Computers should be turned off
I agree with your Publisher’s Letter in the January 2000 issue
on the issue of turning computer equipment off. Running at elevated temperatures
reduces the lifetimes of equipment. In fact I thought that the manufacturers
establish the MTBF of new equipment by running a sample of units at elevated
temperature, noting the failure rate and extrapolating the expected lifetime
from that data.
I have dimmers installed on several of my room lights at home
and I rarely have the lights up full and the light bulbs rarely blow. I was once
told that if you run an incandescent light at about one third of its rated
voltage its lifetime is indefinite.
Equipment often fails at turn-on but that doesn’t mean that it
failed solely due to power cycling. Sure, the final power-up broke the camel’s
back but not without the effects of ageing accumulated through various
Just last year I was asked to repair the power supply of a
work-station that had not worked after a power down. It was fairly old and
inspection showed that the filter capacitor in the switcher was as dry as a
bone; in fact, the can rattled on its innards.
This did not occur through power cycling; it was caused by
years of running in a hot environment and when the unit was eventually switched
off, it couldn’t handle the stress of being turned back on.
I always turn my PCs off if I am leaving the house for more
than 30 minutes and when I go to bed. Nor do I leave my PC on at work when I go
home. At the same time I don’t think it is good to be cycling the power
unnecessarily, so if I am working on a PC that I have to power up and down
frequently I try to connect the monitor so it is always on. I also reckon it is
false economy to turn the room lights off but leave the computer on, as most PCs
use more power than the lights. If a PC is not going to be used overnight then
it should be turned off.
Another issue with PCs left running is the fan. The bearings
tend to dry out and seize, then the temperature can rise beyond the endurance of
parts and they start dying.
There is always a risk of fire when electrical equipment is
left turned on. Just a few years ago I heard in the news of a woman who died in
Melbourne in a house fire started by a faulty monitor. I expect that the risk of
this is probably quite low, especially with equipment made by reputable
manufacturers, however we should not ignore the consequences.
In the end we need to strike a balance: we need to get good
lifetime from the equipment and we need to not waste energy for several reasons.
I think that if you turn your PC on you should leave it on for an hour, and if
you are not going to use if for a couple of hours, then turn it off.
One should also bear in mind that on hot days a PC would make
the room less comfortable but it can help warm a cold room in winter.
Stipulating that PCs should be left on all the time is a
simplistic approach; one that can prove fatal in extreme cases.
Neons don’t like the dark
I noted the letter in "Ask Silicon Chip" in the April issue
about an electric fence tester which works only when there is some ambient
light. For some time now, I have been intending to write a short note about this
problem, or rather the basis of it.
When we moved to our present address, in late 1966, there were
no street lights. I was working night shift and noticed that sometimes the
fluorescent light in the bathroom failed to light until I turned on a
battery-powered torch. Eventually, I came to recognise that it was only under
very dark conditions that the fluorescent light failed to strike.
Later, I noticed that the neon indicator on the controller for
an electric blanket showed much the same phenomenon. The blanket was a
commercially-produced device with a "thermostat" and a neon indicator which
should have lit whenever power was switched through to heat the blanket. In this
case, the neon indicator did not glow at first (sometimes) but lit normally
after a few minutes.
At first I suspected a loose connection, such as a dry-soldered
joint, but eventually came to the conclusion that it was similar to the problem
with the fluorescent light.
My younger son has been an electronics engineer for quite a few
years and when I mentioned the two examples above, he was able to relate it to a
machine he had to fix. It used a neon lamp as part of a timer circuit. After he
had made the necessary repairs, the control was adjusted to the desired time and
everything worked OK until the covers were replaced. This caused the time-out to
change. After a few trials, he came to the conclusion that light was affecting
the striking voltage of the neon gas discharge device.
North Mackay, Qld.
Hot wire cutter can be simplified
I just read the article on making a hot wire cutter in the
April 2000 issue and I thought I’d let you know how I made mine; I think it’s a
fair bit simpler.
For the power supply I used a dimmable 12V lamp transformer.
You can an usually buy a complete down-light kit for $20 at the hardware stores
and occasionally a transformer on its own for $15. For the temperature control I
used a standard lamp dimmer - I bought a dimmer, switch and switch-plate for $10
at K-Mart. For the cutter wire I bought nichrome resistance wire from Dick Smith
Switching off computers
I read your January 2000 editorial about "turning computers off
when not in use" and agreed totally. I was therefore astounded to read the
letter of rebuttal in the Mailbag pages of the March issue.
I completed an apprenticeship as a "Radio & TV Mechanic"
and have spent the past 25 years repairing and designing new electronic
equipment from consumer gear to industrial products and even pinball machines.
I am flabbergasted that a person from a TAFE college can assert
with total confidence that monitors will not catch fire. This is
Actually my pet hate is cheap equipment with shonky switchmode
power supplies and dubious mains wiring. This gear includes TV sets, computer
monitors and VCRs. I have seen dozens of units that have either caught fire or
were very close to doing so and this troubles me.
When I build any piece of mains-powered equipment my first
concern is that it is correctly wired, earthed and fused, according to good
240VAC practice. This seems to be little followed in some very popular and well
known makes of consumer goods. I am sure that some house fires could be traced
to dodgy TV sets running in "standby" mode".
To conclude, I agree 100% with your comments – "if it’s not
being used, pull the plug".