Preview only show first 10 pages with watermark. For full document please download

On The Efficiency And Reliability Of High-current Rectifiers

On the efficiency and reliability of high-current rectifiers

   EMBED


Share

Transcript

  On the Efficiency and Reliability of High Current Rectifiers Pablo E. Aqueveque, Eduardo P. Wiechmann  University of Concepcion, CHILE Rolando P. Burgos CPES Virginia Tech, USA Corresponding Author Eduardo P. Wiechmann P.O. Box 4003, Correo 3 −  Concepción   CHILE Tel: +56-41-204 775 Fax: +56-41-246 999 [email protected]    Abstract- Industrial electrolytic rectifiers work with high currents (in the order of kilo Amps) and very low dc voltage levels (few hundred Volts). The preferred industrial solutions for this type of application are Thyristor-based double-bridge and four-star multi-pulse rectifiers. A number of alternative more advanced rectifiers have been recently proposed presenting improved electrical properties, such as reduced harmonic distortion and reduced reactive power consumption. The various tradeoffs existent between these topologies however are presently only evaluated qualitatively. This paper addresses this present lack of proper design tools and proposes an engineered quantitative solution for the evaluation and selection of high current rectifiers. The proposed method is based on the reliability, electrical performance, efficiency and required equipment investment for these rectifiers. The IEEE Standards 493-1997 and IEEE C57.18.10 for reliability and efficiency are the basis of this method, being used specifically to configure the evaluation platform. Complete results for the standard double-bridge and four-star rectifiers, together with improved rectifier topologies including the chopper-rectifier, the sequential rectifier, and the filterless rectifier are presented and thoroughly discussed in the paper. From these, it is concluded that the modular multi-cell technology used in the chopper high-current rectifier is capable to challenge the long-used Thyristor-based industrial solutions.     page 1/4 On the Efficiency and Reliability of High Current Rectifiers I.   I NTRODUCTION   High grade metal production using electrowinning (EW) or electrorefining (ER) has increased by 100% during the last ten years [1]. Thyristor-based phase control rectifiers are used in these processes, where the double-bridge (DB) and four-star (FS) rectifiers represent the conventional rectifier configurations used by the mining industry [2]. Alternative rectifier configurations aimed to improve the electrical performance of conventional rectifiers have been proposed. An extensive rectifier revision of the operation, configuration and electric properties of conventional and modulated rectifiers was presented in [2]. It is concluded that PWM current source rectifiers are adequate for medium voltage ac drives; however as presented in [2] this configuration requires a relatively high commutation frequency and ac input capacitors, and hence is not covered in this paper. The chopper-rectifier (CR) presented in [3] is a multi-cell parallel arrangement, where each cell is configured as a single three-phase diode rectifier and three paralleled choppers. This rectifier offers a 0.96 input power factor for the entire range of operation and does not require tuned filters. However, its operation requires 4 semiconductors in series impairing its efficiency for high current low voltage applications. A more conventional thyristor-based approach is the phase controlled sequential rectifier (SR) presented in [4]. This configuration uses two sequentially connected double star series rectifiers. The first incoming rectifier delivers 85% of the output voltage and the second bridge provides the remaining 15% in sequence. This sequential operation results in a significantly smaller input filter. An alternative approach driven by the technological trend to perform commutation at medium voltage with IGCT´s has been presented in [5]. The converter uses a medium voltage VSI to modulate the voltage delivered to a step-down transformer. This filterless rectifier FR offers a 0.99 power factor. Energy quality indexes such as harmonic distortion and reactive power have been used to compare industrial rectifiers. However, industrial requirements call for higher efficiency, reduced size, lower investment and near perfect reliability. This requires high efficiency transformers, low current distortion, close to zero reactive power, low conducting voltage semiconductors, and elimination of tuned filters on the input ac side. Consequently, this paper takes into consideration for the evaluation of the rectifiers the following costs: equipment, industrial space, operational efficiency and rectifier reliability. The aim is to produce an optimization design criteria for the rectifier and to provide a powerful tool to precisely quantify the economic impact of the rectifier selection based on current IEEE standards.     page 2/4 II.   H IGH CURRENT RECTIFIERS RELIABILITY   The rectifier reliability is tied to the topology and electric components of the corresponding circuit. A usual design practice for rectifiers is to employ paralleled semiconductors. Naturally, the system reliability also depends on the control circuitry, electric sensors and transducers, power transformers, cable connections, switchgear, tuned filters and the complete system configuration. The phase-controlled rectifiers DB, FS and SR require tuned filters and associated switchgear equipment. Both FS and CR share the capability of continued power delivery to the load thanks to their parallel structures; however, this operation may limit the output current capability. The highest reliability is obtained by the FS configuration. This is the result of a structure using two transformers, each one delivering power to twin parallel rectifiers. The lowest is exhibited by the FR. This topology requires series power conversion stages including a voltage source inverter VSI and forced commutation semiconductors. However, the rates of failure associated to VSI´s have been improved with the development of IGCT´s. So, future IEEE standards revisions will reflect this trend and may modify present results. Table I presents reliability figures computed for each of the rectifiers based on IEEE Standard 493-1997. This information can be used to modify a particular rectifier topology to increase its reliability. For example, designing with redundancy criteria the DB configuration would require a second rectifier transformer. Naturally, an increase in equipment investment should be expected. III.  HIGH CURRENT RECTIFIERS E FFICIENCY . Losses associated with high current rectifiers for ER and EW applications can be classified in three groups: a) Power semiconductor losses :   by far the major contributors of this group are the conducting losses. The most relevant parameter of these conducting losses is the on state voltage and the circuit configuration requiring one or more semiconductors in series. Fig.1 depicts these losses for the rectifiers under evaluation, where the FS unit exhibits the best conducting losses performance and the CR the worst. b) Rectifier transformer losses :   To evaluate these losses the nonlinear nature of the converter currents is an important factor. Traditionally transformer losses are proportional to the square of voltage (iron) and proportional to the load current (copper); therefore the converter power factor will influence the transformer losses. Additionally, harmonic distortion (,i.e., the k   factor) may increase total losses by a significant amount. It is clear that a better power factor and a linear current will help in reducing the total transformer losses. Further losses reductions require modification of the construction characteristics of the transformer. Increasing the conductor diameter should improve losses by reducing     page 3/4 conducting losses and allowing the transformer to operate at lower temperature. As it is known, operating at lower temperatures improves the transformer efficiency. Again, a solution balancing efficiency (operational costs) should be compared with the extra investment in transformer quality. Fig.2 shows the result of this losses evaluation. The DB an FR rectifiers outperform their competitors. The reliable FS nonetheless produces the highest rectifier losses. c) Power distribution system losses :   Losses are produced in power lines, substations and motors. To improve losses at the system the reactive power should be kept at low levels. A low power factor increases system currents (power lines and substations) and impairs voltage regulation (motors). The global efficiency of the rectifier system is shown in Fig.4, while Table I summarized the expected failure rate, down time per failure, electric losses, space required, equipment investment and overall economic performance. IV.   C ONCLUSION   This paper provides a powerful design tool for the evaluation of high current rectifier technology using IEEE standards. The results obtained in this paper show that the two main contributors required to achieve cost-effective solutions are efficiency and reliability. Regarding emerging topologies for high current industrial rectification, the chopper-rectifier employing multi-cell technology seems the only capable of challenging long-used thyristor-based configurations. This trend is expected to be reinforced with the development of PEBB-based multi-cell rectifiers [11]. V.   R EFERENCES   [1]   “Annual Statistics report for Copper and Metals 1985-2004”, Chilean Copper commission, COCHILCO, Mining Ministry, Chile. www.cochilco.cl [2]   J.Rodriguez, J.Pontt, C.Silva, E.Wiechmann, P.Hammond, F.Santucci, R. Alvarez, R.Musalem, “Large Current Rectifiers: State of the Art and Future Trends”, IEEE Trans. Ind. Elect., vol. 52, no. 3, pp. 738-746, june 2005. [3]   V. Scaini, W. Veerkamp, “Specifying DC Choppers Systems for Electrochemical Applications”,  IEEE Trans. On Ind.  Applicat  ., Vol. 37, N°3, May/June 2001, pp. 941-948. [4]   E.P.Wiechmann, R.P.Burgos and J.Holtz, “Sequential Connection and Phase Control of a High-Current Rectifier Optimized for Cooper Electrowinning Applications”,  IEEE Trans. On Ind. Applicat  ., vol 47, no. 4, pp. 734-743, August 2000. [5]   E.P. Wiechmann and P.E. Aqueveque, “Filter-less High Current Rectifier for Electrolytic Applications”,  Industry Applications Conference, 2005. Volume 1, 2-6 Oct. 2005 Page(s):198 – 203.  [6]   Nando Kaminski, Thomas Stiasny , “Failure Rates of IGCTs Due to Cosmic Rays”, Application Note 5SYA 2046-01, ABB Switzerland Ltda, July 2005. [7]   L. Pierce, “Transformer Design and Application Considerations for Nonsinusoidal Load Currents”,  IEEE Trans. On Ind.  Applicat  ., vol 32, no. 3, pp. 633-645, May/Jun 1996. [8]   Derek A. Paice “Power Electronic Converter Harmonics, Multipulse Methods for Clean Power”, ISBN 0-7803-1137-X, IEEE Press, 1995. [9]   IEEE Std 493-1997, “IEEE Recommended Practice for the Design of Reliable Industrial and Commercial Power Systems”. [10]   IEEE Std 57.18.10-1998, “IEEE Standard Practices and Requirements for Semiconductor Power Rectifier Transformer”. [11]   T. Ericsen, “Power Electronic Building Blocks: A Systematic Approach to Power Electronics,” IEEE Power Engineering Society Summer Meeting, vol. 2, pp. 1216-1218, July 2000.     page 4/4 Table I.- Reliability Economic Performance Equipment λ Failure rate (failures per unit-year)   r hours of downtime per failure Losses [kW] Site space cost MUS$ Energy cost MUS$/year Reliability cost MUS$/year Estimated equipment cost MUS$ Annual cost MUS$/year DB 0.01983 89 349.3 0.156 9.48 0.30 2.0 10.15 FS 0.00452 261 420.5 0.156 9.68 0.17 2.1 10.24 SR 0.05782 56 421.4 0.132 9.50 0.62 2.2 10.53 FR 1.32261 103 347.2 0.110 9.43 22.63 2.3 32.47 CR 0.01530 39 512.5 0.102 9.76 0.13 2.6 10.36 Copper price: US$ 1.45 lb Electric energy: US$ 0.05 kW/h Site space: US$12 / ft 2  Initial cost per failure: US$100,000 Hourly cost during failure: US$ 150,000 Interest rate: 8% Annual cost computed for an eight year project VI.   F IGURES   0204060801001201401601802000 5 10 15 20 25 30 35 40 Load current [kA]    S  e  m   i  c  o  n   d  u  c   t  o  r  s   L  o  s  s  e  s   [   k   W   ] Double bridgeFour-StarFilterlessSequentialChopper-rectifier   0501001502002503003504004505000 5 10 15 20 25 30 35 40 Load current [kA]    T  r  a  n  s   f  o  r  m  e  r   L  o  s  s  e  s   [   k   W   ] Four-starDouble BridgeFilterlessSequentialChopper-Rectifier  Fig.1. Semiconductors losses Fig.2. Transformer losses 0200400600800100012000 5 10 15 20 25 30 35 40 Load current [kA]    T  o   t  a   l   L  o  s  s  e  s   R  e  c   t   i   f   i  e  r   [   k   W   ]  Double bridgeFour starFilterlessSequentialChopper-Rectifier   90,091,092,093,094,095,096,00 5 10 15 20 25 30 35 40 Load current [kA]    G   l  o   b  a   l   E   f   f   i  c   i  e  n  c  y   [   %   ] Double bridgeFour-starFilterlessSequentialChopper-Rectifier  Fig.3. Rectifiers Total Losses Fig.4. Global efficiency View publication statsView publication stats