September 2021

36 AFTERMARKET SEPTEMBER 2021 TECHNICAL www.aftermarketonline.net Relevant live data Back to the job at hand. I then looked at relevant live data for engine speed sensors and cranked the engine. My scan tool only listed engine speed from the crankshaft sensor and it displayed around 250 RPM. From that quick test we can see the sensor is working and the signal is reaching the ECU. As there was no camshaft information in live data, I then decided to form a plan. My first step was to test the camshaft sensor with an oscilloscope to see if a signal was being produced and there was good supply voltage and a ground. I decided to connect the oscilloscope to both camshaft and crankshaft sensors to capture both signals, this was done so that if required I could see if a known good waveform could be acquired to check timing of the engine. Thanks to the reference points in both signals, we can compare waveforms to a known good engine and if all lines up we know our timing is good, which saved a massive amount of time as no stripping was involved. Doing both these tests would also give me good direction where to go next with my plan. Fig. 2 is what the scope showed. A nice clean camshaft signal (red) but a very noisy crankshaft signal (blue). This could very well be the issue, and the ECU was possibly having a hard time understanding the signal and converting it to a speed to allow the engine to start. Why then did my live data read okay beforehand? It was at this point one of the other technicians in the workshop needed the scan tool I was using, so I unplugged it for him to use. I then decided to scope the sensors again to see if the signal was repeatable, as this could rule out an issue with the sensor or pick up wheel where the sensor reads on the engine. Much to my surprise, after cranking the engine for a few seconds, the engine fired into life and idled smoothly. Looking at the scope trace, we now had a clean camshaft and crankshaft signal and clearly the engine timing was correct otherwise the engine wouldn’t run. So where to next? The vehicle was now running okay. but I had not actually done anything bar connect the oscilloscope leads to both sensors. Perhaps there was a bad connection in a multi plug? I considered disconnecting everything and road-testing the vehicle. While it was idling away, my colleague who borrowed the scan tool brought it back over. I decided while it was there to clear the stored fault code, but when I gained communication with the ECU and selected clear fault codes, the van cut out instantly. Cranking the engine now it would not start as per the initial complaint of the customer. How could this be? All I had done was connect my scan tool! Lost in the K Line Clearly, we had some strange unexplainable faults going on and I could not answer why they were happening. Why did it start okay before and not now? Why has introducing a scan tool affected things? Rechecking fault codes, the fault had indeed cleared so after cranking the engine over I checked for faults again and had my initial cam/crank coherence fault stored again. With the scope still connected, I cranked the engine over in its non-start condition to check if I still had signals and all looked fine. I also had good engine RPM in live data. However, I had not stopped the scope running and with the ignition on and using the scan tool I noticed out of the corner of my eye the crankshaft signal would rise and fall whenever I used the scan tool which is clearly wrong. Checking a wiring diagram, I found this vehicle uses K Line communication. As it is an older model it does not use Can bus like most vehicles you are likely to see. I then decided the next step was to connect a third channel from my scope to the K Line wire to see if there was any relation to the crankshaft signal. As can be seen in Fig 3. With just the ignition on the K Line signal is causing the crankshaft signal to go high or low. This should not be happening, so I now had direction and could modify my test plan accordingly. Due to the tests I had already carried out this limited the possible causes to either be a wiring issue or an internal fault in the ECU itself. I decided to confirm the wiring on the crankshaft sensor, camshaft sensor and K Line to the control unit for a short together as possibly the loom had rubbed somewhere and bare wires were touching together but all was ok. I then confirmed all power and ground wiring to the control unit was ok which they were. Referring to my test plan, we had confirmed the sensors logging the fault code to be okay. The engine timing was also okay as the vehicle ran ok. Lastly, the wiring was okay so it left the engine ECU itself. Clearly, we had some strange unexplainable faults going on ” Above: Fig.1 Below: Fig.2