Effect of the Harmonic Voltage Distortion on the Efficiency of a Compact Fluorescent Lamp

This paper evaluates the effect of the voltage harmonic distortion over the efficiency of a compact fluorescent lamp that is fed with a constant RMS voltage and constant frequency. Several works have been published about the assessment of compact fluorescent lamps, but the effect of the voltage distortion over the efficiency is still an open topic. This work focuses on designing an experiment to estimate the efficiency of a compact fluorescent lamp while changing the voltage harmonic distortion of the power supply. First, a mathematical model that represents a bus susceptible to harmonic distortion (high impedance) that feeds the compact fluorescent lamp is analyzed. Then the mathematical model is reproduced through a test bench in a laboratory of rotating electrical machines. The test bench produces a three-phase bus with constant voltage and frequency, and variable voltage harmonic distortion. The compact fluorescent lamp is subjected to varying harmonic voltage distortion while recording its electrical variables and the produced lumens to estimate its efficiency. That is a practical approach to calculate the lamp efficiency while several 1 Universidad Nacional de Colombia (Medellín-Antioquia, Colombia). earojas@unal.edu.co. ORCID: 0000-00025385-3328 2 Ph. D. Universidad Nacional de Colombia (Medellín-Antioquia, Colombia). ajsaaved@unal.edu.co. ORCID: 0000-0002-2529-5082 3 Ph. D. Universidad Nacional de Colombia (Medellín-Antioquia, Colombia). caramosp@unal.edu.co. ORCID: 0000-0003-2231-4177 Effect of the Harmonic Voltage Distortion on the Efficiency of a Compact Fluorescent Lamp Revista Facultad de Ingeniería (Rev. Fac. Ing.) Vol. 29 (54), e11604. 2020. Tunja-Boyacá, Colombia. L-ISSN: 0121-1129, e-ISSN: 2357-5328, DOI: https://doi.org/10.19053/01211129.v29.n54.2020.11604 works limit their scope measuring only the efficiency of the input converter. The experimental results show that a variation of the voltage harmonic distortion of 8 % on a compact fluorescent lamp reduces its efficiency. Those results put into evidence the importance of regulating harmonic distortion limits to reduce or prevent the increment of power losses caused by harmonic components.


I. INTRODUCTION
The need to improve the efficiency of energy consumption has been evident in recent decades. Therefore, every discipline is working to improve the processes aimed at improving the energy production and consumption [1][2]. The new developments in technology have increased the electrical energy demand for nonlinear loads, which reduce the power quality of the distribution systems. At the residential level, the nonlinear loads correspond to computers, mobile phones, LED lamps, compact fluorescent lamps (CFLs), etc. Those loads request discontinuous current and therefore deform the voltage waveform of the distribution systems [3].
Academic authorities and researchers developed a workshop on power system harmonics, on January 2014. One of the discussions was about the main issues on harmonics at that time and in the close future. As a result of that workshop, a roadmap was written and published [4]. There is a specific topic in voltage distortion that says: "A reduction in efficiency of end-user equipment was reported in cases with high voltage distortion. This phenomenon should be urgently studied further. If it is shown that this reduction in efficiency is significant, it could form the basis for new voltage-distortion limits." Based on the last statement, it is inferred that all the nonlinear final user devices that are connected to the distribution systems, should be evaluated to identify the effect of voltage distortion on their efficiency. In this way, determining the effects of the voltage harmonic distortion into the efficiency of electrical loads will make it possible to revise the harmonic distortion limits defined in regulation documents.
In recent years, replacing incandescent bulbs by CFL has been a successful strategy to reduce electrical losses in lighting systems [5]. However, it has been demonstrated that CFL and LED lamps, when massively installed, can introduce a voltage harmonic distortion into the connected bus; for example, a 12 % system load due to CFLs produces a 9.64 % voltage harmonic distortion [6]. That is a good The rest of the paper is organized as follows: Section 2 presents the methodology used in this work; in particular, it presents the mathematical model used in the experiment design. Section 3 describes the experiment design, reporting the experimental results and the discussion. The conclusion closes the paper.  Before closing SW1, the nonsinusoidal current i(t) requested by the variable nonlinear load can be divided into the fundamental and harmonic components as it is shown in (1).

II. METHODOLOGY
Where the I1 is the magnitude of the fundamental current, Ꞷ1 is the fundamental frequency, and θ1 is the phase angle of the fundamental current. Since the load is connected through an impedance Zs, the demanded current produces a voltage drop. The bus voltage reflects the fundamental and harmonic components of current over the impedance and therefore the voltage at the load node is given by (2).
The apparent power demanded by the load is defined by (4), where both load current and voltage are expanded into their fundamental and harmonic components.
The first term corresponds to the fundamental component, the second term is defined as the power of current distortion, the third term is defined as the power of voltage distortion, and the fourth one is defined as harmonic apparent power. A first hypothesis is that the last three terms exhibit an active power component entering into the load that will not have an effect on the load output, which will reduce the load efficiency.
The system presented in Figure 1 is adequate to show how a current with harmonic content, drawn by a nonlinear load, can produce a voltage in the load bus with harmonic content. The system is also adequate to show that the increment of the load reduces the RMS voltage of the load bus. If a constant RMS voltage is required during the increment of the load, it is necessary to introduce a voltage regulator.
The system presented in Figure 1 is implemented in a laboratory of rotating electrical machines. Figure  On the other hand, to produce a variable harmonic content of voltage in the load bus, a variable nonlinear load is connected to the SG, see Figure 2. The nonlinear load is implemented with another set of electrical machines. The set is composed by a speed driver feeding an induction motor (IM), which drives a dc generator (DC G).
The increment in the voltage distortion in the load bus occurs due to the increment in the nonlinear current of the speed driver, which flows throughout the synchronous impedance of the SG producing harmonic voltage drops as can be seen in the third term of Equation (2). The increment of the current of the IM occurs due to the increment in the load of the DC G. The load of the DC G are incandescent bulbs connected to its terminals. Therefore, connecting incandescent bulbs to the terminals of the DC G will produce an increment in the voltage harmonic content of the load bus.      Contrasting the active power demanded by the CFL with the change of the produced lumens and the variation of the THDV, it is concluded that increments on the THDV reduced the CFL efficiency. This is observed in the increment of the active power requested without significant increments on the lumens produced by the CFL. Therefore, higher THDV forces the CFL to consume higher power for the same lumens production.

IV. CONCLUSION
The effect of the voltage harmonic distortion over the efficiency of a CFL has been experimentally evaluated in this paper. The analysis of results shows that increments in the voltage total harmonic distortion, on a CFL, between 8.2 % and 15.3 %, and keeping constant both rms voltage and frequency, produce increments in the current, active power, reactive power and apparent power, but the light production remains almost constant. The increment in the input current of the CFL implies an increment in the electrical losses of the final user devices, hence it confirms the efficiency reduction. The results put into evidence the requirement of defining standard tests to evaluate the efficiency of final user devices, and also the need to review the harmonic distortion limits in regulation documents.