March 2023 - How to fault find

Greeting again, my dear subscribers, Closing in on the target of a thousand subscribers for March. Thank you very much for the continued support.

So far the newsletters have focused on case studies of which I have undertaken at my day job - Schmiedmann Suomi Oy in Vantaa Finland. This edition is more orientated to those who are new to the industry and are interested in starting out in fault finding but are struggling or who, perhaps, want to know more about how to go about solving these, at times, challenging problems. So I thought I would put together a basic example of fault finding which can get some of you with less experience a little bit further along in your careers by applying some of the tips and techniques that this article will cover.

The image example above is a typical look at an average working day for a fault finder. A basic check of the output of an alternator reveals, with a running engine, a non-charging event. In this sort of case it is usually enough to look at this and determine that the brushes and slip ring are more than likely worn out. Advanced fault finding could be used if we were to strip this down and measure diodes, stator and windings with a view to rebuilding but usually simple faults like this are easy to diagnose with a good quality Voltmeter such as a Fluke. What if the fault is more involved? This is quite a straightforward fault find. Let us consider, however, how exactly would a novice fault finder go about diagnosing other types of faults without experience on possibly more complex circuits? The answer is knowledge and practice - lots of practice. Let us take a look at a simple headlamp circuit and start the diagnosis process!

Take a look at this diagram schematic and familiarise yourself with how it works. This is a simple system that uses a CAN signal from a headlamp switch or module which is recieved by a body module that processes the CAN message and switches transistors to send 12 Volts to each headlamp on two individual lines. The lamps are grounded to a common ground that has been simplified in this illustration for the sake of clarity. In reality each lamp would be grounded on the left and right sides respectively. These type of BCM controlled lamps feature self diagnosis by monitoring the bulb lines. So how do we start the diagnosis process if there is an issue with one or more of these headlamps?

• Read the work order and glean information - what exactly is the customer reporting?
• Confirm the fault - is it working or not working? What is not doing that it should be doing? Permanent or intermittent?
• What is the likely cause?
• What tools will be used to diagnose this fault?
• Create a test plan - either from the tester or create your own test plan.
• Locate and rectify the fault.
• Re-test the operation before handing over to the customer.

A typical sort of job that states that one headlamp has failed. The first thing to do is to confirm the fault by actually looking at the car, turning the light switch on and off etc, noting the fault. Fault codes, can of course, be read from the fault memory or, alternatively, one can dive straight in and check the basics, voltage supply to the left lamp etc. The biggest issue that I have experienced when coaching beginners in this field is that they tend to get confused at the wiring diagram contents or not having an idea where to start in the first instance. This is where the skills of interpreting a wiring diagram, applying knowledge and practicing helps long term.

In this example the left headlamp has stopped working and the scan tool shows a typical EOBD fault code that the lamp has an open circuit on the input line. But where is this open circuit? And how to find it? What about the wiring diagram, what can be ignored altogether based upon this fault code? These are the sorts of questions that I, as a fault finder, ask myself everytime I embark on a fault finding process. In this case it is a relatively simple circuit but eventually in your career more complex faults will appear involving multiple circuits so getting the basics right first time is essential before tackling larger fault finding problems.

In the case above it is clear that there is no need to check the first part of the circuit to the BCM so time can be better utilised focusing on the output to the left headlamp. The next thing to consider is where is the easiest place to check the Voltage supply to the lamp. Since this is an open circuit the BCM has not switched off the power supply since the supply is open and the BCM has detected that the nominal current draw of the left Halogen lamp is not at the threshold for a working lamp. The usual current draw for a 55W H7 bulb would be around 4.5 Amps and since current needs to flow over a consumer drop and return back to the battery source the BCM knows that, in this example, there is no current draw hence the fault code stored in the BCM.

The easiest place to start to take a Voltage sample is at the headlamp itself, maybe. If the headlamp connector is almost impossible to reach then try to find a place to tap into the line as close to the lamp as you can. If there is no Voltage found at the test area the next place to go is to the BCM output terminal itself. If No Voltage is found at this point then the line can be disconnected temporarily at the BCM and powered up the full length from point A to point B. If the line manages to light up the bulb then the full line can be assumed to be ok and the fault lies in the BCM itself or an open in the terminal connector. Powering up an entire circuit from source is often a great way of confirming the total integrity of a circuit. When faced with this sort of dilemma whether to replace a faulty BCM or not ask yourself the following questions:-

• Are there any manufacturer technical bulletins that apply to a hardware fault within the BCM?
• Is there a way to reset "open or short events" in the BCM?
• Is there anything else that could influence the BCM into failing? Water leaks onto the connector?
• Is the harness connector itself a good fit? male to female etc. Should this be checked by removing the pin and performing a drag test?
• How does the system deal with open circuit events? Does it keep the output active or does it switch it off in the event of an open circuit?

How about that last point? How can we establish how a BCM manages an open circuit, short circuit event if access to functional descriptions are limited? For example, many manufacturers will publish technical information that describe in relatively good detail how things work. If this is not at hand then the easiest thing to do is to disconnect the working headlamp, in our case study the working lamp was the right lamp so it makes sense to then check the ouptut of the right headlamp at the BCM. If there are 12 Volts present and the wire on the right from point A to Point B is in more or less the same state as the left, non-working side yet there is still a 12 Volt ouptut from the BCM then once you have the answers to the checklist above the only thing that can be faulty is the BCM.

The next step is to understand the reason for the BCM output to fail on the left side. Ask yourself:-

• Is this a known type fault, is this a common item that has been known to fail and if so how often?
• Where can I research online about this fault that might give me access to more reports of this item having a weak point or a known component failure?
• Do you have personal experience of this particular fault occurring more than once? How did you cure the fault last time?
• Are there any other technicians working at your organisation who may have also experienced issues similar to this?
• If a new BCM is fitted am I sure that no external factors could damage the new BCM? Bad wiring, shorts etc.

The worst thing that can happen is to fit a new component without fully finding out as far as possible the reason for it to fail in the first place. Sometimes internal faults can be beyond the scope of automotive technician responsibilities and in such cases troubleshooting circuit boards may be best left to specialist companies who can do these types of repairs. The main thing is to at least determine that anything on the vehicle that may be faulty cannot damage a new replacement unit.

How about troubleshooting this circuit above? You will notice that it is similar to the previous diagram except in this case it has no CAN bus, no transistor switching and no control units. In this case a simple mechanical switch has continuity to ground when in the on position. This then provides a ground at the relay coil which is energised at terminal 86 with the ignition on. The relay clicks over and terminal 30 provides a permanent voltage that is output on terminal 87 to the lamps on one line. Lots of things can go wrong with this circuit such as:-

• Switch failure
• A bad ground at the switch
• Non-conducting relay contacts, pitted, burnt, corroded etc
• Fuses can fail due to short circuits to ground
• Wiring failure between fusebox and relay
• Wiring from relay to headlamps
• Loose fitting terminal connectors
• Headlamps themselves
• Bulbs

This list is not exhaustive and there are many other issues than may be added to the list. Here is the same diagram with some suggestions of how to deal with a scenario with two lamps not illuminating and how to go about diagnosing it. The first step is always evidence gathering. What can you test, see or feel? What is this component, wire, connector doing or not doing that it should be doing?

In this sample case, using a special tool called "Human Ears," we may be able to utilise a second person to operate the headlamp switch while the fault finder listens at the relay. If the relay clicks the output can be measured. Just because a relay clicks it should never be assumed that it is actually ouptutting a usable Voltage. Many relays will pit and fail on the contacts yet the coils are often very hardy and can far outlast a relay contact points surface.

If this relay is not emitting anything then we can use a Fluke meter or, my personal favourite, a Power Probe Maestro to check for grounds, lives and to provide a safe and fused power supply to power up and test components. The live feeds to the relay can be effortlessy checked by touching the Power Probe tip to the terminals. Remember to make notes as you check off each item. In the end deductive reasoning will leave very little things on the table left to check and the culprit will rear its head rapidly.

Here is another example of a headlamp control system. We can read fault codes from the headlamp switch and the body control unit. If the body module reports something like "LIN communications fault headlamp module" it is easy to check first of all (in the most acessible place) is there any voltage on this LIN wire? Can I measure at the headlamp switch module and also at the body module? Is the voltage the same? If a wire is suspected can I remove this wire from both ends and load test it? In these cases always check what is easiest and quickest. In this case if you suspect a LIN failure but not the wire itself then check the headlamp switch module, is there 12V or a ground? Does the module have a communication problem on the CAN bus (not shown in the diagram). Sometimes scoping the wire may be prudent and although I have featured many case studies using an oscilloscope I will always try to find faults in a more simple way if possible since I have to be efficient, we still need to make a turnover for the company profit margin.

By using evidence, a wiring diagram and a checklist one can quickly narrow down a fault in 90% of all cases presented at a workshop. Of course there are times when a job becomes an absolute nightmare due to a plethora of failing components that can end up running over a hundred saved fault codes from various different systems and across multiple different BUS protocols but I will leave those for another day and another newsletter. Remember to make a big effort to understand how systems work and then practice how you will go about testing them using the tips in this newsletter.

I hope that this very basic introduction to tips and tricks of a fault finder will help you along the way in this challenging industry. Please feel free to get in touch via the platform if you have any questions. Until next time keep up the learning and keep up the good workmanship! See you in the April newsletter. Thank you all for the continued support and positive feedback :)

#troubleshooting #faultfinding #learning #teaching #technicaltraining #knowledgesharing #knowledgeispower #bmw #schmiedmann #mechanic #mechanics #problemsolving #problemsolvingskills #helpingothers #teamworkmakesthedreamwork #automotiveindustry #diagnostics