Thread: No power ...
View Single Post
  #37   Report Post  
Posted to rec.crafts.metalworking
Robert Nichols Robert Nichols is offline
external usenet poster
 
Posts: 101
Default No power ...

On 01/20/2016 08:59 AM, Ed Huntress wrote:
On Wed, 20 Jan 2016 08:48:09 -0600, Robert Nichols
wrote:

On 01/19/2016 04:44 PM, Jim Wilkins wrote:
I took a chance on the HF 2200W inverter generator #61169, $399.99
with a coupon. Hopefully it will be unnoticeably quiet from the street
under my usual wintertime load of less than 200W and able to run the
A/C or microwave if we have a summer hurricane outage.


Good luck trying to start an A/C with that little inverter generator.
I also see no mention that the output is true sine wave, so some
computer power supplies might not be happy with the output waveform.


I'm curious about that. I've never investigated those power supplies,
but I thought that the output from battery batckup, like my APC, was
square wave or something jagged, anyway. I've never put a 'scope on
mine but have thought about doing it.

Anyway, for anyone experienced with these power supplies, what's the
story?


Modern computer power supplies with "Active PFC" (Power Factor
Correction) aren't rated to run with PWM (Pulse Width Modulated) square
wave input. You see, with the old power supplies that just rectified the
line voltage straight into a big capacitor, all the current is drawn as
a big spike when the AC waveform rises above the voltage on the
capacitor. That current waveform has an RMS value that's way above the
average current. It's similar to the situation with a cheap electric
motor with a lousy power factor -- you're drawing a lot more
Volt-Amperes than you are Watts.

Power supplies with Active PFC have input circuitry that tries to make
the input current more closely track the (presumed sine wave) input
voltage. Feed one of those with a PWM square wave and it's trying to
track the fast rising leading edge of that pulse. That puts you back to
the current being drawn in big spikes, but that input circuitry wasn't
designed to handle that. Of course if you've got a 700W power supply on
a PC that's actually using only 120 Watts you should be fine. If you're
pushing the power supply anywhere near its ratings, you could fry it.
You need a power source that more closely simulates a sine wave.

--
Bob Nichols AT comcast.net I am "RNichols42"