I think you are getting terrible advice here.
> OK, you are talking VA, I was alking watts. If you look at
> the UPS ratings
> it will have both with watts usually being half or less
> than the VA rating
> and the usual math will not get the equivelant VA to watts
> rating.
Everyone better be careful here.
VA is the volts times amperes. It is not watts, and connot
be turned to watts unless you know power factor.
Power factor on a cap input power supply is terrible, and
the better the supply components the worse it gets. Power
factor is caused by the short period of load on the mains as
the capacitors charge on every half cycle. The peak current
can be 50 amps or more with a good quality supply that draws
only 15 amps! That means the VA load would be 50 times 240,
or 12 kVa for a 3600 watt amplifier load.
The second issue everyone is ignoring is bounce of the load.
Most UPS supplies are desiged for slow load changes, not
repeated abrupt bounces from no load to full load. Some
handle it OK, some don't. They often use a 150 or 300V dc
supply that they switch as a stepped square wave to simulate
the RMS value of a sine wave. The regulation is on the dc
supply, and when you bounce the load off and on like CW the
regulator is too far upstream to hold the 150/300 dc steady.
As I say, a small ALS600 with choke input (so PF was good)
set one on fire for me. An FET in the switching bridge
actually arced through to the heatsink and ignited the
insides.
>> That's a 1000VA unit.
>>
> Which might be good for 400 or 500 watts.
Of nearly pure resistance or steady load. Not charging a big
old capacitor bank 120 times a second for 45 degrees of the
power mains cycle or less. The better the supply components,
the worse the power factor. The exception is choke input.
Also, look out for the stepped square wave they call a
modified sine...and the pulsed load.
IMO it's not a good idea at all for a big amp with stiff cap
input supply. Been there, tried it.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|