DG Fuel Consumption Calculator

Consumption follows load, not rating

A generator does not burn a fixed amount of fuel. Consumption follows the power it is delivering, through a figure called specific fuel consumption — litres per kWh generated:

kW output = kVA rating × pf × load fraction
litres/hour = kW output × SFC

The curve is nearly flat above 75 % and climbs sharply below 40 %, because friction, pumping and auxiliary losses are roughly constant while the useful output shrinks. Every unit you generate at 20 % load costs about half as much again as one generated at 80 %.

These are representative figures for a modern industrial diesel. Engine make, age, altitude, fuel quality and how well it is maintained all move them, and an old set out of tune can be 15 % worse. Measure yours across a month — litres delivered against units generated — and use that number instead.

What a generated unit really costs

The comparison people want is against the grid rate, and it is not close. In most markets diesel generation lands somewhere between three and five times the industrial tariff.

That number is fuel only. The full cost of running on DG also includes:

• Lubricating oil and filters, on hours-based intervals.
• Service labour, and major overhauls at a few thousand hours.
• Capital recovery. The set has a finite life measured in running hours; every hour spends some of it.
• Fuel losses — evaporation, spillage, and the difference between what was delivered and what reached the tank.

A reasonable rule is to add 15–25 % to the fuel figure for a fully loaded cost per unit. If you are comparing DG against a solar or storage investment, that is the number to use, not the fuel cost alone.

Load factor, and the trap of a comfortable size

Oversizing a generator feels safe and is expensive twice: once at purchase, then continuously in fuel and maintenance.

Below about 30 % load a diesel runs too cool to burn its fuel completely. Unburnt fuel and carbon accumulate in the exhaust and turbocharger — wet stacking — injectors glaze, bores polish, and the engine progressively gets worse at the thing it was running badly at to begin with. The fix is either a smaller set, or periodic load-bank exercising at 70 % or more to burn the deposits off.

Aim for 60–80 % load in normal running. If your load varies too much to hold that, two smaller sets in synchronism beat one large one: run one at a healthy load factor most of the time and bring the second in at peak. It costs more to buy, less to own, and gives you redundancy a single set cannot.

Size the set properly — including the motor-starting case, which usually governs — with the DG sizing calculator.

Reducing the bill without buying anything

1. Correct the power factor. An alternator is rated in kVA at 0.8 pf. Poor power factor hits the current limit while the engine still has kW to give, so you are carrying capacity you cannot use. Beware fixed capacitor banks on a lightly loaded set though — see APFC panels.
2. Shed non-essential load before starting. Every kW you do not carry on DG is roughly four times the grid cost saved.
3. Consolidate running hours. Two hours at 70 % beats four hours at 35 % for the same energy, and does the engine good.
4. Meter the DG feed separately. Until you know units generated against litres consumed, you cannot tell an efficiency problem from a fuel-loss problem. The metering is straightforward — gateway build.
5. Compare against solar. On a site running a diesel through daylight hours, solar competes against the DG cost per unit, not the grid tariff — which is why the payback is far shorter than the usual arithmetic suggests. Try it with the solar sizing calculator.

The costs that are not diesel

Fuel is the largest line and it is not the only one, and a cost per unit built from diesel alone understates what generated electricity actually costs by a meaningful margin.

Servicing is the predictable part. A diesel set needs an oil and filter change every 250 to 500 running hours, coolant and air filters on their own intervals, and a top overhaul somewhere around 10000 to 15000 hours. Spread across the units generated in that period, routine maintenance commonly adds 8 to 15 per cent to the fuel cost, and more on a small set where the service interval is short relative to output.

Then there is the set itself. A generator has a finite life measured in running hours, and every hour of operation consumes a fraction of the capital. On a set that runs regularly, depreciation per unit generated is frequently comparable to the maintenance cost — which is why a plant deciding between running its own generator and paying a demand charge should compare against the full figure, not the fuel figure the operator quotes from the day tank.

Where the fuel goes that you did not buy

Measured consumption exceeds calculated consumption on almost every site, and the gap is rarely the engine.

Density is the honest part of it. Diesel is bought by volume and burnt by mass, and it expands with temperature — roughly 0.08 per cent per degree Celsius. Fuel delivered at 40 degrees and consumed from a tank at 30 contains about 0.8 per cent more energy per litre than the delivery note implies. Small, but it runs the opposite way to what most people assume.

The rest is losses. Evaporation from a vented tank in Indian summer conditions is real and rises with tank ullage, so a large tank kept half-empty loses more than a small tank kept full. Water accumulates from condensation on the same cycle and has to be drained. And on many sites the difference between the fuel purchased and the fuel accounted for is simply pilferage, which is why an hour meter reading compared against a consumption calculation is one of the more useful routine checks a facility can run — a persistent gap of more than 5 to 10 per cent is worth investigating rather than absorbing.