2011 IEEE International Symposium on Power Line Communications and Its Applications
AC-DC Smoothing Capacitor Current Coupling for Improved Powerline Signal Reception A. J. Snyders‡, P.A. Janse-van Rensburg†, HC Ferreira‡ and AJ Han Vinck§ ‡
University of Johannesburg: Telecommunications Research Group, Johannesburg, South Africa asnyders,
[email protected] † Walter-Sisulu University: Electrical Engineering Dept, East London, South Africa
[email protected] § University of Duisburg-Essen: Institute for Experimental Mathematics, Essen, Germany
[email protected]
Abstract—The proliferation of Switch-Mode Power Supplies (SMPS) has led to a significant size decrease in step-down transformers for PC power supplies, moving away from the traditional linear power supplies which are heavier and bulkier than their modern counterparts. Step-down transformers are becoming more and more redundant in AC-DC converters due to higher blocking voltages of MOSFET’s and the need for reducing the size and the power efficiency of power supplies for mobile equipment. Still, an integral part of any PC power supply is the AC-DC conversion stage. A diode bridge, smoothing capacitor and a load resistor is the very least number of components any power supply requires to be classified as an unregulated power supply. This paper presents live results showcasing a specialized coupling technique of signal coupling for Power-Line Communications. This technique improves signal reception, for the AC-DC smoothing capacitor charging intervals of the mains cycle, as compared to that of a standard PLC receiver during the same intervals. This paper investigates the insertion of a special coupling transformer in the parallel capacitor branch of the rectification stage of a typical AC-DC convertor, for the case where no step-down transformer is present. The technique shows time-domain diversity in signal coupling and utilizes the capacitor charging mechanism which occurs twice during the mains cycle. Frequencies in the CENELEC bands are used for the study.
reception, combined with the impact of AC-DC rectification is investigated. II.
BACKGROUND
A. 220V Live Experiment revealing the impact of Transformer-less AC-DC Rectification on standard PLC signal coupling The parallel smoothing capacitor impedes the low frequency “power” current, and passes through the high frequency PLC current. A standard PLC receiver operating on the same AC phase in the CENELEC A-band as a transformerless AC-DC converter and connected back-to-back to the PLC transmitter, suffers severe signal attenuation during the charging period of the capacitor. During the capacitor charging period the current is non-zero as shown in upper left corner of Figure 1. During the remaining time (current zero) the standard PLC receiver will have sufficient received signal power. The capacitor exhibits extremely low impedance for the CENELEC A-band frequencies during its charging interval, lower than 1Ω. The standard PLC receiver exhibits much higher impedance in comparison and thus receives almost no current from the PLC transmitter.
Keywords- AC-DC converter; CENELEC; Smoothing Capacitor; coupling; current; Power-Line Communications
I.
INTRODUCTION
In [1], it was shown that powerline signal-to-noise ratios are negatively impacted by the low impedance present on the mains network due to connected loads. The paper also showed how the frequency response of the standard coupling circuit is altered when the powerline impedance varies. In this paper, impedance variation synchronous to the mains cycle is investigated; as it occurs twice in every mains cycle for the case where a standard coupler is connected close to AC-DC rectifier circuits. The charging of the smoothing capacitors found in AC-DC rectifier circuits results in signal degradation in the standard coupler. Figure 1. Implemented Signal Coupling with “Rec” denoting Receiver and PLC transmitter located elsewhere
In [2], the effect of distance on signal reception was also investigated. In this paper the impact of distance on signal
978-1-4244-7750-0/11/$26.00 ©2011 IEEE
341
IV.
on the ac input. Standard Receiver at ac Input and specialized receiver in Smoothing capacitors’ branch.
III.
MEASUREMENT TECHNIQUE
Time and Frequency domain results were obtained from the setup which was connected to the 220V 50Hz mains network. An 85kHz unmodulated test signal of 1V was used for the experiment. The setup was configured so that the load current could be changed by using the two-position switch located on the PC power supply metallic case.
EXPERIMENTAL SETUP
The setup uses a standard AC-DC rectifier configuration consisting of 2 x 220µF electrolytic capacitors connected in parallel to a load which are variable, and consists of 3 x 50W resistors. One 4.7kΩ and 2 x 8.2kΩ resistors are used. By using only the 4.7kΩ a load current of 100mA is established; alternatively by using the parallel combination of all 3 resistors a load current of 200mA is maintained. A Photograph of the experimental setup is shown in Figure 2.
Measurements were done with the Transmitter coupling circuit connected back–to-back to the complete Receiver setup contained in the power supply case. Two channels on the Digital Oscilloscope recorded the signals at 100mA for the back-to-back configuration. Further measurements were performed for load currents of 100mA and 200mA with a distance separation between the Transmitter and Receiver of 20 m.
A communications transformer with only 1 turn on the winding connected in series with the electrolytic capacitors and 40 turns on the winding connected to the a Tektronix DPO7254 Digital Phosphor Oscilloscope, was used in the capacitor current coupler to receive the signal. This transformer was placed in series between the smoothing capacitors of the AC-DC rectifier circuit.
The setup was configured in such a way as to facilitate the comparison of the two types of signal coupling, namely capacitor current coupling (also referred to in the measurement results as “ccc”) as opposed to standard coupling (also referred to in the measurement results as “std”).
The standard receiver’s secondary primary winding ratio was 1:7 with the higher turns winding being on the side of the Digital Oscilloscope. The standard receiver used a 220V 0.22 µF in series with the communications transformer specified above.
V.
RESULTS
The results are divided into time and frequency domain results. The measurement data were exported from the Tektronix Digital Oscilloscope to a computer and were then processed using MatlabTM. The results depicted in Figure 5 and Figure 6 were only formatted in Matlab. These results are shown at the end of this paper.
A transmitter coupling circuit was used to send the test signal onto the powerline. The coupling circuit used a 220V 0.22 µF capacitor in series with a 7:1 communications transformer.
The results in Figure 3 and Figure 4 depict Fast Fourier Transform plots from the 1024 samples corresponding to the first time period in which the charging of the dual capacitors occurred. This was done to fairly compare the signal reception performance of the two ways of coupling, since the specialized current capacitor coupling only occurs approximately 10% of the power mains cycle, whilst the standard coupling receives signal power for the full mains cycle.
The inputs to the digital oscilloscope were switched to 50Ω; as was the output of the signal generator.
Figure 2. Photograph of experimental setup. In the top part the fuse, diode bridge, dual capacitors, transformer and the load resistors form the specialized capacitor current coupling, whilst to the center of the power supply case the standard coupler is located. Both couplers are connected to Live and Neutral conductors of the mains network.
342
transmitter and the receiver, the difference has already diminished to only 10dB. Further distance separation should show equal or better reception for the capacitor current coupling, even when considering that this technique only receives signal for 10% of the mains cycle. C. Matlab Processed Results for Charging Period only – to allow for fair comparison To allow for a fair comparison, a 1024 point FFT was performed in the Matlab environment which produced the graphs in Figure 3 and Figure 4. In these figures a relative comparison is established, which clearly shows a 1-2 dB signal reception gain for the Capacitor Current Coupling technique. Further dB gains are expected for larger distance separation between the transmitter and the receiver. In general, further study will include the varying of the time constant of the charging cycle to establish whether further gains in signal reception is possible for increased time constant. Larger distance and also higher load values will also be investigated. A signal-combining approach should also lead to overall improved signal reception.
Figure 3. Relative Magnitude Comparison for 100mA Load at a distance of 20m.
VII. REFERENCES [1] A. J. Snyders, P. A. Janse van Rensburg, H. C. Ferreira, “M-FSK Carrier Gain Adjustment for Improved Power-Line Communications,” IEEE Transactions on Power Delivery, vol. 25, no. 2, pp. 674-679, Apr. 2010. [2] P. A. Janse van Rensburg, H. C. Ferreira, “Coupler Winding Ratio Selection for Effective Narrowband Power-Line Communications,” IEEE Transactions on Power Delivery, vol. 23, no. 1, pp. 140-149, Jan. 2008.
Figure 4. Relative Magnitude Comparison for 200mA Load with a distance of 20m. VI.
CONCLUSIONS
A. Time Domain Results The first plot in Figure 5 clearly shows the severe signal attenuation the standard coupling technique are exposed to during the capacitor charging period. In the same time period the capacitor current coupler receives the signal with similar power and with increase in distance from the transmitter, this coupler also shows stability in signal reception, whereas the standard coupler shows increasing attenuation over distance (see remaining time domain plots). B. Frequency Domain Results For the back-to-back configuration the capacitor current coupler shows approximately 20dB weaker signal reception. However, with a distance separation of only 20m between the
343
Figure 5. Time Domain Results
344
Figure 6. Frequency Domain Results
345
