[j-nsp] EX4550 true power consumption

[Michael Loftis]

On Thu, Oct 24, 2013 at 3:02 PM, Chuck Anderson <cra at wpi.edu> wrote:
> On Thu, Oct 24, 2013 at 09:38:43AM -0700, Michael Loftis wrote:
>> I don't know anyone that assumes that the peak capability of a PSU
>> (especially in network gear) is it's actual consumtion, but thats just
>> me.  I do agree I wish they'd publish at least approximate figures.
>> It can be a deciding factor in a lot of environments today.  Keep in
>> mined 650W is high-line mode (~220/240VAC) - in 120V it's only going
>> to be around 325W.
>
> This deserves clarification.  The power usage in Watts for a given
> hardware configuration should be the same either way--it is just that
> in high-line mode 208/220/240V, you will draw less current in Amps, by
> about half for twice the voltage (ohms law approximately, with a Power
> Factor Correction thrown in), and because of this the power supply is
> designed so that you can consume up to 650W of power only if you use
> high-line mode, and if you max-out the supply in low-line 120V mode
> you will only be able to consume up to 325W.  This implies that if you
> use low-line mode, you won't be able to install as many
> modules/optics/whatever-else-uses-power before hitting the max or
> losing power supply redundancy (being that both PSUs will be required
> to power everything).

Yeah sorry, I was quoting the PSU wattage ratings from memory at the
input voltage ranges, not usage.  So yes, barring efficiency
differences between 120V (low-line) and 240V (high-line) "mode" the
wall usage is, of course, the same.