Deepdark said:
Do I need to take in count the negative rail ??
Yes, yo do!
No, it's not! W is active power, VA is apparent power, var is reactive power. Draw a square angle triangle and all that,
while all are volts X amps there are different timing consideration. VA is the volts your DMM reads times the amps your DMM reads (proper reading of course, sometimes TRMS needed, doesn't make much sense in DC). W is the used power in the system, referred as how much heat a resistive load will dissipate, independently if DC or AC. The var is the geometric difference between them and refers to the power that goes in and out of the system but never gets used. If your load isn't linear (guess what, it isn't) this last one stops making sense and there are others ways to call it.
Thanks. Is it safe to assume that:
DC Watt for the phantom power is about, in the worst case, 10ma x 48v = 0.48W x 2 = 0.96W AC
DC Watt for the bipolar rail = about 80ma x 15v = 1.2W x 2 = 2.4W
So 0.96W + 2.4W = 3.36W. 1W = 1VA so I need about 4VA. If I need to include the negative rail, so it would be about 6.8VA.
I should be allright with a 15v dual rail, 10VA toroidal from Antek, then. Right?
So, for a single channel, 10mA for the phantom, 80mA for each rail.
10mA x 48V = 0.5W
80ma x 15v x 2= 2.5W
(2.4W+0.5W) x 1.5* = 4.5VA
* YMMV, rule of thumb numbers depend on the person holding the thumb...
You are missing here the losses on the PS, and some cases you might want to be able to handle, I guess using the 10VA is the way to go unless some other constrain. That 1.5 can change, depending hugely on the PS design, there are some decent spread sheets out there to help with that, I guess 1.5 is the smallest number I'd use for a quick calculation, it might be short in some cases, if the PS isn't that efficient or you are over estimating your filter caps, for example.
JS