Power Consumption in kilowatts of over one month

  • Thread starter Thread starter Paul Samiljan
  • Start date Start date
Gary W. Swearingen said:
FWIW (not much), until a few months ago I long thought that modern
switching power supplies were about 90% efficient, and was suprised to
read somewhere (as you have confirmed) that PC PSUs are considerably
less efficient than that.
Click to expand...

Many single output, higher voltage (i.e. 24v output) switching power
supplies are 90% or greater in efficiency. But the output diode drops
in the high current 3.3v and 5v outputs of ATX PSUs make the 90% figure
unobtainable. Add the -5v and 5vsb circuitry, and 70% is a good figure.
 
Many single output, higher voltage (i.e. 24v output) switching power
supplies are 90% or greater in efficiency. But the output diode drops
in the high current 3.3v and 5v outputs of ATX PSUs make the 90% figure
unobtainable. Add the -5v and 5vsb circuitry, and 70% is a good figure.
Click to expand...
ROFLMAO. I don't know what books you're reading but it helps if you
actually understand whats printed.
 
Conor said:
ROFLMAO. I don't know what books you're reading but it helps if you
actually understand whats printed.
Click to expand...

Any particular portion of the above draw your ire, or are you just showing
your overall ignorance again? The lower the output voltage, the higher
percentage the power dissipated by the output diode is when compared to
the output power. For example the 0.6v forward drop times 100 amps in a
5v 500w output is 60w, but the 0.8v drop times 21 amps in a 24v 500w output
is only 16.8w. Is it any wonder that the 24v supply would be more efficient?

Does math scare you that much?

[Redfaced Conor sighted again!]
 
ric said:
'Tis also correct.
Click to expand...

Re efficiency,
http://www.tomshardware.com/howto/20030609/power_supplies-02.html
says: "we tested [...] from 64% to 78%." And the very-relevant
http://www.cepe.ethz.ch/download/staff/bernard/28_formated.pdf
says "The efficiency of computer power supplies operated at least at
20% of their nominal power lies between 60% and 80%." And I've
seen several claims of 78%-79%.

As for 90% being "unobtainable", I don't know much about switching PSU
design, but I'm guessing that output diodes could be replaced by
switched FETs (which would require more control circuitry and $) to
bring the figure above 90%.
 
Back
Top