--- title: "VA to Watts Calculator" description: "Convert VA to watts using power factor. What a 1000 VA UPS or a 5 kVA stabiliser will really run, with a chart and the loads to check it against." standard: "W = VA × PF" source: "https://energycalchq.com/tools/va-to-watts" --- You have a VA rating on a nameplate — a UPS, a stabiliser, a small transformer — and you want to know what it will actually run. That answer is in watts, and getting there needs the power factor of the device itself. ## The formula ``` W = VA × PF ``` Where `VA` is apparent power in volt-amps, `PF` is power factor, and `W` is real power in watts. For the reverse — you know the load and want the rating — use the [watts to VA calculator](/tools/watts-to-va). ## A worked example A 1000 VA UPS with a power factor of 0.6: ``` W = 1000 × 0.6 = 600 W ``` So that unit runs 600 W of equipment, not 1000 W. If the nameplate already prints both figures — and it should — use the printed watts and treat this calculation as a check. ## What common VA ratings actually deliver | Rating | At PF 0.6 | At PF 0.7 | At PF 0.8 | At PF 0.9 | At PF 1.0 | | --- | --- | --- | --- | --- | --- | | 600 VA | 360 W | 420 W | 480 W | 540 W | 600 W | | 800 VA | 480 W | 560 W | 640 W | 720 W | 800 W | | 1000 VA | 600 W | 700 W | 800 W | 900 W | 1000 W | | 1500 VA | 900 W | 1050 W | 1200 W | 1350 W | 1500 W | | 2000 VA | 1200 W | 1400 W | 1600 W | 1800 W | 2000 W | | 3000 VA | 1800 W | 2100 W | 2400 W | 2700 W | 3000 W | | 5000 VA | 3000 W | 3500 W | 4000 W | 4500 W | 5000 W | | 6000 VA | 3600 W | 4200 W | 4800 W | 5400 W | 6000 W | | 10 kVA | 6000 W | 7000 W | 8000 W | 9000 W | 10,000 W | The spread across one row is the whole point. A "1000 VA" unit might run 600 W or 1000 W depending on nothing except its own design, which is why the VA figure alone is not a specification you can buy from. ## What that will actually power Taking a common 1000 VA / 600 W UPS: | Equipment | Typical draw | Fits in 600 W? | | --- | --- | --- | | Desktop PC + 24" monitor | 200 W | Yes, comfortably | | Two desktops + monitors | 400 W | Yes | | 1U rack server | 350 – 500 W | One, with little spare | | Network switch, 24 port PoE | 100 – 400 W | Depends entirely on PoE draw | | WiFi router + modem | 30 W | Trivially | | NVR + 8 IP cameras | 120 W | Yes | | Laser printer, printing | 900 W peak | **No** — never put a laser printer on a UPS | | 1.5 ton air conditioner | 1600 W | No | | Refrigerator | 200 W running, 1200 W starting | Marginal — the start is the problem | Two entries there deserve attention. A **laser printer** draws a large pulse when its fuser heats, and it will overload a UPS sized for the office it sits in — it belongs on a normal socket. A **refrigerator or any motor load** draws several times its running current at start, so it must be sized on the starting surge, not the running figure. ## VA to watts on a stabiliser Voltage stabilisers are also rated in VA, and the same conversion applies — but with a second consideration. A 5 kVA stabiliser at 0.8 PF passes 4 kW. However, a stabiliser's rating assumes its **input voltage is within its working band**. Most units state something like "140 – 280 V input", and their capacity falls as input voltage falls: at 150 V input the transformer is working much harder for the same output, and many units derate substantially. If your supply routinely sags, size on the low-voltage capacity, not the headline figure. ## Where the missing power goes Nothing is lost. The difference between VA and watts is **reactive power**, measured in VAr: ``` VAr = √(VA² − W²) ``` For our 1000 VA unit at 600 W: ``` VAr = √(1,000,000 − 360,000) = 800 VAr ``` That 800 VAr is energy moving into the load's magnetic or capacitive fields during part of the cycle and coming back out during the rest. It does no work, but it flows through every conductor and every winding on the way, heating them exactly as much as working current would. Which is the answer to "why not just rate everything in watts?" — because the conductor, the winding and the switching device do not care whether the current is useful. They only care how much of it there is. That is what VA measures. ## Three-phase, briefly For a three-phase source the conversion is unchanged: ``` W = VA × PF ``` Phase only enters when you convert to current. A 10 kVA three-phase supply at 415 V delivers 13.9 A per line; the same 10 kVA single-phase at 230 V delivers 43.5 A. Same apparent power, same watts at a given PF, three times the current. See [kVA to amps](/tools/kva-to-amps). ## Common mistakes **Assuming VA equals watts.** Only true at unity power factor. On a typical UPS you are overstating capacity by 40 %. **Using the load's power factor.** For rating a UPS, the UPS's own PF governs. Your load's PF matters for how much current it draws, not for what the UPS can supply. **Sizing on running watts for motor loads.** Fridges, pumps and compressors draw a large starting surge. The UPS or stabiliser has to survive that surge, not the running figure. **Reading a peak rating as continuous.** Some inverters advertise a surge figure prominently and a continuous one in the small print. ## Power factor is not a fixed property of the load The conversion treats power factor as a number you know. On a linear load it effectively is — an induction motor at full load sits near 0.85 and stays there. On the electronic loads a UPS usually feeds, it moves with the load itself. A switch-mode power supply without power factor correction draws current in narrow pulses at the peak of the voltage waveform. Its displacement between voltage and current is small, but the distortion is large, and the resulting true power factor is commonly between 0.5 and 0.7. Modern supplies with active correction reach 0.95 or better, and the two behave completely differently behind the same VA rating. Which means the 0.8 assumption printed on most UPS marketing is a compromise rather than a measurement. A rack of servers with corrected supplies will draw close to its rated watts from a UPS sized in VA at 0.8, leaving capacity unused. A mixed load of older equipment, small wall adapters and unfiltered electronics can exhaust the VA rating well before the watt rating, and the UPS will report itself full while the power meter reads comfortably low. ## Leading power factor, and the UPS that refuses the load Everything above assumes current lagging voltage, which is the normal case for motors and transformers. Corrected electronic supplies can do the opposite. An active power factor correction stage presents a slightly capacitive load, and a large number of them in parallel — a data rack, a floor of workstations — can push the aggregate power factor leading. Older UPS and generator designs were specified only for lagging loads, and a leading load makes their output voltage regulation unstable. The symptom is a UPS that runs a modest load without complaint and then reports an output fault, or a generator whose automatic voltage regulator hunts. Anything specified in the last decade generally handles leading power factor down to about 0.9 and states so on the datasheet. It is worth checking on equipment being reused rather than bought, because the failure appears only once the load is connected and looks nothing like an overload.