Hardware Reference
In-Depth Information
Fault-finding and
troubleshooting
techniques
A popular misconception concerning electronic fault finding is that good
troubleshooters are borne and not made. The implication of this is that the
skills of a service or test engineer cannot be acquired unless the person con-
cerned happens to possess the equivalent (in electronic terms) of 'green fingers'.
Nothing could be further from the truth - indeed it is quite possible for any-
one of moderate intelligence and manual dexterity to successfully locate faults
on even the most complex systems. The secret lies with adopting the correct
approach
to troubleshooting. This is the real key to successful fault finding.
With experience, the right technique will come as second nature. Indeed, a
practised service engineer may not even be conscious of the technique which
he or she is applying when tackling a fault. They may appear to get right to
the cause of the problem without even thinking. By applying a little logic and
reasoning, you can do the same.
Fault finding is a disciplined and logical process in which 'trial fixing' should
never be contemplated. The generalized process of fault finding is illustrated in
the flow chart of Figure 13.18. The first stage is that of identifying the defec-
tive equipment and ensuring that the equipment really is defective! This may
sound rather obvious but in some cases a fault may simply be attributable to
maladjustment or misconnection. Furthermore, where several items of equip-
ment are connected together, it may not be easy to pinpoint the single item
of faulty equipment. For example, take the case of a process control system
in which the user simply states that there is 'no output'. The fault could be
almost anywhere in the system; computer, display, printer, or any one of several
connecting cables.
The second stage is that of gathering all relevant information. This process
involves asking questions such as:
•
In what circumstances did the equipment fail?
•
Has the equipment operated correctly before?
•
Exactly what has changed?
•
Has there been a progressive deterioration in performance?
The questions used are crucial and they should explore all avenues and eventual-
ities (particularly when the repairer has no previous experience of the equipment
in question). The answers to the questions will help to build a conceptual model
of the symptoms - before and after the fault occurred. Coupled with knowledge
of the equipment (e.g. its performance specification) this model can often point
to a unique cause.
Once the information has been analysed, the next stage involves separating
the 'effects' from the 'cause'. Here the aim is simply that of listing each of
the
possible
causes. Once this has been accomplished, the
most probable
case
can be identified and focused upon. Corrective action should be applied (to this
cause alone). Such action may require component removal and replacement,
adjustment, or alignment, etc.
Next it is necessary to decide whether the fault has been correctly identified.
A component may have failed (open circuit or short circuit) or a fuse may have
blown. This will confirm that the cause has, in fact, been correctly identified. If
so, the fault can be rectified and the equipment brought back into service. If not,
any new information that has been generated can be evaluated before reverting
to the selection of the next most probable cause. In practice, the loop may have
to be executed several times until the fault is correctly identified and rectified.
