computer quit, couple questions - Page 2

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View

Re: computer quit, couple questions

Quoted text here. Click to load it

  Common sense without underlying knowledge makes one his own fool.
For example, what happens when all outputs from a powered power supply
are shorted together?  Naive know that a power supply is damaged.  But
even 30 years ago, a short circuit must not cause power supply
damage.  Short all the outputs from a power supply to create no damage
and no human threat.  Why?  Because anything dangerous is elsewhere.

  Why does Mark N use profanity?  A symptom of those who somehow know
before learning facts.  Mark N actually thinks a meter will cause
damage or harm humans?   Its instructions were too complex?  Even
cavemen had no problem reading.  Simple.  Set meter dial to range (ie
20 VDC).  Touch probes to wires.  Read numbers.  Even Geico is more

  How can one damage a computer?  Disconnect things - also called
shotgunning.  Disconnect nothing inside a computer when using a
meter.   Just another reason why a meter is a preferred solution.   No
one needs profanity to know that.

  Those who promote fear also would know an Ipod is too complex.
After all, a 3.5 digit multimeter is only sold in K-mart, Lowes, and
Wal-mart because - as Mark N implies - one must be a genius to use
one.  Those who first learn before posting know why that meter is even
sold in Radio Shack and Tru-value hardware stores.  It is a tool as
'dangerous' as a screwdriver.  And sometimes sold on sale for $10.

  What is wrong with friesian's computer?  Two minutes with a meter
provides useful numbers to learn and to post for useful replies.  Back
in cavemen days, we kids used VTVMs.  Even VTVMs were safe and simpler
than an Ipod.  We cavemen did not fall for myths, fear, and profanity
promoted by Mark N.  No mocking.  No insults.  Just reality.
Apparently Mark N warns us to not buy a too complex and too dangerous
Ipod.  Others instead learn from reality; gain experience and
knowledge; don't fear.

Re: computer quit, couple questions

On 21 Mar 2007 03:03:12 -0700, ""

Quoted text here. Click to load it

The very condensed version, enough to get you started-

Any cheap multimeter is sufficient for the basics, accurate
enough to take n.n Volt readings.  Pick what seems
appropriate to your budget, they can be had for about $10-15
at many hardware stores, some general purpose superstores,
or even cheaper online (though shipping begins to erode the
cost savings if it's a one-item order).

Any multimeter will have voltage reading capability.  Set it
to that DC voltage scale with the (dial?) method available.
If there are multiple increments in DC voltage, instead of
an "autoranging" meter (which costs slightly more), set it
to the double-digit setting, often 20V.  The idea is that
whatever you maximum anticipated range is (which in this
case, in a PC the highest voltage is about 12V), you want
the multimeter setting immediately larger than that range,
the closest one to it.

With it set to DC you have red, positive lead and black,
negative lead.  For our purposes, the negative is also
ground, because all the voltages from a PSU are referenced
from a common ground, and the system case metal is also
(supposed to be) at this same ground potential (voltage ~

Anything you measure, will have the multimeter negative lead
touching (or clamped to, if your meter has attachments like
an alligator clip), in firm contact with conductive metal.
I write conductive because a case that's painted,
clear-coated, any other kind of covering may interfere.  If
the surface is in question, put bot meter leads on it and
see if it reads very close to 0 volts.  Ideally the
negative, ground point would be taken from the black wire
connector on the plug going to a load (powered device), but
for basic measurements where a lot of accuracy isn't needed,
any free plug will suffice.

Be sure your multimeter is in voltage reading mode before
proceeding, a common mistake some people make is to have
formerly used their meter to read current, then when they
being to read voltage they forget to change the dial.
Current reading mode causes a very low resistance connection
between the two leads, nearly a short-circuit, which could
damage the meter or at least blow it's fuse when talking
about higher current available in a PC.

In voltage reading mode put the red meter probe on the
target point to measure.  For exampe a yellow hard drive
connector socket for 12V or a red (wired) connector socket
for 5V.  On some OEM systems the colors are different,
occasionally you might find blue corresponding to 12V on a
PSU plug for a drive, but the coloring used should be
consistent, and you can tell what the voltage should be by
observing the polarity of the plug and what it's plugging
into, what voltage was supposed to be on that plug position
for the device... and since we know a hard drive uses 5V and
12V, if the connector were facing with the tapered corners
facing upwards, and looking at it from the pin side, the 12V
socket would be on the left side of the plug.

For an ATX motherboard connector, see this picture,
On the newer 24 pin connectors there will be addt'l 3V, 5V,
12V, and ground, but you need not be concerned about the
extra 4 pins, the first ten columns will still correspond to
the ATX picture linked above and the remaining 4 are
*redundant*, only addt'l lines from the same voltage source
in the PSU.  If your ATX connector doesn't match the
picture, you might have a proprietary PSU like those
sometimes used by Dell, though as briefly mentioned above,
occasionally only the colors are changed and with your
multimeter, you can compare what voltage should be at what
position on the ATX connector to see if a connector with
different colors has the correct ATX voltage pin-positions.

Measurements of voltages are then straightforward when
system is turned on.  With system off, measurements of PS-On
and 5VSB should be taken.  System off means plugged into AC

Having gotten this far, we are brought to the point where
numerous other usenet posts describe the tolerances for
voltages, what levels you should have and what might
indicate a problem.  With the system unplugged from the ATX
connector, all other things unplugged from PSU as well, and
a paperclip or similar shorting the PS-On pin to ground, and
a minor load plugged in (like a hard drive or optical drive,
ideally something of low value when questioning the ability
of a PSU to work properly), the PSU should start with fan
spinning and roughly appropriate voltages on all pins - but
some like the -5V, -12V (if these are present, modern
systems don't use them anymore), and sometimes 3V, may be
off-spec some as they won't have a load on them from just a
Hard or Optical drive.  Also note that a PSU that is failing
might be able to power such a slight load, but fail when
pressed for more power from a whole system - whether it be
due to a component failure in the PSU, a component being
powered BY the PSU, or general insufficiency of the PSU to
produce enough power, IOW a PSU incapable of supplying
enough current.  Such a latter state could mean a PSU
initially works to power a system but has a short life once
degraded from it's new state, it then begins struggling more
at this overload condition.

A basic power-on and voltage test cannot find all faulty
PSU, some may appear to be working properly still with such
a test, but any that appear to be improperly or not working
at all, would tend to indicate the PSU has failed.  There
are a few exceptions to this, but they are beyond the
intended scope of a usenet reply, being situations present
to uniquenesses in particular PSU or systems, they're
atypical, arguably nonstandard parts combination problems
where the problem would tend to surface at first attempt to
use such parts, not a problem that only appears later on.

Site Timeline