In my last blog posts, I talked about how Power Quality affects the entire electrical infrastructure and that it’s important to consider every system element needed to keep your facility on line when voltage sags and other short term disturbances – including the cooling / mechanical side of the system – rear their ugly heads. Fortunately, a good power monitoring system that includes software-based tools and meters with built-in intelligence will keep you ahead of the game.
A few years back, one of our customers had a hard voltage sag in their system. Their UPS worked perfectly and handled it just fine, but they still had to take their data center off line. Why? Because while their UPS system kept the servers running, their chiller went off line and the cooling to the entire data center was lost. The controls on their mechanical system were not designed to ride through voltage sags, and their design did not include any UPS-based control power or ride-through. Another customer had a large enterprise lab go down for a similar reason – they had a single-pole 20 amp / 120 volt branch breaker trip, which just happened to feed the control circuits on their chiller plant. These events are sticky situations, because not only does the data center go off line, but the servers get very hot, very fast. It’s definitely worth considering how your entire system, and not just the IT load, will respond to events.
What really matters is keeping your facility running
Advanced PQ meters and power monitoring software have the potential to tell you more about the quality of your power than ever before. Is there a useful way to provide a summary view, or is what really matters what happens on each individual event? For the data center an index report like EN50160 can be helpful, but the bottom line is that your facility needs tools that can both provide a summary view of what happens over time, as well as capture, define, and notify staff when a serious, ‘off line’ incident occurs. The tried and true CBEMA and ITIC curves, which plot each individual event against a curve of duration and amplitude, is a very useful report that tends to carry the weight of industry credibility when it is shared with your utility. In my experience you will see a dramatic decrease in monthly power quality disturbances.
Power monitoring with analytical intelligence is important
In a power monitoring system, intelligence is more than just an email or SMS message notifying you when an event occurs. For example, a GPS-based satellite time clock that synchronizes all the clocks on your power quality meters and status monitoring devices enables you to go back and understand exactly what happened when, should your system respond to a hard sag or outage by transferring a UPS to battery, a static transfer switch to an alternate source, or event by starting and transferring the load to an emergency backup generator. Software based tools that enable you to compare power quality disturbances with equipment status changes are a critical resource to making sure your system is responding properly. Use a power quality meter with as much on board intelligence as possible. It helps. For example, some power quality meters include algorithms that compare the voltage and current waveforms to determine whether the source of the disturbance was upstream or downstream of the meter.
For more information, download this FREE white paper, and discover the ways power monitoring systems provide insight into the operation of your data center infrastructure.