Hardware faults are generally attributable
to component malfunction or component failure. Electronic components do not
generally wear out with age but they become less reliable at the end of their
normal service life.
It is very important to realize that
component reliability is greatly reduced when components are operated at, or
near, their maximum ratings. As an example, a capacitor rated at 25V and
operated at 1OV at a temperature of 20°C will exhibit a mean-time-to-failure
(MTTF) of around 200000 hours.
When operated at 40°C with 20V applied,
however, its MTTF will be reduced by a factor of 10 to about 20 000 hours.
TIP: The mean-time-to-failure (MlTF) of a
system can be greatly extended by simply keeping it cool. Always ensure that
your PC is kept out of direct sunlight and away from other heat producing
sources (such as radiators). Ventilation slots should be kept clear of
obstructions and there must be adequate air flow all around the system
enclosure. For this reason it is important to avoid placing tower systems under
desks, in corners, or sandwiched between shelves.
Hardware fault, what to do
If you think you have a hardware fault, the
following stages are typical:
1. Perform functional tests and
observatioos. If the fault has been reported by someone else, it is important
to obtain all relevant information and not make any assumptions which may lead
you
along a blind alley.
2. Eliminate functional parts of the system
from your investigation.
3. Isolate the problem to a particular area
of the system. This will often involve associating the fault with one or more
of the following:
(a) power supply (including mains cable and
fuse)
(b) system motherboard (includes CPU, ROM
and RAM)
(c) graphics adapter (includes video RAM)
(d) disk adapter (includes disk controller)
(e) other IjO adapter cards (e.g. serial
communications cards,
modem cards, USB devices, SCSI devices,
etc.)
(f) floppy disk drive (including disk drive
cables and connectors)
(g) hard disk drive (including disk drive
cables and connectors)
(h) keyboard and mouse
(i) display adapter
(j) monitor
(k) external hardware (such as a printer
sharer or external drive)
(I) communications or network problems
4. Disassemble (as necessary) and
investigate individual components and subsystems (e.g. carry out RAM
diagnostics, gain access to system board, remove suspect RAM).
5. Identify and replace faulty components
(e.g. check RAM and replace with functional component).
6. Perform appropriate functional tests
(e.g. rerun RAM diagnostics, check memory is fully operational).
7. Reassemble system and, if appropriate,
‘soak test’ or ‘burn in’ for an appropriate period.
TIP: If you have more than one system
available, items such as the system unit, display, keyboard, and external
cables can all be checked (and eliminated from further investigation) without
having to remove or dismantle anything. Simply disconnect the suspect part and
substitute the equivalent part from an identical or compatible system which is
known to be functional.
No comments:
Post a Comment