AnonLinuxUser wrote:
On 1/4/2020 11:51 PM, William Unruh wrote:
On 2020-01-05, AnonLinuxUser wrote:
On 1/3/2020 8:54 PM, nospam wrote:
In article , vallor
wrote:


I have built Preamps and power amps of both types and prefer the
solid
state ones hands down.

i also did long ago. i never much cared for designing analog
circuits.

A good portion of solid state circuits are analog.

a much larger portion of them are digital.


Not in the sense of power amplification. A lot of integrated circuit
designers do the analog work inside the chip that goes out to the power
amps. The power amps are analog and will always be analog.
Just less analog design to do at that point.

Well, class D is sort of digitial (pulse width modulation).


The only thing that would be classed as I know is some IPS units that
put out staircase power for PCs... I don't like them because of the
problems they created for wifes iMac... makes it run funny at times.
Best not to introduce spikes to the iMac.

You're referring to a UPS (uninterruptable power supply).

They come in modified sine and pure sine.

https://www.researchgate.net/figure/...fig2_261048059

You will also find a lot of fanciful waveform diagrams,
not all of which are correct.

https://imeon-energy.com.au/wp-conte...-sine-wave.png

*******

Inverters produce square waves (cheap ones, from the past).
And in the dim and distant past, square waves were OK for
a lot of the "naive" loads, like a light bulb perhaps.

*******

Modified sine and square waves were bad for
some of the first Active PFC ATX power supplies.
You could get "coil noise" from the front end,
indicating it wasn't happy or wasn't the right
design for it.

Later, the Active PFC front end was changed (somehow),
to make these supplies less upset by non-sine signals.

The first PFC was designed on an algorithm of
"the current draw waveform should look like the
voltage waveform from the line". This was intended
to reshape the current draw waveform, as well as
bring it into phase with the voltage waveform.

If you were at the campsite, and connected a
desktop PCs to a car battery plus a 200W inverter
(square wave), the desktop PC with APFC would go nuts.
In some cases, the PSU would switch itself off. Or,
it could make "weird noises" in the coils because
of the PFC trying to make a square current waveform,
from a square voltage waveform.

The modern designs must be doing something different.

Summary: If you know a PC has a power supply which
is bad for this sort of thing, stick with
pure sine UPS boxes.

This one, for example, runs *two minutes* at full power
according to the advert text near the bottom of the
main pane.

So naturally, you're not expecting miracles for $190/$210
or whatever the real price is.

The spec tab gives a different number.

https://www.newegg.com/cyberpower-cp...-133-_-Product

AVR
VA Rating 1350 VA Watts 810 Watts
Full Load: 3 Minutes
Half Load: 9 Minutes
10.40" x 3.90" x 14.20"
Weight 20.30 lbs.

This one is $369.99. You're paying by the pound. The
formfactor of the battery, may be easier to find a knockoff
battery for it later.

https://www.newegg.com/apc-smc1500c-...=pure+sine+ups

AVR
VA Rating 1440 VA Watts 900 Watts
Full load: 4 Minutes
Half load: 11 Minutes
8.45" x 6.70" x 17.30"
Weight 45.00 lbs.

You can test this with 9 x 100W lightbulbs, because the
filaments will be hot during switchover, and there won't
be nearly as large a load surge as with cold light bulbs.
900W of incandescent light bulbs would draw 1800W when cold.

Neither unit mentions switching time, which is likely
to be 8msec to 15msec. These aren't double conversion supplies
with zero switchover. Finding a cooling fan on a UPS, is
a hint it's a double conversion type.

Paul