Installing generation behind the meter changes what you buy from the utility, but it does not change what the utility has to have ready for you.
The distribution transformer, the service conductors and a share of network capacity are still sized for your full load, because on the day the generator is down, that full load arrives from the grid with no notice. Standby tariffs exist to price that readiness, and for many on-site generation projects they are the difference between a case that works and one that does not.
The two services
Tariffs generally distinguish two things, though the vocabulary varies by jurisdiction.
Supplemental service is the ongoing grid supply for whatever portion of your load the on-site generation does not cover. A 2 MW site with a 1.2 MW generator takes about 800 kW supplemental in normal operation.
Standby or backup service is the supply that must be available when the generation is not — during a fault, and during scheduled maintenance. It is not usually consumed. It is reserved.
The charge structure follows the distinction. Supplemental is priced much like ordinary service. Standby typically carries a reservation component billed on contracted capacity whether or not it is drawn, plus a usage component when it is.
Why the reservation charge exists
It is the same cost causation argument that produces demand charges generally, applied to a customer whose consumption pattern conceals its capacity requirement.
An ordinary customer's peak reveals what the system must be able to serve. A customer with on-site generation has a metered peak far below their actual requirement, because the generator is covering the difference — until it is not. The network still has to be built for the full load. If that customer were billed only on metered demand, other customers would fund the capacity held for them.
Whether the reservation rate a given tariff sets is a fair measure of that cost is exactly what gets litigated in rate cases, and the treatment of distributed generation under standby tariffs has been an active regulatory subject for years. The answer differs by state.
The effect on a project case
This is where projects are won and lost, because standby charges are routinely omitted from the first version of the model.
On-site generation with and without standby charges
Annual figures for a site installing 1,200 kW of generation.
- Grid energy displaced by the generator7,400,000 kWh
- Grid energy rate avoided, all-in6.9 ¢/kWh
- (Gross energy saving)$510,600
- Fuel and operating cost of the generation$318,000
- (Net saving before standby charges)$192,600
- Contracted standby capacity1,200 kW
- Standby reservation rate$4.20 / kW-month
- (Annual standby reservation charge)$60,480
- Supplemental service demand charge, unchanged$0
Net annual saving after standby charges$132,120
Standby charges took 31% of the project's net saving. A model that omits them overstates the return by nearly half. Rates and quantities illustrative — take yours from the standby tariff itself.
Thirty-one percent is enough to move a project from approved to rejected, and it is entirely knowable in advance. The standby tariff is a public document.
What to check in the standby tariff
- How contracted standby capacity is determined — by nameplate, by contract election, or by measured demand during outages.
- Whether the reservation charge is billed monthly regardless of use, or only in months where standby is drawn.
- Whether scheduled maintenance outages are treated differently from forced outages, and what notice is required to qualify.
- Whether there is a limit on standby hours before a different rate applies.
- How supplemental service demand is measured, and whether it interacts with the standby determinant.
- Whether a ratchet applies to either determinant: ratchet clauses.
- Whether the tariff has an exemption or reduced rate for particular technologies or sizes under state rules.
- What happens if the generation is retired or removed — whether the account returns to a standard schedule automatically or by election: switching rate schedules.
The third item is worth real attention. Many tariffs price a scheduled maintenance outage more favorably than an unscheduled one, on the reasonable basis that a utility given notice can plan around it. Capturing that requires giving the notice the tariff asks for, in the form it asks for, and it is a saving that costs nothing but administration.
Sizing changes when standby is priced
Standby charges billed on contracted capacity change the optimal size of the generation, not just its economics.
A generator sized to cover the entire site load requires standby capacity equal to the entire site load, and pays a reservation charge on all of it. A smaller unit covering the base load leaves a portion served by ordinary supplemental service, with a smaller standby reservation behind it.
The efficient size is therefore not the largest one that fits. It is the one where the marginal energy saving from another increment of generation exceeds the marginal standby reservation charge that increment brings with it. That is an optimization worth doing properly rather than assuming, and it belongs in the same appraisal as the rest of the project: building the business case for demand reduction.
And the demand charge you still pay
On-site generation reduces metered demand while it is running, which does reduce an ordinary demand charge. It does nothing during an outage, and an outage that coincides with a peak period produces a metered demand at full site load — potentially the highest interval of the year, and under a ratchet, the floor for the year afterward.
That is an argument for treating outage timing as a demand management question rather than purely a maintenance one, and for understanding exactly which determinants your ordinary schedule bills: demand charges explained.
