Measure generator natural gas usage by subtracting two gas-meter readings taken over a recorded operating period, then divide the volume by elapsed hours. To attribute that volume to the generator, account for every other appliance supplied through the same meter.
Open the Natural Gas Generator Consumption Calculator.
What does a gas-meter reading actually measure?
A meter records accumulated gas volume, while a consumption rate describes volume per unit of time. Those are different quantities: a change of 300 cubic feet during two hours means an average rate of 150 cubic feet per hour, usually written 150 ft³/h or CFH.
Read the unit printed on the register before taking any numbers. A display may represent cubic feet, hundreds of cubic feet, or cubic metres, and a small test dial may have a separate value per revolution. Do not assume that one digit equals one cubic foot because the meter belongs to a familiar utility.
National Grid’s meter-reading guide explains digital registers and alternating dial directions. Use your supplier’s instructions for the actual meter, especially when a pointer appears exactly on a number. A photograph of the entire register preserves the units and dial positions for later checking.
Billing energy is another step beyond volume. A therm represents 100,000 Btu, but the Btu contained in each cubic foot depends on gas composition and billing corrections. Keep the original volume record even if the bill presents only energy units; otherwise a later conversion becomes difficult to audit.
Build a log before running a measurement
A useful observation can be made without opening the meter enclosure, adjusting a regulator, or modifying any gas pipe. Choose a normal, authorized operating period and record information from accessible displays. Leave installation and gas-supply concerns to the relevant qualified service provider.
Create a row for the start and another for the finish. Include the exact time, full register reading, displayed units, generator status, and major electrical loads. Record whether heating, hot water, cooking, or another gas appliance operated during the interval.
| Log field | Illustrative entry | Why it matters |
|---|---|---|
| Start reading | 12,480 ft³ | Baseline volume |
| Finish reading | 12,780 ft³ | End volume |
| Elapsed time | 2.00 hours | Converts volume to a rate |
| Average generator output | 12 kW | Allows a load comparison |
| Other gas use | 24 ft³ estimated | Separates shared consumption |
| Operating notes | Loads cycled normally | Explains variation |
Use actual elapsed minutes divided by 60. A run from 14:10 to 15:40 lasts 1.5 hours, not 1.3 hours. Decimal hours are a frequent source of error because clock minutes use a base of 60.
Make the test long enough that register resolution is small compared with the volume used. A meter that advances only in coarse units may give an unreliable short test. Extending the observation is usually more useful than reporting extra decimal places from an uncertain reading.
Calculate total and generator-only consumption
For an illustrative two-hour observation, the meter moves from 12,480 to 12,780 ft³. The building therefore uses 300 ft³ during the observation:
Total rate = (12,780 − 12,480) ÷ 2 = 150 ft³/h
Suppose a separate, reasonably matched baseline suggests that other gas appliances used 24 ft³ over those same two hours. The generator attribution becomes:
Generator volume = 300 − 24 = 276 ft³
Generator rate = 276 ÷ 2 = 138 ft³/h
The arithmetic is exact for the entered numbers, but the attribution remains an estimate. A water heater that did not fire during the baseline might fire during the generator observation. Subtracting the baseline does not erase that uncertainty.
If other appliances can remain inactive using their ordinary user controls without compromising essential heating or safety, a dedicated observation can be easier to interpret. Do not extinguish pilots, operate unfamiliar service valves, or disable protective controls to improve a spreadsheet result.
Where unrelated gas use is large or unpredictable, record the result as whole-building usage during generator operation. A professionally installed dedicated submeter can offer better attribution, but its design and installation are separate from a simple reading exercise.
Compare the reading with the calculator
The embedded calculator estimates volume from electrical energy, assumed electrical efficiency, and gas heat content. Its relationship is:
Gas volume = kW × load fraction × hours × 3,412 ÷ efficiency fraction ÷ Btu per ft³
The EIA energy-conversion page gives the standard conversion of one kWh to 3,412 Btu. That converts electrical output into an energy unit compatible with the fuel input; it does not mean all fuel energy becomes electricity.
For a separate illustrative estimate, enter a 20 kW generator, 60% average load, two hours, 30% electrical efficiency, and 1,000 Btu/ft³. Electrical output is 24 kWh. The resulting volume is:
24 × 3,412 ÷ 0.30 ÷ 1,000 = 272.96 ft³
That corresponds to about 136.5 ft³/h. The example’s adjusted meter result of 138 ft³/h is close, but the agreement does not validate a 30% efficiency assumption for all operating conditions. Several uncertain inputs can partly offset one another.
Use the heating-value basis supplied for the generator fuel data when available. Mixing a lower-heating-value efficiency figure with a higher-heating-value gas value creates a systematic mismatch. If the basis is unknown, document it as an unresolved assumption instead of adjusting efficiency until the figures agree.
How do changing loads affect the result?
The percentage input should describe average electrical output during the measured interval. A 20 kW generator producing an average 6 kW is operating at 30% of that rating, even if it briefly supplied 14 kW when a compressor started.
A rated capacity is not a measurement of energy delivered. Leaving the calculator at 100% because the engine ran continuously can greatly overstate consumption. Conversely, estimating load only from devices that happen to be running at the finish may omit heating or pumping cycles earlier in the interval.
When a suitable installed energy monitor provides accumulated output, divide kWh by operating hours to obtain average kW. Use a display or approved monitoring arrangement; gathering this data does not require opening energized equipment. If only an appliance inventory is available, identify duty-cycle assumptions explicitly.
Repeat observations at different recognizable load conditions rather than combining everything into one unexplained average. A low-load evening, a pumping interval, and an air-conditioning interval may produce different rates. Each observation can be valid for its own conditions without defining the generator’s complete fuel curve.
Do not extrapolate a single rate down to zero load. An engine normally still consumes fuel while running without useful electrical output, whereas the calculator’s simple efficiency model approaches zero fuel as calculated output approaches zero.
Turn readings into an outage budget
Once a measured rate is representative, multiply it by expected operating time. At an illustrative 138 ft³/h, eight hours uses 1,104 ft³. If that gas is assumed to contain 1,000 Btu/ft³, the corresponding energy is 11.04 therms.
Apply the appropriate variable gas charge only after checking the bill’s units. A price per therm cannot be multiplied directly by cubic feet. Fixed monthly charges do not become an additional operating cost merely because the generator starts.
For a changing outage schedule, calculate separate blocks. For example, four hours at 138 ft³/h and six hours at a separately measured 95 ft³/h total 1,122 ft³. A ten-hour calculation using the higher rate would be more conservative, but would also conceal the planned load reduction.
Record a range when the other-appliance correction is uncertain. If the two-hour interval’s unrelated use could plausibly be 10–40 ft³, the generator estimate is 130–145 ft³/h. That range is more informative than an unsupported figure such as 138.00 ft³/h.
Troubleshoot an unexpected reading
First check transcription and scale: register multiplier, metric versus imperial units, elapsed time, and an overlooked decimal. Then review unrelated appliances and generator load. A change in household heating can easily make two whole-building observations incomparable.
Also check whether the generator spent part of the interval warming up, exercising unloaded, or cooling down. These periods consume gas while contributing little or no useful delivered energy. Include them when estimating total outage fuel, but distinguish them when comparing steady-load performance.
A meter log cannot verify gas pressure under load, pipe capacity, combustion quality, or fuel-system condition. If operation is unstable or the result differs materially from the model’s published data, give the service provider the readings, times, loads, and observed symptoms. Those details are more useful than an unexplained claim that the unit uses too much gas.
Can the utility bill replace a timed reading?
A bill can help check long-term totals, but its interval normally includes unrelated appliances and periods when the generator was off. It is best used as a reasonableness check unless generator operation dominates a carefully documented billing period.
Should startup consumption be ignored?
Include startup and cooldown when budgeting a real operating session. For a steady-load comparison, begin and end the selected measurement interval consistently and clearly state which operating stages were included.
References
- National Grid: How to Read Your Meter
- EIA: Energy conversion calculators
- Natural Gas Generator Consumption Calculator