Watts to Amps Calculator

The formula#

DC circuits — the simple case, no power factor:

I = P / V

Single-phase AC:

I = P / (V × PF)

Three-phase AC:

I = P / (√3 × V × PF)

Where P is real power in watts, V is volts, PF is power factor, and I is current in amps.

Worked examples#

A 1500 W heater on a 230 V single-phase supply. A heater is a resistive load, so the power factor is 1:

I = 1500 / (230 × 1) = 6.5 A

Comfortable on a 16 A socket circuit, and the reason a 6 A socket is not enough for one.

The same heater on a 120 V supply:

I = 1500 / (120 × 1) = 12.5 A

Halve the voltage and the current doubles for the same power. On a North American 15 A branch circuit, that single heater occupies 83 % of the circuit before anything else is plugged in — which is why the 12 A / 1440 W limit on portable heaters exists.

A 1500 W motor at 0.85 power factor, 230 V single-phase:

I = 1500 / (230 × 0.85) = 7.7 A

Same watts, 18 % more current than the heater, because the power factor is below 1.

Appliance chart at 230 V#

Current drawn by common single-phase loads, with the power factor each type actually runs at:

Two things this chart does not show. Motor-driven appliances — fridges, air conditioners, washing machines — draw a starting current several times these figures for a second or so. And an inverter air conditioner varies its consumption continuously rather than cycling, so its running figure is whatever the compressor is doing at that moment, not a fixed number.

Watts, VA, and the UPS trap#

This is the conversion that costs people money.

A UPS is rated in VA, apparent power. Your equipment is rated in watts, real power. They are not the same number, and the ratio between them is the power factor:

VA = W / PF

A 600 W load at 0.6 power factor needs 1000 VA of UPS. Buy a "1000 VA" UPS for a 1000 W load and it will overload immediately — that unit is designed for about 600 W.

Older UPS units state a power factor of 0.6, newer ones 0.8 to 0.9. The nameplate always shows both figures; use the watts one. And leave headroom: a UPS running at its rating runs hot, and hot is what kills the battery.

The watts to VA calculator does this conversion directly.

Three-phase, and the size of the difference#

A 15,000 W load:

• Single-phase, 230 V, PF 0.9: 15000 / (230 × 0.9) = 72.5 A
• Three-phase, 415 V, PF 0.9: 15000 / (1.732 × 415 × 0.9) = 23.2 A

A third of the current for the same work, spread across three conductors instead of one. This is why anything much above 5 kW is supplied three-phase — the cable, the switchgear and the losses all scale with current, not with power.

Sizing the circuit around the answer#

The current is where the design starts:

Continuous loads get 125 %. A load running more than three hours continuously — heating, lighting, a compressor on a duty cycle — is conventionally sized at 125 % of its current. Our 6.5 A heater wants a circuit rated at least 8.1 A, so a 10 A device.

The cable must survive the device, not the load. If the protective device is 16 A, the cable has to carry 16 A after derating for its ambient temperature and how many other cables share its route. Cable sizing handles the derating; MCB and MCCB sizing handles the device.

Volt drop over the run. On anything longer than about 20 m, volt drop rather than heating usually decides the size. Limits are 3 % for lighting and 5 % for power. The voltage drop calculator covers it.

Common mistakes#

Assuming PF 1 for everything. True for heaters, kettles, filament lamps and geysers. Wrong for motors, air conditioners and anything with a switch-mode supply, and it understates the current by 10 to 40 %.

Using the wrong voltage. 230 V line-to-neutral for single-phase, 415 V line-to-line with √3 for three-phase. Mixing them up is a factor of nearly two either way.

Reading the surge rating from a nameplate. Some appliances state peak input rather than continuous. If the figure looks unusually high for the appliance, check whether it is a surge or a sustained rating.

Adding up nameplate watts for a whole installation. Nothing runs everything at once. Connected load times a diversity factor gives maximum demand — see connected load and diversity for the factors that apply.

The neutral carries the single-phase current too#

On a single-phase circuit the current calculated here flows in the line conductor and returns, in full, through the neutral. That is obvious stated plainly and is routinely forgotten in two places.

The first is cable selection. A reduced-neutral cable is legitimate on a balanced three-phase circuit and is wrong on a single-phase one, where the neutral is a full current-carrying conductor and has to be the same size as the line. The second is switching and protection: the neutral must be isolated by the same device where the installation requires it, and a single-pole breaker on a circuit that needs double-pole isolation leaves the neutral live relative to earth when the circuit is supposedly dead.

Where a distribution board mixes single-phase circuits across three phases, the neutral current at the board is the vector sum rather than the arithmetic one, and on a well-balanced board it is small. That cancellation is the reason a common neutral is acceptable at all — and the reason it stops being acceptable where the loads are electronic, because triplen harmonics add in the neutral instead of cancelling.

Continuous loads and the 80 per cent rule#

Protective devices are rated for the current they can carry indefinitely under standard test conditions, and those conditions are more generous than a real enclosure. Where a load runs for three hours or more at a stretch, common practice — explicit in NEC and implicit in the derating tables elsewhere — is to size the circuit at 125 per cent of the calculated current, which is the same as loading the device to no more than 80 per cent of its rating.

Water heaters, air conditioners, EV chargers, lighting circuits and anything on a process that runs a shift all qualify. A 3 kW geyser at 230 V draws about 13 A, and the circuit for it should be built around 16.3 A rather than 13 — which is the difference between a 16 A device that will spend its life near its limit and a 20 A one that will not.

The reasoning is thermal rather than bureaucratic. A breaker in a full panel, surrounded by other breakers, in an Indian summer, is running considerably hotter than the bench it was calibrated on, and its actual trip point drifts down accordingly. The margin is what stops a correctly sized circuit becoming a nuisance-tripping one in July.