PSU watts reading

  • Thread starter Thread starter Frank
  • Start date Start date
Frank said:
Anyone knows if it's possible to monitor or take a reading
of how much watts the psu is giving out. I have a 400W psu
and I would like to know if I'm running it at full capacity
or if it's just giving out 50% of its capacity.
Click to expand...

You could put a very low value resistor, in the order of .01 ohm, in
series with each voltage output and measure the voltage drop across it
and calculate the current, but that would require building an adapter
cable with those resistors. Or you could use a clamp-on current
sensor for each set of wires (red, orange, and yellow), but this can
be costly because magnetic current sensors for DC require linear Hall
effect sensors.

It would be simpler to just measure the power at the AC side with a
plug-in device like a Kill-O-Watt meter and assume that the power
supply is putting out 65-75% of its reading (the efficiency of the
average ATX supply). Or if the power suppy has active power factor
correction, you can simply separate the wires of a power cord, put a
clamp-on ammeter (AC-only type is fine) on one of the leads, and
multiply the reading by the voltage. Without active power factor
correction, the current reading can be 50% too high.
Reason is the fan pulls out air which is 2x hotter that my
case fan
Click to expand...

If it's 25C going in, 2X hotter would be 621C going out. But 50C
would be only .08x hotter.
 
250 watt power supplies remains, as posted, typically
sufficient for most every desktop computer. A simple wattage
reading can be rather difficult- unless one is aggressive.
Return to my original post. Get the many peripherals running
and measure DC voltages as defined earlier. The quickest way
to determine a sufficiently sized power supply was originally
posted.
 
Not todays modern desktop computers.
Click to expand...

It depends...

P4 Prescott, Radeon 9800, 3 hard drives, and you'll be wanting more than
300W.

Athlon 64, Radeon 9200, 1 hard drive, 250W is sufficient.
 
It depends...
P4 Prescott, Radeon 9800, 3 hard drives, and you'll be
wanting more than 300W.
Athlon 64, Radeon 9200, 1 hard drive, 250W is sufficient.
Click to expand...

You have to check the capacity on +3.3V +5V and +12V independent
and have the normal >30% margins on each, + margins for future
expansion.
 
Good luck doing a load analysis for each voltage. Load on
one voltage changes the power capacity of another output.
This being a function unique to supply design. Furthermore,
many clone supplies don't even provide valid power ratings for
their outputs since supply is marketed to 'computer
assemblers' who have insufficient grasp of electrical
principles. Tom's Hardware is often cited as the example of
power supplies that even self destruct before maximum rated
power is achieved. IOW PSU fails twice over. 1) does not
meet power specification AND 2) load causes power supply
damage. Each is an unacceptable failure. This typical power
supply will be proven sufficient load plus 30%? I don't think
so. So what is the OP to do?

Short of expensive equipment, the OP can only determine
power supply output by measuring voltages with the computer
operating all peripherals simultaneously. A procedure
originally posted at start of this discussion on 9 May 2004 by
this author.

In the meantime, systems designed by engineers (not by
computer assemblers) routinely provide more than enough power
with only 250 watts supplies. But those are supplies that
really do output 250 watts as rated. Kony has accurately
posted typical power requirements for some systems. But since
too many computer assemblers don't even know the real power
limits of their supply and don't understand basic electrical
principles (may even condemn the use of a multimeter), then
those computer assemblers also must resort to the classic
solution: Home Improvement's .... more power. They just
'feel' it solves the problem; multimeter readings and spec
numbers be damned.

As demonstrated by this discussion, most computer assemblers
would never know what power requirements are for their
computer - shortage of electrical knowledge and insufficient
test equipment. Nor would they really know what the maximum
output is for any DC voltage. One possessing knowledge of how
ATX power supplies work would also understand that maximum
load on any one output is significantly changed by loads on
other outputs. Each output is not independent. A
relationship unique with each manufacturer's design. Computer
user cannot determine the "capacity of +3.3V +5V and +12V
independent" especially if he cannot even measure the input
power.

Best solution, to repeat, is to fully load power supply with
all peripherals running. Then measure DC voltages with a
multimeter. After all the discussion, this remains the OP's
(Frank's) only alternative other than buying major, expensive,
test equipment.
 
Good luck doing a load analysis for each voltage. Load on
one voltage changes the power capacity of another output.
Click to expand...

Antec True Power don't have that problem.
Most other have.

Best solution, to repeat, is to fully load power supply with
all peripherals running. Then measure DC voltages with a
multimeter.
Click to expand...

I don't think so.
Then you don't know if you are near the limit on a output.
 
Good luck doing a load analysis for each voltage. Load on
one voltage changes the power capacity of another output.
Click to expand...

Antec True Power don't have that problem.
Most other have.

Best solution, to repeat, is to fully load power supply
with all peripherals running. Then measure DC voltages
with a multimeter.
Click to expand...

I don't think so.
Then you don't know if you are near the limit on a output.
The margins are very important for stability.
 
Back
Top