Robert M wrote:
Hi everyone,
APC said step wave UPSs _can_ cause problems for computers that have PFC power supplies. Apparently, PFC power supplies are more sensitive than non-PFC power supplies to fluctuations in the waves...
APC didn't mention anything about initial power draw. Just the sensitivity to non sine wave output.
I get the confusion now. We're addressing the same thing from two different perspectives.
This is APC's official statement on the matter of PFC power supplies:
http://nam-en.apc.com/app/answers/detail...-their-use
Relevant excerpt:
Although computer power supplies draw only a fraction of their full capacity during it’s steady state(normal operation), PFC power supplies have the potential to draw their full capability during initial inrush. "Inrush" or "Inrush Current" refers to the maximum instantaneous input current drawn by an electrical device when first turned on.
A computer’s power supply may also be subjected to a period of inrush, while the UPS is changing state (switching from utility power to battery power and back). Back-UPS and Smart-UPS SCs may experience up to an 8ms transfer time during this period. This is just long enough to remove power from the PFC power supply, resulting in a momentary inrush of the PFC. Once the UPS changes states from "Online" (passing utility power) to "Onbattery" (passing power from the UPS's internal battery), the momentary inrush from the attched equipment subjects the UPS to the PFC power supply’s maximum power draw, resulting in a potential Overload condition or dropped load.
And this is their white paper on the subject:
http://www.apcmedia.com/salestools/RMUZ-..._R1_EN.pdf [pdf]
Notice that they discuss overloads, not wave-forms. The overload is the primary concern. The waveform effects the type of behavior that you might expect after an overload. But the overload is the really bad and easily preventable thing.
When there's an overdraw condition, if the timing is perfect there's the potential for a tiny gap in power in a stepped sine wave product where a pure sine wave product will simply provide less power at that moment. Instead of a brownout you get a blackout.
But it's not gonna happen unless you exceed the normal operating capacity of the UPS.
It's an issue with servers because of redundant power supplies. Peak draw from one PSU on startup may be far less than the UPSs rated capacity, but the second power supply -- which ordinarily draws negligible power -- may double that peak draw when first powered up. If you don't factor in the second PSU when calculating the capacity of the UPS for your rack, you could blow the whole rack.
But it's not an issue if you've calculated the capacity of the UPS properly. And the rule of thumb that says to get a UPS with twice the capacity that you think you'll need covers that situation very well.
