used K-Line on DLC pin 7 also used something else. The main driver for K-Line on DLC pin 7 was so they could meet the US OBDII requirements, specifically CARB. A vehicle had to have one of the 4 specified protocols (pre 2008), so J1850VPW, J1850PWM, ISO9141-2 or ISO14230 had to be wired up on a specified pin. This was more of an annoyance to the Euro brands (and Honda/Acura) as they were already using CAN for intra-module communication.
What you ended up with was a Generic K-Line on DLC 7 that only was Emissions and then everything else wired elsewhere (20 pin, 38 pin, other J1962 pins etc.).
Honda as an example on many models wired scan tool to ECM and a gateway. The rest of the ECUs are B-CAN or F-CAN, with the cluster as the gateway. Here's an example where the scan tool talks to the MICU only for non-Generic communication.
I focused this on DLC pin 7 since the topic was diagnosis no comm with scan tool.