Unit of Competence:

Implement Maintenance
Procedure
 LO 1:
Determine best practices for
equipment and software maintenance
Identifying computer hard ware to be maintained
• There are several ways to identify the normal operation of a personal
computer.
• Most people use diagnostic software packages like PC Tools, Norton
Utilities and/or Check It to test a computer.
• Those diagnostic packages provide user-friendly operations to perform
testing of a computer.
• However, you can initially make measurable observations using your
senses, that is, the sights and sounds to identify the normal operation
of the computer.
• The table below suggests where to look and what you might hear to
get an indication of normal behaviour of a PC.
Cont.…
Device Sights Sounds

System unit • Floppy disk drive activity indicator Floppy disk drive mechanisms
(light) Speaker (beep)
• Front panel indicators such as: Fan
➢ Power on LED (light emitting diode) Hard disk drive

➢ Hard disk drive activity LED

Display unit (monitor) Power on indicator

Text displayed on screen

Keyboard Num lock indicator

Caps lock indicator

Scroll lock indicator

Cont.…

Table 1.1: Reference points for indications of normal behaviour of a PC

Potential sources of damage to computer hardware
and software
There are a number of common causes of damage to a computer or its
components. These are:
• Temperature variations
• Power cycling
• Static electricity
• Power line noise
• Radio frequency interference
• Phosphor burn on a monitor
• Dust and pollutants
• Water
Temperature variations
Cause
• Temperature variations (expansion and contraction of components from
temperature change) can lead to serious problems.

Damages
• Chip creep : where the heating and cooling of components can cause movement,
usually out of the socket that holds the component.
• Signal traces on circuit boards can be cracked and separated.
• Solder joints can be broken.
• Contacts undergo accelerated corrosion.
• Solid-state components can be damaged.
• Read and write problems on hard disk drive (due to expansion and contraction of
the platter of hard disk the data may be written at a different location relative to
the track centre).
Cont.…
Advice
• Ensure a computer operates in correct ambient temperature. Refer to
the computer User’s Manual for this information.
• Ensure the ambient temperature when the:
❖system is on it is in the range of 15-32 °C
❖system is off it is in the range of 10-40 °C.
Power cycling
Cause
• Turning on a cold computer subjects it to the greatest possible internal
temperature variation.
Damages
• Same as for temperature variation
Advice
• Power on a computer only once daily. Don’t turn a computer on and
off several times every day.
Static electricity
Cause
• This problem usually appears during winter months when humidity is low, or in
extremely dry climates where the humidity is low year-round.
• Some static-sensitivity problems are caused by improper grounding of computer
power.
Damages
• Electronic components
Advice
• Always use a grounded power cord plugged into a properly grounded outlet. You
could use an outlet tester to check that it is properly grounded, but today, OH&S
requires that all power equipment be properly tested and certified. This includes
the outlets, cables and connectors.
• Use a grounded static mat underneath a computer, or an antistatic wrist-strap,
before touching internal components of the computer.
Cont.…

Fig.1. Grounded static mat Fig.2. Antistatic wrist-strap

Power line noise
Cause
• This problem is caused by poor quality power being
supplied to a computer system.
• It can also be caused by sharing a power source with other
higher power consuming equipment, such as coffee
makers, copy machines or a laser printer.
• The wire size and length will affect the resistance of a
power circuit.
Damages
• All system components
Cont.…
Advice
• A computer system should be on its own circuit with its own circuit
breaker.
• A three-wire circuit is a necessity.
• To decrease resistance, avoid extension cords unless absolutely
necessary and then use only heavy-duty extension cords.
• Avoid using too many items on a single outlet.
• Add an Uninterruptible Power Supply (UPS) as a power conditioner.
Radio frequency interference
Cause
• Mobile phones, cordless phones, fax machines and any radio
transmission equipment
Effects
• Sporadic random keystrokes will appear, as though an invisible entity
were typing on the keyboard
• White spots and lines appear on the screen
Advice
• Install specially shielded cables (built-in toroid core cables) outside a
system unit.
Phosphor burn on a monitor
Cause
• The phosphor on a cathode ray tube can be burned.
Damages
• Reduces the life of monitor (cathode ray tube)
Advice
• Turn both brightness and contrast levels to the minimum.
• Use a screensaver that displays different patterns on a screen.

NB: CRT phosphors are specific chemical compounds, which are

arranged on the inside of the screen tube in a certain order and are
available in various colors, starting with basic red, green and blue.
Cont.…

Fig.3. A monitor with phosphor burn

Dust and pollutants
Cause
• A power supply fan carries airborne particles through a computer.
• Food crumbs are attracted by magnetic media, while cigarette ash and smoke are
drawn toward disk drives.
Damages
• Floppy disk heads and media
• Electronic components (dust on the surface of components prevents necessary heat
loss)
Advice
• Use power supply unit with air filter (the filter must be cleaned and changed
periodically).
• Don’t operate an unprotected computer in a dusty environment, e.g an industrial
workshop.
Water
Cause
• On a desktop, coffee or tea spills over a keyboard or into a monitor.
Damages
• Keyboard malfunction
• Monitor explosion (if a monitor is on)
Advice
• Never eat, drink or smoke inside a computer room.
The first steps towards troubleshooting
Reflect
Here is a typical scenario reported to the help desk.
• A client phones the help desk and reports that the computer hangs each
time they try to run a particular application.
❑What might be the source of the problem?
❑What steps will you take to find out?
This is the trouble shooter’s challenge!
Cont.…
Feedback
• In all cases where you are trying to troubleshoot a problem, you need
to use a logical step-by-step approach. For example, two questions
that you would always ask in this situation are:
❑When did the problem begin?
❑Has any new hardware or software been added between the time that
the problem appeared and when the system was last working
correctly?
Cont.…
Here is a list of reasons why a computer might hang each time a specific
software application is run. It could indicate:
• a corrupted file
• an incorrect installation
• hard disk failure
• a virus
• a new application causing conflict
• new hardware causing conflict
• new device drivers causing a conflict with older software.
General troubleshooting guide
Here’s a general troubleshooting guide that you can use when a
computer develops a fault.
1. Don’t panic.
2. Observe:
• What are the symptoms?
• What conditions existed at the time of failure?
• What actions were in progress?
• What program was running?
• What was displayed on the screen?
• Was there an error message?
• What functions are still working?
Cont.…
3. Use your senses (sight, hearing, smell and touch).
• Is there any odour present?
• Does any part of the system feel hot?
4. Check power supply:
• Is the plug inserted snugly /closely/ into the computer?
• Is the power cord plugged into an appropriate wall power outlet?
• Is the wall power outlet working?
Cont.…
5. Documentation (fill in a pre-designed check list):
• What is the computer doing?
• What is the computer not doing?
• What is being displayed on the screen?
• Is there any error message?
• What is still operating with everything connected?
• Is power still operating on each part of a computer?
6. Assume one problem:
• Use correct data and resources
• Use relevant technical manuals and information
7. Use proper test equipment.
Cont.…
8. Isolate units one-by-one:
• If a system worked when all peripherals were disconnected, turn power off
and reconnect one of the peripherals. Power on and test. If that unit works,
turn the power off and reconnect another peripheral. Again, power up and
test. Follow this procedure until a unit fails.
9. Consult your index of symptoms:
• Using your logbook, help desk database, or any relevant flow charts in
reference books and manuals.
10. Localise to a stage.
11. Isolate to the failed part.
12. Test and verify proper operation.

Remember that: After diagnosing and rectifying the fault, you need to
document it in the log book or help desk database for future reference.
A hardware fault-finding checklist
Here’s a useful checklist that you can use to help you diagnose faults in
hardware.
• First, consult any service level agreements (SLA) to ascertain if or
clarify response time obligations and internal/external responsibilities.
Determine also if there are there any other organisational guidelines
you need to follow.
• Consult documentation logged from previous related or similar
situations. Determine a set of questions can you ask the user, your
colleagues and your supervisor that might assist you in finding a
solution.
Cont.…
• Remember to keep safety as your highest priority by observing OH&S
precautions, that is, ensure your own safety first, and then consider
other precautions such as static discharge, etc.
• Check the power supply. Ensure it is working and that it is powering
the motherboard.
• If no video is displayed try swapping the monitor with a known good
one.
• If the video controller is built in, disable it and try another known
working video card. To disable the built in video controller you will
need to access the system CMOS or BIOS setup. On some systems,
simply inserting a new video card will automatically disable the built
in video.
Cont.…

Fig.4. video card

Cont.…
• Remove all expansion cards. If the machine boots, replace the cards
one by one until the problem reappears.
• Check the CPU fan is operating.
• Check the RAM chips by swapping them with known good ones.
• Check the motherboard for signs of blown components.
• If still no success, you might swap the entire motherboard and CPU.

➢Remember to document everything you do according to organisational

guidelines.
Is the problem with the hardware or the software?
• A computer system consists of a hardware sub-system and a software
sub-system.
• However, when looking for the cause of the fault, sometimes it can be
difficult to determine if the fault is hardware or software.
• Once you have determined that the fault is confined to one of these
two sub-systems, you can then isolate it, focus on the fault-finding
process, and rectify the fault.
Cont.…
• The easiest way to determine whether a problem is hardware or
software is to test the hardware with software packages that are
known to be good and that have successfully run on the system before.
• If the system boots and operates correctly, then the fault can be put
down to software.
• If the system does not boot or operate correctly then the fault can be
put down to hardware.
Configuration problems

• Configuration problems are problems that arise when a computer

system is set up for the first time or when new peripherals/components
are added to the system.
• When the component is first added, the system is not ready to receive
the hardware, until the system is prepared to support the device.
• This mismatch can be rectified by:
❑installing the appropriate software device drivers
❑configuring CMOS/BIOS
❑configuring the operating system.
The POST
• The Power On Self Test (POST) operates whenever a computer is
switched on.
• Whenever you start up the computer system, the computer
automatically runs a series of tests.
• At this time the computer tests the basic functionality of vital
components such as the CPU, RAM, video card, motherboard, and
input and output devices.
• POST tests are not particularly thorough, but they represent the first
line of defence, especially in handling severe motherboard problems.
Cont.…
• If the POST test finds a problem which is severe enough to keep the
system from operating properly, it halts boot up of the system and
produces audio beeps and/or error messages.
• You can find the meanings of these error signs in documentation from
the system manufacturer — this is often required for an accurate
understanding of audio beeps and messages.
• But if the POST fails, then at least you know the problem is hardware-
related.
The boot up process
• Carefully watching the steps in the boot process can reveal a lot about
the nature of problems in a system.
• By doing this you can include or exclude various possible causes of
faults.
• The absence of one or more of the following during booting can
indicate a fault:
✓When power is applied, the power supply fan should work.
✓The keyboard lights should flash as the rest of the system components are
reset.
✓A POST memory count and test should be seen.
✓A BIOS message should be visible on the monitor.
Cont.…
✓The floppy drive access light should come on briefly.
✓The hard disk access light should come on briefly.
✓An audible short beep should be heard.
✓The floppy disk access light should come briefly before a check of the hard
drive starts.
✓An operating system prompt, message, or logo should be visible.
• By observing the above sequence you should be able to work out
where the problem might be, that is, isolate the fault.
• For instance, if any of the above steps (except the last one) fails in
some regard, then you know the problem is hardware-related.
Hardware toolkit
• What equipment are you likely to need when carrying out the fault-finding? The
most useful tool, that you should never leave home without, is a good quality
Philips-head screwdriver. However, other tools in your hardware toolkit may
include:
✓screwdrivers — a full set and range of sizes
✓anti-static strap
✓pointy-nose pliers
✓multimeter
✓known good components such as video card, cables, mouse, hard-disk drive,
network interface card (NIC), CD-ROM drive
✓serial and parallel loop-back connectors
✓boot disks
✓a range of testing software for the loop-back plugs and NIC
✓a POST card.
Cont.…
• A POST card is a device that plugs into an empty slot in the
motherboard.
• When the system boots up, the card runs a series of diagnostics.
• In some cases these cards replace the normal functions of the BIOS.
The great advantage of using these cards is that you do not have to
resort to software running off the hard drive or a floppy disk.
• POST cards are normally used when systems are ‘dead’, or when the
system cannot read from the hard drive or the floppy drive.
• Some POST cards also come with a series of light emitting diodes
(LEDs) that produce coded error signals that you could interpret
together with a manual. Other cards produce audio beep signals.
Cont.…

Fig.5. A POST card

Diagnostic software
• There is a wide range of diagnostic tools available that can help you
identify all sorts of computer problems.
• Generally, the diagnostic software used for testing system components
and/or performance falls into two categories:
❑ Generic
❑ Pproprietary.
Generic diagnostic software
• The generic tools available are usually sold as software packages and
are very limited. To evaluate the usefulness of generic software you
have to assume that the software supplier has tested their software
with all original equipment manufacturer (OEM) hardware and
software you may want to test. This is hardly possible, so you should
not be surprised when the results of such packages fail to live up to
expectations.
• This is not to say the providers of these packages are supplying a
defective product — just that they can really only test the
functionality of devices and systems functions known to the
program writers and this will exclude many proprietary devices. A
good example of this would be network interface cards (NICs).
Cont.…
• Most of the generic diagnostic packages will probably be able to determine that
the NIC is installed in the system, however, if the exact functionality instructions
of the NIC are not built into the diagnostic software, an accurate result will
probably not be achieved.
• This is better understood when considering that the same NIC OEM may provide
a different diagnostics utility with each of many similar NICs. If the wrong
version of the diagnostics utility is tried with the wrong NIC, even the most basic
loop-back test will often fail. This highlights the need to pay close attention to the
software provided by OEMs.
• Another factor to consider is that general diagnostic software uses information
gained from the underlying operating system, which gains its information from the
BIOS. This means that the operating system (using its device driver) may not be
able to accurately provide information on the device. If the information provided
to the diagnostic software is not extremely accurate, then valid test results are not
a reasonable expectation. For truly accurate testing of a device you should use the
software provided by the manufacturer or use specialised diagnostic software that
can bypass the operating system all together.
Propriety diagnostic software
• If diagnostic software is operated on its own specially designed operating
system, then direct access (via the BIOS) to the hardware will likely yield
accurate and thorough details. Having unimpeded access to the low-level
functions of the hardware means the diagnostic software is able to run
rigorous testing and reporting.
• After all, it’s unlikely that rigorous memory testing could be performed
while there are several other programs currently running in memory.
• Two examples of good diagnostic software are:
❑Micro-Scope Diagnostic Suite from Micro2000
(http://www.micro2000.com/)
❑PC Certify Lite from Pro Tech Diagnostics
(http://www.protechdiagnostics.com/)
Common diagnostic tools
• All operating systems come with utilities that are used for general
checking, repair and reporting of faults.
• Each operating system is different but they do have some tools in
common such as hard disk scanning tools, e.g.
• Scandisk for Microsoft
• fsck (file system check) for Unix clones like Linux
• Disk First Aid for Apple MacIntosh systems.
Leave a trail to follow
• Apart from the use of diagnostic software (and a little trial and error)
examining documentation on past faults and their solutions can reveal a lot.
To give two examples:
• The introduction of a new device (say a different brand of NIC) may bring
with it some configuration problems. If there is documented evidence of
previous problems, together with information on how the problem was
solved, you can more efficiently remedy similar problems. The remedy may
come in the form of an automated configuration file (script), or a decision to
purchase less trouble-prone devices.
• The reports generated by some help desk software may indicate that a
particular user consistently experiences problems with certain devices, or
software applications. This information could then be used to reduce future
incidents of support calls by providing or recommending targeted training
for that user. The term currently in use for this is ‘just in time training’ (JIT).
Finding more information
• Experience, a logical approach and reviewing previously documented
problems and solutions, can form the backbone of being able to analyse and
determine the cause of system faults. However, there is always something
new and no single person or group can form the only valid source of good
information. There will always be a need to perform some research as a
result of some fault or problem caused by hardware or software. The
Internet provides other sources of information that can be relied upon.
• When you do research on the Internet, you should check the source of the
information, and bookmark (add to favourites) those that you consider valid
and useful.
• In the Research section of this Learning Pack you’ll find many websites to
carry out further research of the topics discussed here, including hardware-
related and software-related websites.
Summary
To be able to analyse and determine system faults you need to:
• have a general knowledge of how computers and individual
components should work
• use your senses to take in all the factors involved, eg sights, smell and
sounds
• take a logical approach using questioning, flow charts, checklists and
general guidelines
• have a good toolkit of both hardware and software tools
• use documentation on previous problems or faults generated by help
desk software.
Cont.…
You should also:
• know the factors that can adversely affect computers and components,
in order to prevent further problems occurring
• know where to find troubleshooting information at short notice
• document the fault and the resolution before the job can be considered
as finished
• keep calm while everything fails around you!
Identifying Vendor documentation, peer
organizations or research information