What It Takes To Create An Off-Grid Household In The Bay Area (California) Using Rooftop Solar & Battery Storage Only (Exclusive)

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

By Indradeep Ghosh, PhD
Cupertino, CA

Indradeep Photo thumbWith the recent announcement of the Tesla Powerwall battery pack, many articles have been written about the possibility of homeowners using it to defect from the grid and live off the grid using rooftop solar generation and battery storage for the times when the sun doesn’t shine. Though such an idea seems very attractive in theory, it comes with significant real-life challenges, especially in a world envisioned to be powered by renewable energy only. This article takes a closer look.

In order to investigate such a scenario, let us start with the energy usage profile of a net-positive household in Bay Area, California – the Energy+ household. This house has been covered in a local newspaper article few years back. It is a standard, all-electric, energy-efficient, 2200 sq. ft. house built to current code. It houses a family of four – two working adults and two kids. For the last four years, this household has remained net energy positive in day-to-day living using an 11 kW rooftop solar array and two EVs – a Chevy Volt which maintains 96% EV mode and a Ford Focus Electric. All EV charging is done at home. As a result, for four years running, this household has not burnt any natural gas, propane, wood, or gasoline in daily life. The local utility company PG&E sends a check to the family at the end of each yearly billing cycle for the excess energy generated.

Ener0 Ener1

Chip in a few dollars a month to help support independent cleantech coverage that helps to accelerate the cleantech revolution!

The complete yearly energy usage and generation profile of the above house on an hourly basis is available for download from PG&E’s customer account using their green button facility. The monthly and average hourly usage data for 2014 are shown in the above graphs. The data does have some unevenness to account for the five weeks when the family was away on vacations. This data is left unchanged as this also comprises the usual energy usage profile of the family on a yearly basis. The average hourly usage profile for the various months very closely resembles the “duck curve” that California ISO is so worried about. It is apparent from the graph that there are two different problems of energy storage that have to be solved here. The first is the daily storage problem whereby the solar energy generated during the day has to be stored for use during the night. The second more significant problem is the seasonal energy storage problem where the excess solar energy generated during summer is banked with PG&E to be used during winter. The instantaneous peak load of the house is 25 kW, which a couple of single-phase inverters should have no problem handling.

To figure out how much battery storage will be needed, a simple software program is implemented that simulates a battery over this usage data. The program simply accumulates the negative generation numbers in the battery until it is fully charged. When the load turns positive, it starts using up the battery until the charge goes to zero. Then this cycle is repeated. To take a house off the grid, the battery state of charge should never go to zero throughout the whole year. Otherwise, the house runs out of energy to use. The battery simulations are started from summer and the year is wrapped around to give the battery enough time to charge fully before winter use.

Ener2Since the household used about 13,500 kWh for the whole year, the daily household energy use is averaged and the simulations are first run using the average daily storage requirement of 40 kWh. The resulting hourly usage profile for the various months is shown in the adjacent graph. It is clear from the curve that though this daily storage solves the problem for all the summer months from May to October, it is completely inadequate to serve the winter where considerable power is still drawn from the grid. To investigate how much storage is needed so that no power is drawn from the grid anytime during the whole year, the battery size is gradually increased and the simulations rerun. It turns out that the battery size needed to go off grid is a whopping 3018 kWh. Such a battery from Tesla will cost close to a million dollars and probably occupy a couple of two-car garages by volume! Clearly, this brute force approach is impractical as it stands now. So, is going off grid only with rooftop solar even feasible in the Bay Area?


The obvious problem over here is winter, when the solar generation is inadequate to service the daily needs of the household. The only way to solve this problem is to reduce winter consumption and increase winter generation. It turns out there is a lot of scope for reducing winter consumption in the household. About 60% of the winter consumption goes into space and water heating for which electric heat pumps are used. The current ones in the house were installed in 2010. Since then, more efficient heat pumps have arrived on the market. Particularly, the efficiency of the heat pump water heater has increased from 240% to 324% (loosely speaking). The average efficiency of the heat pump space heater has improved from 300% to 400%. Also, the heat pump space heater has a nasty stand-by load of 60 W which makes it waste almost 500 kWh every year. However, the current mini-split heat pumps have practically reduced the stand-by load to zero. Finally, the electric clothes dryer in the house can be replaced with a heat pump dryer recently introduced to the US market. Such a dryer will reduce dryer energy consumption by 50%. If the above three appliances are replaced in the house, the winter energy consumption can be reduced by 25% and the summer consumption by about 15%.

Satellite image of the maxed-out roof.

To increase generation, the solar array has to be oversized. Up till now, solar panels were expensive and the idea was to just cancel out yearly energy consumption completely. From the above discussion, this sizing strategy will not work as it will be deficient in winter generation. Since winter insolation in the Bay Area is half the summer insolation, for an off-grid configuration, the solar array should be sized as double what was necessary before, to balance consumption on a monthly basis. This will lead to huge over-generation in summer, where generation will have to be curtailed by disconnecting the array at times. Also, there can be the issue of running out of roof area. As can be seen from the adjacent picture, the roof is maxed out. However, if the current 15% efficient solar modules are replaced with the 22% efficient ones currently available on the market, the solar array size can be increased to 16 kW.

These two strategies are implemented in the simulator whereby the load is decreased and the generation increased in the usage profile according to the above possible scenarios and the resulting hourly usage profile for the various months is shown again without any battery. It can be observed now that there is substantial solar generation even in winter to offset the consumption on a monthly basis and huge over-generation in summer.


Next, the battery simulations are again run on this new load profile until there is no power drawn from the grid at any time. This time, the battery requirement is a much more manageable 349 kWh. The final resulting hourly load profile is shown in the adjacent graph. The estimated current cost of such a system with a 16 kW highly efficient solar array is about $150,000 and hence out of reach of all but a few.

However, within 10 years, installed solar costs are projected to drop to $1/watt and Li-on battery costs to about $100/kWh. In such a scenario, about $50,000 will yield a system that will provide all household energy practically for life throughout the year and will not need a grid. The current 5,000 cycle capacity in the battery will be adequate for more than a 50-year life as most of the battery will remain unused most of time.

Also, an intelligent battery charge/discharge controller is needed to use the battery cells evenly over time. Such a system will be much more feasible and cheaper as we go towards the tropics, where seasonal insolation variation is less. Conversely, it will be much more difficult to pull this off as we go more towards the northern latitudes and alternative strategies for winter renewable energy generation and seasonal energy storage will be needed. However, based on the above analysis, the possibility of going off the grid with only rooftop solar and battery for a Bay Area household is a dream that seems to be within reach in less than 10 years.

Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.

Latest CleanTechnica.TV Video

CleanTechnica uses affiliate links. See our policy here.

Guest Contributor

We publish a number of guest posts from experts in a large variety of fields. This is our contributor account for those special people, organizations, agencies, and companies.

Guest Contributor has 4387 posts and counting. See all posts by Guest Contributor