watts = volts x amps

(for reactive loads this actually over estimates the watts slightly)

An estimate can also be obtained by using the bar chart further below.

Starting Load Requirements
Determining the starting requirements can be a bit more complicated. Certain electrical devices require additional power and current when initially turned on.  This is true for motors because the rotor of the motor and the shaft driven load (fan, pump, compressor, saw, etc.) is initially at a standstill. It requires more energy to accelerate these rotating parts to operating speed than it does to keep them rotating. Therefore, during the period of acceleration, the demand on the power supply is greater. To precisely evaluate the motor starting capability the detailed motor characteristics need to be known. However, a rule of thumb is usually sufficient. Most engine driven generators will start a motor with up to 1/5th the horsepower of the engine, if it is the first load connected. For example, a 2500 watt generator driven by a 5 horsepower engine will usually start up to a 1 horsepower motor. This assumes a common type of motor design with NEMA code G starting characteristics. This data can be found on the motor nameplate.

Power Quality & Distortion
Perfectly pure AC power is a sine wave for both the voltage and current.  Resistive loads such as incandescent lights and heaters are linear loads since the current is always proportional to the voltage applied. Some types of generators and non-linear loads can alter this perfect sine wave. A non-linear electrical load does not have a linear relationship between the voltage applied and the current that flows into the load. Certain types of electronics, lighting ballasts, arc welders and other devices are non-linear. Welding generators due to their design and poorly designed generators may also produce a distorted AC wave. When a significant portion of the load on a generator (or any power source for that matter) is non-linear, all the loads fed by the source will see this distortion. A measure of this distortion is called THD, or total harmonic distortion. If the distortion is severe enough, motors and transformers will operate hotter. Over a long period of time this can cause a reduction in life. And some other sensitive electronic equipment may not operate.  A specific example is an uninterruptible power supply (UPS) system powering computers or communications equipment. These types of devices cause some distortion of the AC wave and at the same time can be negatively affected by it. A UPS system powered by an inadequately sized backup generator may continue draining the internal battery rather than switching over to generator power and charging the battery.

To reduce chances for THD problems, the rule of thumb is to select a backup generator kW size at least three times the kW of non-linear loads to be powered. For example, if you have 2000 watts of computers fed by UPS systems and 1000 watts of incandescent lighting to be fed by a generator, first total the power:

The generator needs to be at least 6000 watts in accordance with this rule of thumb.

UPS manufacturers usually have specific guidelines for each type of UPS that they sell stating how much to oversize a standby generator.

Typical Running and Starting Loads
The graph below shows typical values for common residential loads. For applications that are approaching generator ratings, the actual nameplate load data, or better yet measured data, should be used to ensure an adequately sized unit.

Measuring Building Electrical Load Using a Stopwatch
If you intend on powering most of the items fed by your utility electric meter you can measure your total building load at any time using simply a stop watch while observing your meter. Follow the steps below to make this measurement.

3.  Now, start the desired appliances, heating or air conditioning for the condition to be measured

4. Using a stopwatch while watching for the black mark on the meter's disk, measure the time it takes for one or more disk rotations. If the disk is rotating rapidly, better accuracy will be attained if you time more than one rotation.

Here is an example calculation for the first meter above.

From the meter's face, Kh = 7.2. The time measured for 5 rotations of the disk was 24 seconds. Thus, Rev = 5 and T = 24 seconds.

Test Reports

This section contains test results and measurements for various electrical loads. Note that these measurements are representative only for the specific models shown and under the particular test conditions. Different conditions can alter both the current values and starting times.

Slightly different line voltages will affect results as will the wire size and length feeding the load. Ambient temperatures and thermostat settings will affect refrigeration compressor current demands. For well pumps, the pressure settings and the depth that the pump is positioned will affect measured results.

Additional test measurements are being added to this page on a regular basis. If you have some tests results that would be helpful to others, send them to us for posting here.

Refrigerator Starting Current
This test measured the starting current for a refrigerator with the following nameplate data.

Results of one measurement are shown in the plot below. It shows a maximum inrush current of about 13 amps which lasts only about one-half second. Also, the running current is significantly less than the nameplate value. It should be noted that most refrigerators (including this one) are "frost free." This means that on a regular basis a timer shuts off the compressor and turns on resistance heaters to clear frost buildup in the freezer section. This defrost current was not measured and is probably greater than the compressor running current. This may explain the large difference between the nameplate current value and the measured value.

Freezer Starting Current
This test measured the starting current for a freezer with the following nameplate data.

Results of one measurement are shown in the plot below. It shows a maximum inrush current of less than 5 amps which lasts only about 0.3 seconds. Also, the running current is significantly less than the nameplate value. It should be noted that most freezers are "frost free." However, this one is not.

Well Pump Starting Current
This test measured the starting current for a well pump motor. The motor is rated 3/4 hp and 240V. Results of one measurement are shown in the plot below. It shows a maximum inrush current of about 18 amps which lasts only about 0.2 seconds.