Watching Residential Photovoltaic Installations Increase

 By Doug Houseman, Sean Morash and Jens Schoene

In the 1970s installing residential photovoltaic (solar) was an expensive mess. Seldom did a solar array exceed one kilowatt (kW) in size and most were smaller than a single kilowatt. Through the 1990s solar arrays on residential structures stayed at one to two kilowatts, as cost and complexity kept installations low. Over the last decade the average size of solar arrays on residential structures has grown.

A net-zero building with solar is feasible, in which the solar array over the course of a year equals the total consumption of electricity. Average consumption of a residential customer at Detroit Edison in 2015 is about 8,320 kilowatt-hours (kWh), which means in the Detroit area solar installation for net-zero is about 7.5 kW. Installing a 7.5 kW or larger solar system is within reach of some residential users, with costs dropping, leasing more prominent and product standardization.

While a sub-kW installation offered fewer challenges for customers or utilities, multi-kW arrays present challenges. The assumption in the rest of this article is that the array listed is isolated from all other solar installations and there is no interaction between solar installations.

The authors created a multi-tab spreadsheet that takes hourly data (available from the authors via email) that looks at a number of characteristics of solar in Michigan, with the capability to substitute data from other areas. From the analysis, it is possible to create the numbers in the table below and create an analysis in the columns to the right.

Starting with the sub-kW solar installation and working up in increments to 30kW (the largest request put to a local zoning committee recently in Michigan by a homeowner) the spreadsheet shows few issues for the grid owner in the smaller arrays; this underscores that utilities tended to have no trouble with the installations in the 1970s through the 1990s. As solar installations grew larger, interaction with the grid increased, and as the density of solar installations grew, the interaction between both the grid and other arrays grew. The table looks at only the size-based issues with solar installations. This list should not be used in place of a qualified inspector approving the final installation.

The table below is the summary of the information developed in writing this article. It all pertains to the data used from the Detroit area.

  • Array Size is the size in kilowatts of the installed array. This is the size that was input into PVWatts to create the hourly profile for production.
  • Total Production is the total number of kWh the solar system would have produced over the course of 2015
  • Hours of export are the total number of hours that the solar system overproduced compared to the hourly consumption of an average home in Michigan. The total number of hours of production was 4,495, so even a 30 kW array did not export every hour it produced energy.
  • Maximum Demand on the Grid is in kW and is bidirectional. The maximum is labeled with either an “I” for import or consumption and with an “E” for export or over generation. With a 5-kW array the demand placed on the grid exceeds an average residence with no solar, even though the dwelling is not yet at net-zero. Demand on the grid only grows from there as the array size increase. As the arrays decrease in size, there is almost no impact on demand, 2 kW and smaller has almost no impact on demand on the grid infrastructure compared to no solar installation.
  • Net Generation in kWh is the total amount of power that is generated in excess of the demands of the residence that the system is installed on. NA in the table indicates that the total amount of energy generated was less than the total consumed on an annual basis.
  • Total Export in kWh is the total amount exported by the system. Even at sizes as small as 1 kW the systems made some use of the grid as storage. Even at 5 kW more than half of the electricity produced is exported to the grid only to be re-imported later to the meet the consumption needs. In Michigan the energy created in April and exported is not needed until July, meaning that is not purely a daily mismatch but rather it is a seasonal mismatch. A total of 207 kWh from a 5 kW array is spring over generation that ends up consumed in July and August, a five-month shift.
  • Physical Changes to check for at the residence are listed in the order they happen, whenever the reader moves down a row, all of the items listed above should be included in the potential items to check. Older homes in this case are homes that were built through about 1970, many of those homes had services that were 60 amps. In many cases these homes have received larger 200 amp meters, but may not have had any of the other infrastructure items changed.
  • Physical Changes to check for on the local grid is similar to the prior column. While small arrays that are isolated offer little in the way of need for physical changes, as the size grows, the needs may grow.
  • Potential Residential Impacts list items the occupant of the residence may see in terms of issues. These issues if they arise should be brought to the attention of the leasing company or the installer.

Not included in this table are the impacts on the grid. That is a subject of further inquiry for another article.

*The authors would like to thank DTE for this information.




Doug Houseman, VP of Technical Innovation at Enernex and chairman of the IEEE PES Intelligent Grid Coordinating Committee. Doug has a broad background in Utilities and Energy. He worked for Capgemini in the Energy Practice for more than 15 years. During that time he rose to the position of CTO of the 12,000 person practice. During that same time Capgemini grew from less than $10 million dollars in Energy related revenue to more than $2.4 billion. Doug was part of the Global leadership team and worked all over the world in a thought leadership and delivery role. During that time Doug founded the Smart Energy Alliance, lead the Distribution Roadmap 2025, and developed the smart metering and smart grid practices. Doug and his sons are makers and tinkers who have built everything from a 3D printer to a greenhouse. Doug started using photovoltaics in the 1970s and geothermal in the 1980s.



Sean Morash rejoined the EnerNex Smart Grid Engineering team in August 2015, having completed an internship at the EnerNex Smart Grid Labs in 2012. Sean produces solutions through research based on a working knowledge of Smart Grid related applications, including communication technologies and protocols, advanced sensing and control, renewable energy, electrical, mechanical and information systems integration, enterprise information architecture, cyber security, information modeling, and related disciplines and methodologies.



Jens Schoene, Ph.D. is the Director of Research Studies at EnerNex. His areas of expertise are transient and harmonic analysis of power systems, distributed generation interconnection studies, induction studies and lightning studies. Jens was a research assistant at the High-Voltage Laboratory, University of Paderborn, Dept. Soest, Germany, where he designed and built a generator to simulate multiple lightning surge currents and voltages. His work at the University of Paderborn earned him the Adam-Herbert Award for outstanding accomplishments in the lightning protection area.

IEEE Smart Grid Newsletter Editors

Past Issues

To view archived articles, and issues, which deliver rich insight into the forces shaping the future of the smart grid, please visit the IEEE Smart Grid Resource Center.

IEEE Smart Grid Newsletter Compendium

The IEEE Smart Grid Newsletter Compendium "Smart Grid: The Next Decade" is the first of its kind promotional compilation featuring 32 "best of the best" insightful articles from recent issues of the IEEE Smart Grid Newsletter and will be the go-to resource for industry professionals for years to come. Click here to read "Smart Grid: The Next Decade"