In the 1970s, the most common application of photovoltaic ("PV") technology was off-grid. But today, more than 95% of solar installation are on-grid, or "grid-tied." Certainly falling PV costs have helped accelerate grid-tied solar, but the other driving factor was the adoption of Net Metering legislation by 40 US states, including California.
Net Metering is an electricity policy for utility customers who operate their own on-site "self-generation" power systems such as photovoltaic systems. PV systems are connected to the utility grid via the customers' main service panel and meter and, when generating more power than is needed at the site, return excess electricity to the grid through the power meter, reversing the meter from its usual direction. As a result of the meter working in both directions – one way to measure power purchased (when on-site demand is greater than on-site power production), the other way to measure power returned to the grid – the customer pays the "net" of both transactions.
In California, there are two important characteristics of Net Metering policies that are particularly relevant to solar power-owning utility customers.
Retail Price Credit for Power "Sold"
When excess electricity is fed through the meter and returned to the utility grid in California, the customer will receive full retail price as credit to be applied to future purchases. To facilitate this, Net Metering customers are on an annual billing cycle as opposed to the usual monthly billing cycle – and the credits received are reconciled with the purchases made annually on the anniversary of the local utility approval of the system installation.
Only Generate What You Need
At the time of this annual reconciliation, there are three possible outcomes. First, if the utility customer has purchased more power than they've returned to the grid, they will be billed for the "net". Second, the "net" could be zero, although this is unlikely since power use varies somewhat from year to year as does weather which impacts PV system performance, so the odds are pretty low that the bill would be exactly zero.
The third possible outcome, which we do our best to prevent, is that the solar power system generates more power than needed in total over the course of twelve months and there's excess credit. This could happen if the PV system is over-designed or if the customer's usage declines significantly. Regardless of the reason, if there is excess credit on account, as per California's Net Metering policies the credit will be kept by the utility company and the twelve-month clock starts anew.
Time of Use Metering Can Often Multiply Savings
Given utility companies in California must credit net metering customers at retail rates for power returned to the grid and that solar-generating customers can reconcile these credit and debits annually, many solar-generating customers can take advantage of their utility company's "Time-of-Use" rate schedules to increase the value of the power sold during peak generating times.
For example, Pacific Gas & Electric has a commercial rate schedule that charges as much as $0.32 per kilowatt-hour from noon to 6 PM weekdays from May through October and rates as low as $0.09/kWh at other times. This rate schedule would normally only appeal to businesses that either use little power at the peak rate times or could be flexible to "demand shift" their usage away from the higher rate exposure.
But for solar-generating customers, this rate is very appealing, because the peak rate period coincides quite well with the period of peak solar power generation. In this case, customers "sell" power at higher rates which increase the value of their PV systems, and can enable them to buy less PV equipment and improve their return on investment.
As the determination of the "net" needs to be done on a hour-by-hour basis, with 8,760 data points per year, the analysis of usage patterns and comparison to PV output must be done carefully to avoid under-sizing PV systems which will leave customers exposed to high peak rates.
Sunlight Electric uses a proprietary software model (with all 8,760 data records – it's a 6.3 MB Excel file) to compare hour-by-hour usage with hour-by-hour production so we can truly assess the benefits of Time-of-Use metering for each customer. In cases where complete data are not available on the usage side, we use our growing database of usage patterns as a proxy and carefully review our assumptions in the proposal so everyone's clear on how we arrived at our recommendations.