On Nov 21, 10:57*pm, (csar244)
I completely agree with what you are saying, BUT in my guy's
defense, he did check the power supply before he changed out.
The old power supply LED light was on, but he did not get any
readings from it. *The computer worked after that, but that
is when it started having the problem. *It would work well for a few
days and then it would suddenly shut off. *
He checked the power supply. Fine. He might have checked it out
using standard technical knowledge, or he might have checked it out as
so many computer assemblers do. IOW since we don't know why - don't
know what he did - then his conclusion provided little to work with.
That is the principle. To know something such as what he knew, we
also must know why. Without whys, little is known that is useful.
A defettive power supply controller can cause a power supply to act
defective. Did he know of other power supply 'system' components? Or
a connector problem is temporarily corrected by connecting a new
supply. Again, without numbers, nobody can be helpful.
Do not assume you know what those numbers will report. Definitive
information is necessary to fix intermittents. A most definitive fact
are numbers. Those numbers mean zero to you and yet may be crying
out, "Fix me." Others would understand that language. Later, those
numbers provide you with a useful conclusion AND a grasp of what the
system does - what those numbers are measuring.
As you have discovered, a "power supply too large" is typical of
many recommendations. Until we have numbers - especially when the
system is not working - then no one will provide a useful reply.
Without any doubt, question, or hesitation, have and be using a 3.5
If we have to replace the supply, you need the meter to confirm the
new supply works properly. Even a new and defective supply can boot a
computer - which many do not understand.
Industry standard supplies use industry standard wire colors. This
two minute procedure describes how to get useful numbers from each
wire both before and when the power switch is pressed. Also provided
is a diagram showing where those wires connect to the motherboard
connector - in case your supply uses a non-standard color scheme:
"When your computer dies without warning....." starting 6 Feb 2007 in
the newsgroup alt.windows-xp at:
Connector chart to locate each color:
Best time to measure orange, red, yellow and purple wire voltages is
when a CPU is multitasking to all peripheral - has all peripherals
drawing power simultaneously. IOW playing complex video graphics (ie
a movie) while searching the hard drive, while downloading from the
internet, while playing a CD-Rom, while soundcard makes noise, while
reading a floppy, etc. Now those voltages are ready to be measured to
three significant digits.
The point is to establish power 'system' integrity (more than just a
power supply) before moving on to other suspects. Idea is to find a
problem before fixing anything. Fixing something before collecting
facts can exponentially complicate this problem.
Also useful are to check for some rare problems; easily identified.
For example, examine the tops of electrolytic capacitors on the
motherboard - especially around the CPU. All should be cylinders with
perfectly flat tops. Of course a CPU fan starts everytime. Rarely, a
defective fan does not start until pushed with a finger.
And finally, the power LED light can light even when voltages are
completely defective. The light only says a power cord is connected
and (depending on what that light does) that the supply does something
in response to a pressed power switch.
Meanwhile Compaq is from HP. Therefore the manufacturer has
provided only what responsible manufacturers provide - comprehensive
hardware diagnostics. Find them. They may prove useful later. These
diagnostics are often found on the hard drive, on a CD-Rom, and from
the HP web site.