As the Internet of Things increasingly generates data about the world around us, we see the rising profile of the communications-and-control systems that make sense of all that data. These systems are critical for driving improvements in productivity, quality and system-availability, which is why it’s increasingly important for companies to examine how they’re protecting their business operations and industrial processes from a very real danger: loss of power.
I like to break the environment down into two areas: “mains” and “brains”
- Mains power is the primary power feed to a system or appliance, often the only feed
- Brains power is a separate power feed to the intelligence of the system
Think of the furnace and thermostat in your home. You have the main power to the furnace and a separate, lower power feed to your thermostat. It would be costly to provide an alternative to utility power for your furnace. If, however, you have an Internet-enabled thermostat, it would be nice to retain access to that system (as well as other systems connected to your home network) during power losses or circuit interruptions. The cost of protecting this intelligence—the “brains” in the system—is much less than protecting the main power to appliances.
Now consider this arrangement with higher stakes. In an industrial environment such as a plant floor there are “mains” loads that warrant full power protection—any process or machinery that would cause significant harm if it were to shut down during a power outage. In a critical commercial environment such as a hospital it might take the form of an MRI machine, where power-outage damage is severe because of the cycle time and revenue cost (not to mention aggravating patients by forcing them to go through the MRI process again).
So what’s the fix?
Critical mains loads should be protected by an uninterruptible power supply (UPS) system capable of handling the electrical load on its own for a reasonable period of time. Typically, these are larger UPS systems, such as a 3-phase, 460-volt units.
In some cases, these UPS systems work in tandem with generators, providing power during transition time after a utility outage. In an industrial environment, “brains” include items like sensors and process-control systems (such as programmable logic controllers), as well as the networks that enable them to communicate. “Brains” might also include building-management/security systems, as well as PCs or servers attached to equipment that collects and controls processes.
And all of these things can go down when the power fails.
Most companies can’t cost-justify supplying main power UPS backup for all of their machinery, processes or business systems. However, the extent to which we rely on the intelligence in these systems means we may want to ensure that the “brains” stay up and running, at least for some time.
In the event of an outage, one or more relatively small UPS installations can enable these systems to keep functioning long enough to provide for effective system shutdown and to collect valuable data. In an industrial environment, this data can help determine the cause and time of the outage, enabling a faster restart. A single UPS may be able to back up a group of “brains” systems and some level of related network infrastructure.
A good example in almost any modern industrial or commercial environment is security systems. We increasingly rely on IP-enabled camera systems connected to/powered over Ethernet networks serving DVR equipment. During power outages, the value of these systems increases—the entirety of these systems constitutes “brains” applications that typically warrant UPS backup.
Examples of machines, processes and devices that can benefit from UPS backup are all around us, increasingly in this IoT era. Companies need to start asking hard questions about what happens to their systems when they lose power.
What does it take to get the systems back up and running?
What sorts of critical data may be lost?
Answering these questions is the first step in determining what level of risk is acceptable. Then, a more formal risk assessment might be justified, enabling you to accurately determine which systems—The mains? The brains? Both?—to back up.
Dan McGinn is director of business development for Secure Power Systems at Schneider Electric.