27 Feb 2024

It’s all in the timing

A job I was recently asked to look at had a bit of an odd symptom; it would run (apparently quite well), as long as the camshaft position sensor was unplugged. The previous history on the vehicle, a 2009 2.5 TDI VW Transporter, was very limited. During my initial inspection, I noticed that the vehicle was suffering from a slightly extended cranking time, which is quite normal for a cam sensor related issue, as the engine ECU cannot easily calculate which TDC is which.

This was the later engine type (BNZ engine code) which is a Pump Duse (PD) design and which has the DPF and associated componentry. This engine employs electronically controlled unit injectors to meter engine fuelling. The fuel pressure is developed mechanically, inside the injector, by an extra lobe on the camshaft acting upon a pump plunger. Once the pressure is built and the time is right, both mechanically and electronically, the ECU gives the injector solenoid some voltage to enable the nozzle to lift and fuel to pass directly into the combustion chamber.

Just to confirm the fault, I reattached the camshaft position sensor, cleared engine diagnostic trouble codes and cranked the engine. Sure enough, it just cranked over and over. Interestingly and relevant to the diagnosis, there was a distinct lack of smoke and no attempt at firing. I stopped cranking and re-interrogated the ECU fault memory, which was clear. I disconnected the cam sensor, got back on the key and after a few seconds of cranking, the engine sprang into life. When I rechecked the trouble codes after the start event, an open circuit cam sensor fault was present. Therefore, the ECU was monitoring the cam sensor circuit as it should and it was seeing the relevant fault code for the open circuit condition. Although I didn’t have the answer, these two pieces of information provided useful diagnostic data, which was helpful in building a picture of what may be happening.

see main image

FIGURE 2: Cranking during non-starting conditions

Mapping the problem

Figure 2

A feature of this engine is that the timing is controlled by gears rather than a traditional chain or a belt. Generally, non-starts are caused by a lack of combustion, fuel problems, compression problems or timing problems. Cam timing anomalies would typically lead to a host of other associated symptoms such as poor running, rough idle, smoke etc., which this engine wasn’t exhibiting. Fuel problems could also be ruled out by the prompt starting whilst the cam sensor was unplugged. A compression fault was also ruled out on the same basis. So, it seemed that we had an injector timing, rather than a mechanical (cam to crank relationship) timing fault. It is not normally possible to see from a waveform, unless a known good comparison is available, whether or not a synchronisation or timing error is the root cause.

Figure 3

FIGURE 3: Engine cranking during a start

A diagnosis

In the analysis of the two data captures it can be seen that the injector fires three crankshaft teeth prior to the TDC reference (or 55 teeth after TDC depending on your perspective) in both cases. With a liberal application of mechanical knowledge, notice how the crank-sensor signal (blue) varies in amplitude (height); the small (low) parts of the waveform are caused by the engine rotating at a slower speed during the compression phase of the 4-stroke cycle. Re-examine the two traces and you’ll see that the fuel injector timing alters. During no start, the injector fires on the ‘easy’ non-compression stroke (indicated by a large crank sensor amplitude); this is likely to be exhaust/intake stroke when the PD injector has very low pressure (and, therefore, very little – if any, fuel enters the cylinder and injector timing completely wrong). No surprise that the engine won’t go.

In the non-start capture, the ECU calculates the injector firing point at the ‘wrong’ TDC, based on the cam sensor information. The reason there is no smoke from the exhaust is because the injection pressure on a PD engine is built by a separate ‘injector’ lobe on the cam. At this point the injector cam is on the heel and no fuel pressure is generated and no fuel is injected. When the cam sensor is unplugged, the ECU eventually ‘guesses’ which of the TDC’s is correct and fires the injector at roughly the right time, during the compression stroke, as indicated by the lower amplitude crank sensor signal. This helps explain the extended cranking time when the cam sensor was unplugged.

The longest part of this diagnosis was the mechanical strip and fit to gain access to the ECU. Applying a bit of logic to the two running condition waveforms was time well spent, as this indicated what the next steps were for the repair process. My training courses will give you the knowledge regarding system operation and component function and I will put this knowledge in a framework of practical application, enabling you to be quicker and more accurate when you are on-the-job diagnostics. The most popular question asked by people continues to be ‘which tool is best for diagnostics’, my reply continues to be ‘the thing that stops your ears from meeting in the middle’, your brain. Invest in regular updates for this all makes ‘tool’ and you’ll see a distinct improvement in your performance and the satisfaction of your customers.

Find out more

Enquire about James Dillon’s training courses by calling Technical Topics on 01278 428 699.