Challenging current techniques

By Frank Massey | Published:  18 July, 2017

Frank Massey looks at how you need to always keep an open mind on diagnostic methods

Accepting my reservations, I always believe in challenging any current techniques with an open mind with the possibility that a better method may exist. My reservations are based on the following reasons.


Electronic measurements are based on sensor signatures compared with known confirmed values. Many engines now employ variable valve lift and timing. Added to this complexity, often there is no mechanical locking of sprockets to shafts. Confirming actual mechanical camshaft position regarding the crankshaft would require sensors monitoring all camshafts and all sprockets.

With exception of highly advanced engines this is not the case. So how can we predict actual valve functionality before opening the spanner drawer?


Recently we filmed a detailed diagnostic process with Pico exploring a reliable method. Determining electronic plausibility is relativity easy. The answer to actual valve operation is reliant on accurate pressure evaluation in the cylinder. This must be overlaid in real time with the timing sensor signatures. There is still an element of error as you will need to calculate camshaft rotational angle against the crankshaft profile. The Pico software helps by converting the oscilloscope base line into the 4-stroke rotational position. So far so good, but can poor mechanical condition be predicted more easily? Well yes it can, remember the vacuum gauge. Volumetric efficiency is directly proportional to correct valve position.

Hang on a moment; Where should the valve position be? It’s not simply a function of a chain or timing belt. The computer now has an opinion on optimum timing angle. It’s not that easy anymore. However, I do support the premise of predictive evidence.


Let’s visit an actual event in our workshop from very recently; a Porsche Boxter 3.4 supported with Bosch Motronic med control. With no obvious cause or warning it developed a drop in power and throttle response. No definitive DTC. You didn’t think I was going to make it that easy for did you?

Serial data confirmed two key symptoms. Misfire count and rich fuel trim more than 20% on one bank only. It is a low mileage car with no obvious signs of poor care. Removing the spark plugs confirmed excessive soot as the symptoms of incomplete combustion. Note my words carefully. Further examination excluded the unlikely cause as poor ignition energy. Back to basics, the vacuum gauge confirmed too high pressure in the intake system specific to the offending bank. The next action was to attach the pressure transducer to each bank in turn comparing the pressure profiles.

Direct measurement

To achieve this, direct measurement from each bank is required. Running the engine with the pressure sensor directly in one cylinder from each bank confirmed a differential in pressure between banks. It must be a mechanical problem then. The next decision is more intuitive than evidence driven as this engine has variable valve timing adjustment via electro-hydraulic control.

Serial data could not provide actual camshaft position data though, as the sensor is only motoring the sprocket position. This is the very point I made earlier. We could get clever and monitor control duty cycle and current through each actuator. Unfortunately, the computer does not monitor a problem, so there will not be any correction made.

Complete restoration

Examination of the service history confirmed long life servicing. This prompted us to flush the engine and replace both hydraulic camshaft control actuators. This resulted in a complete restoration of performance; obviously, we reset the adaption values to fuel trim, throttle angle and misfire count.

As is often the case, a combination of hard evidence and common sense prevailed. I often say to entry level diagnostic students, the most complex problem in diagnostics is you.

Further information
Please contact Annette on:
01772 201 597 or email for further information on upcoming training courses and events.

Related Articles

  • Glowing, going, gone! 

    I decided to share this case study for my first article because what I expected to be a simple job turned into something a little more complex and gave me an opportunity to study a and learn about a system that until now I’d probably taken for granted.

    We were presented with a 2010 Skoda Fabia 1.6 TDi by a car dealer who had recently taken it in part exchange. The engine management was light illuminated, however with no other symptoms. The previous owner told the dealer that the MIL had been on for around a year and her local garage had failed to repair it. It had also recently been recalled for the ‘Dieselgate’ VAG emission software update. The dealer told the customer there were DTCs stored for the glow plugs and that they needed replacing to which she declined as she was sure they had previously been replaced. We already had a reasonable amount of vehicle history to start with, and were ready to take a look.

    Voltage and current
    A code read revealed DTCs for all four glow plugs being open circuit and a glow plug module communication fault. A quick inspection of the engine revealed that the glow plugs were not that old and also there was a new glow plug module fitted, plus an old one found in the boot.

    While checking the resistance of the glow plugs may tell us something, measuring the voltage and current with an amps clamp paints a much clearer picture. The oscilloscope was connected and the ignition was cycled. The screen capture revealed a healthy 12 volts for around 10 seconds then pulsed at random, however there was zero amps flowing (on all glow plugs). It was clear the plugs had gone open circuit for some reason so they were removed for inspection. It was then we noticed that the heater plugs fitted were rated at 4.4 volts, so now we know why they burnt out! Could they be the wrong glow plugs? Could it be the wrong control module? We checked and found the part numbers were correct.

    At this point it was crucial that we understood exactly how the system is wired and how it should operate. By studying a wiring diagram we were able to plan how we were going to test the system (see image 1). Starting with the power supplies and ground, it is always best to test a circuit in its normal environment which means we really need the current load of working heater plugs. If we were to fit new heater plugs at this point there was a high risk of them being damaged which is expensive so we substituted four headlamp bulbs instead. The fuse rating for the circuit was 50A so with a quick bit of maths we calculated the current required for four bulbs was safe. The main live feed, ground and ignition switched live were all good so we moved on to the two communication wires that link directly to the PCM.  

    If the PCM can log individual codes for each glow plug then we know that it must have a two-way communication system. Scoping both wires with the module connected and disconnected showed us that there was clearly a command signal from the PCM and although it was random and rather messy (see image 2), the glow module responded directly by activating the glow plugs at the same rhythm.

    The second wire had totally different digital signal which had to be the feedback to the PCM. The noise and irregularity of the command signal was clearly an issue so we checked the wiring back to the PCM and with the aid of the good old-fashioned wriggle test the fault was identified as a poor connection in the PCM harness connector. The connection was cleaned and the system retested which revealed a much healthier scope pattern and the communication DTC was cleared (see image 3).

    Reliable repair
    At this point we could have fitted new glow plugs but to save unnecessary expense we wanted to make sure it was a reliable repair so we decided to monitor the system with the faulty glow plugs still installed and the leads connected to the bulbs. We started by monitoring all four glow plug voltages on the oscilloscope. Using the scan tool to activate the glow plugs showed us that the 4.4 volts is achieved by pulse width modulation at a duty cycle of around 13% with a frequency of around three times per second. What was more interesting was that all four plugs were individually triggered in a sequence (see image 4) so there is never more than one glow plug energised at any one time. The logic behind this is that it makes a substantial reduction in power consumption.

    Our next test was to observe the control strategy of the PCM from a cold start and warm-up phase. The objective here was to ensure that there was no software related issues. From the point of key on there is a 1.5 second supply phase to heat the plug as fast as possible then temperature is maintained by the 13% duty control.

    Decade box
    Of course, after a period of time, once the engine starts to warm up the system turns off and the communication wires go quiet. If you want to test it more than once then you’d have to wait for the engine to cool so to save time we connected a decade box in place of the engine coolant temperature sensor and by observing the coolant temperature in serial data on the scan tool we were able to select a variety of resistances that would represent low temperatures and fool the PCM into commanding glow plug activation.

    The decade box has proved to be an extremely useful tool really is a must in any diagnostic technician’s tool box. It is great for substituting in place of certain sensors and components to check the integrity of a circuit or to observe an ECU responding to a variation in signal (resistance).

    The final test was an observation of voltage over current on one glow plug. The other interesting thing we noticed was the simplicity of the digital feedback signal. By unplugging each glow in turn you could see the pattern in the signal change and when all were connected and working it was a regular pattern.

    Summing up
    Clearly more time was spent on this job than necessary and the labour charge remained fair. In a busy workshop it is hard to find spare time for these situations but my point is that sometimes sacrificing a lunch hour or staying behind for half an hour gives an opportunity to learn so much which can only aid you in speeding up diagnostic time and process on future jobs.

    Winning the Top Technician 2017 competition was unexpected. It has not only introduced me to some very inspiring, like-minded people, but has also taught me you can never have too much training, whether it’s self-training like in this instance or on a professional training course. There are some fantastic training companies offering a variety of courses available now. Also, some of the best and most respected all regularly write for Aftermarket!  

  • New to the UK: Bosch BAT 645 and BAT 690 battery chargers  

    The new BAT 645 and 690 battery chargers from Bosch feature 45 or 90 amperes of continuous current output.  They are suitable for conventional lead-acid and gel batteries, as well as absorbent glass mat (AGM) and enhanced flooded battery (EFB) technologies and lithium-ion LiFEPP04 (LFP) batteries. When the tool identifies an ‘exhausted’ or empty battery, the battery chargers use a specific setting to maximise battery life. Using back-up mode, the devices ensure a stable power supply even when the starter battery has been removed. Meanwhile, buffer mode supports on-board power supply during control unit updates.

  • Spin the wheel 

    I have been asked several times about ABS wheel sensors. Like many other components, the technology is changing. The changes reflect the expansion in integrated chassis dynamics.

    Just imagine how many functions require wheel speed and rotational differential data.

    ABS, dynamic stability, hill start, audio volume, navigation, self park, all wheel drive, active steering assist, electronic handbrake etc. Sharing this data on a high speed can network ensures very accurate vehicle motion dynamics.

    Older variable reluctance sensors (VRS) rely on a coil generating an alternating voltage when rotation occurs. The problem is they are not directional sensitive and cannot report motion at very low speed. Air gaps were critical as they affect signal amplitude. They are often referred to as passive sensors. So, the introduction of digital or active sensors was inevitable.

    How do we tell them apart? Active sensors require a voltage supply from the ABS PCM, with a ground or signal return. They operate with different principles of signal generation; hall, and magneto resistive. Pure hall effect sensors will switch between the supply potential voltage and ground. Magneto resistive sensors operate on the principle of current and voltage change in response to a change in magnetic induction. This change can be introduced in several ways reflected in wheel bearing and sensor design. Smaller sensors with integrated magnetic field rings are now the norm. Developed by NTN at their Annecy facility they are called encoded bearings. A small ring mounted at one end of the bearing carries a series of north south poles. These have now been replaced by dual encoding, two sets of magnetic rings with a unique offset. This enables the abs module to determine direction of rotation.

    Subtle differences
    There are two very subtle differences in the digital outputs. They can be called pull up or pull down. The sensor supply voltage will be slightly lower than battery voltage this is due to the different internal resistance values. However, it will be around 10.5/11.5v.

    The ground or return signal value will vary between 0v or 1.4/1.8v. You could have a sensor or circuit fault; let me try and explain the subtle differences, and how to prove which is which. Remember the golden rule if in doubt compare a wheel circuit that works normally.

    First unplug the sensor and measure both circuits in the loom. With no load applied the supply voltage should jump up to NBV

    Next check the ground circuit if its true ground then it’s a pull-down type and the signal will be on the power line, and may only be around 200mv

    If a small voltage exists then it’s a pull up type and the signal will be on this wire not the supply. The digital signal will be very small when the wheel rotates. It could be small around 200/400mv, or as high as 0.5/1.8v, depending on the manufacturer variant

    Common sense would dictate the serial route is easiest, however how would you determine an intermittent fault? It could be a faulty sensor, faulty encoder, or a circuit error. The only way is using a scope. Should we measure voltage or current though? Both change in the circuit. Unless you have a very special current clamp, go for voltage and select a AC coupling.

    The specific question I am often asked is current measurement, well I can tell you in a pull-down circuit its around 7-15 ma with a 400mv voltage change. The pull up type will produce around 6/13ma with 0.2/0.35mv.     However, these voltage values can vary due to the value of the two parallel internal sensor resistors these are normally 1.4k ohms, with a much higher resistor in the meg ohm range, within the ABS pcm.

    I hope this helps. The pico image was taken from a VW Golf 1.4 TSI. The easy bit is replacing the wheel sensors. Ever since metal housings were replaced with plastic they never corrode in the housings
    do they…?

  • UFI Filters enhances UK offering 

  • Ignite your interest in ignition  

    This month’s subject was prompted by a recent conversation with a colleague in Australia. The conversation included an invitation to a technical festival in October, where it was said that ignition would be one of the subjects of interest. Many years ago, when I began developing our training programme, ignition was a subject of primary concern when diagnosing gasoline
    engine problems.

    This is a complex subject often not fully understood and often overlooked. Its vital importance recently became apparent in our workshop, when we were presented with two Audi rs6 engine failures. One failure has yet to be investigated the other suffered piston failure due to combustion faults.

    The increasing complexity of homogenous and stratified fuelling, split injection delivery and variable valve timing geometry has placed critical responsibility on ignition performance. Often within the diagnostic process there is no serial evidence of an ignition problem, or that what evidence is available is incomplete especially at the early stages of failure. The process has not changed in over 30 years;  You must scope it.

    Process overview
    So here is an overview of the process. Firstly, you must understand that it requires a specific amount of energy to completely combust the air fuel charge. Ignition energy is measured in joules, our task it to ensure the energy is created and delivered correctly. The primary circuit bears the responsibility of energy creation with current profile as the focus of our measurement. The secondary circuit has the responsibility of delivery, our focus is burn time and slope profile.

    I accept that both circuits have a shared responsibility at the point of induction where energy within the primary is transferred into the secondary. The physical challenge is the method of accessibility. With static or direct ignition it is often not possible to connect to the coil primary circuit, leaving the option of induction as the method of measurement. The primary will always have a power and switched ground, so current measurement using a suitable hall clamp is always possible.

    Diagnostic observations
    The four critical diagnostic observations in order of priority are:
    Ignition burn time measured in milli-seconds with a range of 1-3ms depending on ignition type. Do not assume length of burn relates to energy value Primary current profile with a range of 3amps (points ignition) 20amps static ignition. Note the expression profile, it includes rise time and rate of collapse Coil ringing, this is the resonance at the end of the burn event it represents the small residual ignition energy returned in to the coil secondary winding Firing line voltage, this represents the value of electrical pressure in delivering the induced energy to the spark plug electrode it includes all components in the delivery process

    You must also understand that the performance of the injector, cylinder turbulence, and mechanical efficiency forms part of the combustion process. Intake air temperature, pumping losses and fuel quality all affect the burn process. Let’s begin with the tool I distrust the most! Serial data is a good first look – there is some very useful information such as cylinder misfire count, ignition timing individual timing retard data, air intake temperature and exhaust temperatures. There may also be additional data on burn time and primary charge time, but I don’t trust or rely on it.

    So, out with my Pico scope. Connectivity can be a challenge, over the years we have built our own probes, however, if the manufacturers can run a circuit there you can scope it. There is a simple logic process.  Begin with burn time, look at the duration and slope it – It should be roughly parallel with the horizon.

    A rising line confirms a difficult transition of energy across the electrode. Lean combustion, glazed plug, cylinder pressure, plug performance. Cylinder turbulence.

    A falling slope represents the opposite condition; low cylinder pressure, fouled or shunting plug circuit, small plug gap. The burn profile should be relatively smooth, a turbulent burn path confirms difficult in cylinder conditions. It can and does point to injector fuel delivery problems especially if a sharp rise at the end of the burn time is present.

    You may appreciate now just how vital scope evaluation is.

    Primary current path confirms good power supply and the performance of the power transistor in its ability to switch and hold load to ground. Note the rise time characteristics and the off switch, under shoot here is a good indication. If you can, observe primary voltage. Note the slow rate on load, it’s the slow rise in voltage during coil charge time, a problem here will affect current flow so go for current first its easier to understand. Remember one of my core diagnostic rules; If it moves, gets hot, or applies a load measure current!

    Coil ringing is the inverted energy returned into the coil secondary. With no path to ground,  it gradually gets weaker, converting its energy to heat. Expect 2/3 rings in current systems. If the coil windings are compromised in any way a reduction in inductance will follow. The rings will disappear, ignition energy may still be present but a reduction in value will result. Be warned this condition will never be known if not scoped and critical engine failure often follows.

    Firing line voltage can only be measured accurately in primary to be honest. Expect the following values:; Conventional rotating ignition 50v, wasted spark ignition 40v, direct ignition30v; Plus or minus 5 v on all values. The problem with exploring this with a coil probe is that the probe attenuation is not known, so its difficult to scale.

    I hope this helps. It is a very complex subject , often neglected and overlooked.

    Just before I go here is a challenge; How many information systems, VMs especially, don’t give these four  vital statistics? So how do they know if there is a problem?

Most read content


Sign Up

For the latest news and updates from Aftermarket Magazine.


Where should the next Automechanika show be held?


Click here to submit an event


©DFA Media 1999-2016