Residential Power Factor Correction

Sponsored by: PacifiCorp

1. Project Overview

Initial Project Objective
Energy efficiency in the household is becoming increasingly important as the costs associated with producing electricity are rising and the negative effects of emissions from fossil fuel generation sources are becoming better understood. In an effort to reduce the demand for fossil fuel generated electricity and to mitigate power losses in the existing electricity transmission infrastructure, many utilities are looking for ways to increase the efficiency of electricity consumption at the residential level. This project’s sponsor, PacifiCorp, has proposed a residential power factor correction system that will serve to increase the efficiency of inductive loads for residential customers. PacifiCorp has given our group the freedom to define the scope and minimum requirements for this project and they will serve in a support role to oversee that the project is headed in the right direction.

The need for power factor correction at the residential level is partially caused by integral ballast compact fluorescent lamps, which are quickly replacing incandescent bulbs for most residential applications because of their much higher energy efficiency. For example, a 14 Watt compact fluorescent light bulb has the equivalent light output as a 60 Watt incandescent bulb, which means significant cost savings for the end user. A drawback of the transition to compact fluorescents is that these bulbs have very poor power factor due to the current limiting circuitry required, which is not present in incandescent bulbs. Having a poor power factor means that additional power (reactive power) is transferred to the load and back to the source in each cycle. This power does no useful work and the increased current associated with it causes additional losses in transmission lines based on the current squared times resistance relationship. In addition, transformers and conductors throughout the transmission system will have to be sized larger to handle this wasteful current.

To prove the need for this project our group conducted an experiment to measure the power factor of a typical screw-base compact fluorescent light bulb. We purchased an Ecosmart 14 Watt compact fluorescent 2-pack for $3.98 from Home Depot. A true rms Fluke multi-meter was used to measure the rms voltage and rms current, which were 121.1V and 0.205 A respectively. Apparent power (A) was calculated using the following equation: A = V*I = (121.1 V) * (0.205 A) = 24.83 VA. A wattmeter was then used to find the real power (P) delivered to the load which was found to be 12 W. With that the following equation was used to determine the power factor of the load: Power Factor = cosϕ = P/A = (12 W) / (24.83 VA) = 0.483. This exercise demonstrated that the power factor for common compact fluorescent bulbs is indeed very poor. It would not be uncommon for a typical home to have 20 or more of these bulbs on at once considering how many it takes to provide sufficient lighting in a single room, and a power factor correction device would be very useful under these conditions.

The scope of this project will be to create a “proof of concept” device that is capable of correcting the power factor of a load bank, consisting of several fluorescent light bulbs. This device will be capable of correcting power factor by a minimum of 10% for a minimum of 300 watts worth of fluorescent lights. In order to decrease costs and complexity, this device will only correct the power factor for loads connected to a single phase 120 Vrms supply, as opposed to the split-phase voltage supply provided to residential customers at the main service panel. This device will consist of three major components: power factor meter, control/switching circuit, and a bank of capacitors.

The power factor meter will measure system electrical properties necessary for calculating the power factor. The control/switching circuit will use the values measured by the power factor meter, calculate the capacitive load needed to correct the power factor towards unity, and send signals to a switching circuit that will connect the proper amount of capacitive load. A bank of capacitors with switching capability will be used to actively add capacitive load to the system.

The final product will prove how effective a residential power factor correcting device can be for improving the efficiency of the electricity transmission system. This device could be used as a framework for creating a product that could be installed at the main service panel of a home and correct the power factor of all household loads. As the transition towards fluorescents in residential lighting applications becomes more prevalent, this device will be an important step in improving the efficiency of these new loads.

Revised Project Objective
The initial assumption that poor power factor in electronic ballast fluorescent lights could be corrected by applying shunt capacitance across the load proved to be false. This phenomenon was not fully understood until the team was able to obtain a current transformer. The current transformer made it possible to closely analyze the current being drawn from the compact fluorescent lights using an oscilloscope. The current waveform being drawn by the load exhibited a sawtooth (not sinusoidal) shape and was so distorted that the actual zero crossing of the current waveform could not be accurately determined for comparison with the voltage zero crossing. The oscilloscope measured current transformer output when connected to a 14 watt compact fluorescent light bulb is shown in the picture below.



Through further investigation and coursework in power electronics it was discovered that true power factor is a measure of more than just the phase displacement between the current and voltage waveforms. True power factor takes into account the harmonic distortion of the current drawn by the load because the harmonics require the power source to deliver more current than would be required for a linear load with a purely fundamental current waveform [1]. This relationship is shown in the equation below where Is1 is the fundamental component of the current waveform and Is is the total current. DPF is the displacement power factor which is equal to cos(ϕ), where ϕ is the phase difference between the voltage and current waveforms [1].

PF = (Is1/Is)*DPF

This relationship shows that you could have a load that draws unity displacement power factor but still have a poor total power factor due to the harmonic distortion in the load current. These harmonic currents can overload shunt capacitors and they can only be improved by using specially designed front-end electronics that are beyond the scope of this project. The project objective was changed from correcting power factor for compact fluorescent lights to correcting power factor for magnetic ballast fluorescent lights. These magnetic ballast fluorescents have low harmonic distortion and very low displacement power factor. A bank of 9 fluorescent ballast lights were tested in the lab and the power characteristics measured were 493 VA, 270 watts, 4.1 amps, and 0.55 PF. The oscilloscope measured current transformer output when connected to a 3 light bulb magnetic ballast is shown in the picture below.



No system changes for the switched capacitor bank were required to move from a target load of compact fluorescents to magnetic ballast fluorescents. The overall objective is to correct power factor for a bank of a minimum of 300 watts of magnetic ballast fluorescent lights by a minimum of 10%.

System Requirements

# Must correct power factor for a bank of 300+ watts of magnetic ballast fluorescent lightbulbs by greater than 10%.
# Device must be easy to operate and have visible on/off indication.
# Device must be safe for the user.
# Device must display the corrected power factor and active power usage of downstream loads.
# Must be compact so it doesn't take up too much space in the home.
# Equipment will be protected from overcurrent conditions with circuit interruption devices.

2. Background Research
   
3. System Requirements and Desired Features
   
4. Design Solutions
   
5. Top Level Block Design
   5.1 Breaker
   5.2 On/Off Switch
   5.3 Current Sensing Circuits
   5.4 Voltage Sensing Circuits
   5.5 DC Power Supply
   5.6 On/Off LED
   5.7 Microcontroller
   5.8 Controller Code
   5.9 LCD Display
   5.10 Switch Drivers
   5.11 Capacitor Bank
   5.12 Capacitor Discharge
   5.13 Fuses
   5.14 Enclosure
   5.15 Capacitor Enable/Disable Switch
   
6. Testing
   
7. Project Timeline
   
8. System Test Evidence
   
9. Expo Materials
   

---

Project Video

Group Members

Left to right: Justin Ammons, Sean Brosig, and Grant Blakeley

Grant Blakeley blakeleg@onid.orst.edu
Sean Brosig brosigs@onid.orst.edu
Justin Ammons ammonsj@onid.orst.edu

Mentors

Adrian Smith Adrian.Smith@pacificorp.com

References

[1] Ned Mohan. Power Electronics. New York. MNPERE Minneapolis. 2009.

Additional Documents
Reference Links