In smart grids, latency is the time it takes for datagrams to travel over a network link, typically one hop at a time. Latency is normally measured in milliseconds or even in microseconds depending upon the medium used to convey the data. The next generation smart grids are all about ultra low latency and coordination of systems aimed at dazzling new levels of customer service and performance. Like an aerial acrobatic team, smart grids operate at the scary edge of performance. With the right technology, processes, systems, and robust communications, situational awareness will drive better decision-making, albeit automated to the maximum.

Latency is one of the many critical parameters for smart grids. The metric is different for transmission grids compared to distribution grids. The time domain is one that often perplexes operators especially when the latency metrics are so fast that they defy human comparison to other time frames often found in life. Here are some examples of the time domain of everyday things.

Milliseconds Event
3 a housefly’s wing flap
4 typical average seek time for a 10,000 rpm hard disk
5 a honey bee’s wing flap
8 1/125 of a second (125), a standard camera shutter speed; fastest shifting time of a car’s mechanical transmission
16.7 (1/60 second) called a jiffy
16.7 (1/60 second) cycle time for North American 60 Hz AC mains grid
20 cycle time for European 50 Hz AC mains grid
33.3 the amount of time one frame lasts in 30fps video
41.708 the amount of time one frame lasts in 24fps video or film (actually 23.976fps for most films.)
50 cycle time for the lowest audible tone, 20 Hz
30 to 100 typical minimum latency for a broadband internet connection (important for online gaming)
100 the time interval between gear changes on a Ferrari FXX
100 to withdraw our hand from the stove
134 time taken by light to travel around the Earth’s equator
150 athletes and others who train themselves can achieve faster reaction times
150 recommended maximum time delay for telephone service
200 to stomp on the brakes of a car

In smart grids, we have a variety of latency requirements depending upon the grid type, architecture, technology, need, and application. Some applications like remotely controlling street lights have very relaxed latency, other applications like reading smart meters have slightly more stringent latency, and still other applications like grid automation and voltage optimization using IEC 61850 have ultra low latency.

SCADA systems that poll in 2, 4, or 8 second windows are easy to accommodate, even over long delay technologies like satellite links. But, a federated grid model for distribution automation with IEC 61850 may demand ultra low latencies for the initial GOOSE messages in the area of sub 10 ms or less than two cycles of power (60 Hz) at 33.34 ms.

The architecture of the control systems is important. Historically, all grid management and telemetry systems for the past several decades used centralized control and SCADA. So, latency was relaxed compared to the next generation of federated grid management whereby the controller and the intelligence is pushed out to the edge of the network and is collocated in the substation managing feeders and perhaps one or two other smaller substations and their associated feeders. These smaller serving areas or clusters are hyperactive and respond in incredibly low latency parameters firing off multiple GOOSE messages in many directions to trip switches, disconnect renewable energy sources, and acknowledge open / closed switch conditions.

There is a lot of autonomy in the next generation smart grids and this demands a distributed intelligence design with lightning fast communications over optical fibre or Layer 2 Ethernet wireless links. Many conventional communication network solutions will no longer suffice to meet the latency demands of these IEC 61850 managed feeders.

Telemetry is now flowing in from many devices to these IEC 61850 substation controllers, including telemetry from residential and C&I smart meters, reclosers, segmentation switches, power line monitors, transformers, and a myriad of other devices living on the distribution feeders. This telemetry is providing power quality measurements back to the IEC 61850 substation controller to optimize the voltage and control segmentation and tie switches. Feeders are now operated in many smaller sections with the ability to isolate and contain faults and power outages to reduce populations of the feeder connections thereby limiting loss of power to fewer customers for shorter durations. Looping feeders bridge together to back-feed customers downstream from a isolated segment. Enhanced customer service and higher service performance levels are the goal as well as reduce line losses.

Managing latency is therefore essential for all next generation grid management systems underpinned with robust communication networks able to accommodate ultra low latency. Grid operators will need to be keenly aware of latency issues and understand how to best manage their distribution grids in this new paradigm.


Michael Martin has more than 35 years of experience in broadband networks, optical fibre, wireless and digital communications technologies. He is a Senior Executive Consultant with IBM’s Global Center of Excellence for Energy and Utilities. He was previously a founding partner and President of MICAN Communications and earlier was President of Comlink Systems Limited and Ensat Broadcast Services, Inc., both divisions of Cygnal Technologies Corporation. He holds three Masters level degrees, in business (MBA), communication (MA), and education (MEd). As well, he has diplomas and certifications in business, computer programming, internetworking, project management, media, photography, and communication technology.