Static and dynamic back-corona characteristics

Similar documents
(1) Non-linear system

PRACTICE NO. PT-TE-1414 RELIABILITY PAGE 1 OF 6 PRACTICES ELECTROSTATIC DISCHARGE (ESD) TEST PRACTICES

Discontinued AN6262N, AN6263N. (planed maintenance type, maintenance type, planed discontinued typed, discontinued type)

Exercise 1-1. The Sine Wave EXERCISE OBJECTIVE DISCUSSION OUTLINE. Relationship between a rotating phasor and a sine wave DISCUSSION

ABB STOTZ-KONTAKT. ABB i-bus EIB Current Module SM/S Intelligent Installation Systems. User Manual SM/S In = 16 A AC Un = 230 V AC

Mixed CMOS PTL Adders

Experiment 3: Non-Ideal Operational Amplifiers

Magnetic monopole field exposed by electrons

CHAPTER 3 AMPLIFIER DESIGN TECHNIQUES

Experiment 3: Non-Ideal Operational Amplifiers

& Y Connected resistors, Light emitting diode.

Multi-beam antennas in a broadband wireless access system

Localization of Latent Image in Heterophase AgBr(I) Tabular Microcrystals

ISSCC 2006 / SESSION 21 / ADVANCED CLOCKING, LOGIC AND SIGNALING TECHNIQUES / 21.5

The Discussion of this exercise covers the following points:

A Novel Back EMF Zero Crossing Detection of Brushless DC Motor Based on PWM

Investigation of Ground Frequency Characteristics

Three-Phase Synchronous Machines The synchronous machine can be used to operate as: 1. Synchronous motors 2. Synchronous generators (Alternator)

To provide data transmission in indoor

THE STUDY OF INFLUENCE CORE MATERIALS ON TECHNOLOGICAL PROPERTIES OF UNIVERSAL BENTONITE MOULDING MATERIALS. Matej BEZNÁK, Vladimír HANZEN, Ján VRABEC

LATEST CALIBRATION OF GLONASS P-CODE TIME RECEIVERS

Carbon Composition Resistors

Compared to generators DC MOTORS. Back e.m.f. Back e.m.f. Example. Example. The construction of a d.c. motor is the same as a d.c. generator.

EET 438a Automatic Control Systems Technology Laboratory 5 Control of a Separately Excited DC Machine

MEASURE THE CHARACTERISTIC CURVES RELEVANT TO AN NPN TRANSISTOR

Investigation of propagation of partial discharges in power transformers and techniques for locating the discharge

Electronic Circuits I - Tutorial 03 Diode Applications I

Lab 8. Speed Control of a D.C. motor. The Motor Drive

A Development of Earthing-Resistance-Estimation Instrument

DESIGN OF CONTINUOUS LAG COMPENSATORS

Module 9. DC Machines. Version 2 EE IIT, Kharagpur

Engineer-to-Engineer Note

Time-resolved measurements of the response of a STM tip upon illumination with a nanosecond laser pulse

Math Circles Finite Automata Question Sheet 3 (Solutions)

Design of UHF Fractal Antenna for Localized Near-Field RFID Application

FATIGUE BEHAVIOUR OF COMPOSITE JOINTS WITH HEXAGON BOLTS

AN ELECTRON SWITCH. by C. DORSMAN and S. L. de BRUIN.

Robustness Analysis of Pulse Width Modulation Control of Motor Speed

Fuzzy Logic Controller for Three Phase PWM AC-DC Converter

Safety Relay Unit. Main contacts Auxiliary contact Number of input channels Rated voltage Model Category. possible 24 VAC/VDC G9SA-501.

Two-layer slotted-waveguide antenna array with broad reflection/gain bandwidth at millimetre-wave frequencies

5 I. T cu2. T use in modem computing systems, it is desirable to. A Comparison of Half-Bridge Resonant Converter Topologies

Geometric quantities for polar curves

Comparison of soundscape on the ground floor of tube-houses in Hanoi and open urban space in Bordeaux

Synchronous Machine Parameter Measurement

Control of high-frequency AC link electronic transformer

Synchronous Machine Parameter Measurement

ARC DISCHARGE AND EROSION BEHAVIOR OF SILVER ELECTRIC CONTACTS BETWEEN STATIC GAP

Power System Stability Enhancement By UPFC Based Power Oscillation Damping Controller

Lecture 16: Four Quadrant operation of DC Drive (or) TYPE E Four Quadrant chopper Fed Drive: Operation

FTU263. Ripple Control Receiver. Technical Data. Load Management Ripple Control

University of Dayton Research Institute Dayton, Ohio, Materials Laboratory Wright Patterson AFB, Ohio,

Network Theorems. Objectives 9.1 INTRODUCTION 9.2 SUPERPOSITION THEOREM

Example. Check that the Jacobian of the transformation to spherical coordinates is

Performance Monitoring Fundamentals: Demystifying Performance Assessment Techniques

High Speed On-Chip Interconnects: Trade offs in Passive Termination

Kirchhoff s Rules. Kirchhoff s Laws. Kirchhoff s Rules. Kirchhoff s Laws. Practice. Understanding SPH4UW. Kirchhoff s Voltage Rule (KVR):

Design and Modeling of Substrate Integrated Waveguide based Antenna to Study the Effect of Different Dielectric Materials

Alternating-Current Circuits

Experiment 8 Series DC Motor (II)

Th ELI1 09 Broadband Processing of West of Shetland Data

Transformerless Three-Level DC-DC Buck Converter with a High Step-Down Conversion Ratio

Rectangular type APS-10 to 15 Series

Effect of High-speed Milling tool path strategies on the surface roughness of Stavax ESR mold insert machining

A New Algorithm to Compute Alternate Paths in Reliable OSPF (ROSPF)

Area-Time Efficient Digit-Serial-Serial Two s Complement Multiplier

THE STUDY ON THE PLASMA GENERATOR THEORY FOR THIN DISC AND THIN RING CONFIGURATION

Operation Manual. Addendum. Embroidery Machine. Product Code: 884-T13

University of North Carolina-Charlotte Department of Electrical and Computer Engineering ECGR 4143/5195 Electrical Machinery Fall 2009

MAXIMUM FLOWS IN FUZZY NETWORKS WITH FUNNEL-SHAPED NODES

Dataflow Language Model. DataFlow Models. Applications of Dataflow. Dataflow Languages. Kahn process networks. A Kahn Process (1)

Simulation of Transformer Based Z-Source Inverter to Obtain High Voltage Boost Ability

AN ANALYSIS ON SYNTHETIC APERTURE RADAR DATA AND ENHANCEMENT OF RECONSTRUCTED IMAGES

Modal Analysis as a Means of Explaining the Oscillatory Behaviour of Transformers

CHAPTER 2 LITERATURE STUDY

PB-735 HD DP. Industrial Line. Automatic punch and bind machine for books and calendars

ABOUT THIS MANUAL ABOUT THIS MANUAL

Radiant systems 0801EN March 2016 Radiant plasterboard ceiling and/or floor system ISO /7

Use of compiler optimization of software bypassing as a method to improve energy efficiency of exposed data path architectures

Calculation of Off-Core Inductance in Dual-Circuit Model of Transformer

SUPPLEMENTARY INFORMATION

This is a repository copy of Design Guidelines for Fractional Slot Multi-Phase Modular Permanent Magnet Machines.

GEH-2024E Supersedes GEH INSTRUCTIONS MULTICONTACT AUXILARY RELAY. TYPE HFASl. www. ElectricalPartManuals. com. GENERAL fl ELECTRIC

Experimental Application of H Output-Feedback Controller on Two Links of SCARA Robot

Application Note. Differential Amplifier

Single and multiple optical switches that use freestanding silicon nanowire waveguide couplers

Available online at ScienceDirect. 6th CIRP International Conference on High Performance Cutting, HPC2014

Experimental evaluation of polycrystalline diamond tool geometries while drilling carbon fiber-reinforced plastics

2N6071A/B Series. Sensitive Gate Triacs. Silicon Bidirectional Thyristors TRIACS 4.0 A RMS, V

Maximum power point tracking of PV system under partial shading conditions through flower pollination algorithm

Joanna Towler, Roading Engineer, Professional Services, NZTA National Office Dave Bates, Operations Manager, NZTA National Office

Section 2.2 PWM converter driven DC motor drives

S1 Only VEOG HEOG. S2 Only. S1 and S2. Computer. Subject. Computer

GEI E INSTRUCTIONS M ULTI-CONTACT AUXILIARY RELAYS TYPE HFA53. GE Protection and Control 205 Great Valley Parkway Malvern, PA

Experiment 3: The research of Thevenin theorem

PRO LIGNO Vol. 11 N pp

(CATALYST GROUP) B"sic Electric"l Engineering

Leaky Wave Antennas Designed on a Substrate Integrated Waveguide

Influence of spectral irradiance measurements on accuracy of performance ratio estimation in large scale PV systems

Software for the automatic scaling of critical frequency f 0 F2 and MUF(3000)F2 from ionograms applied at the Ionospheric Observatory of Gibilmanna

Transcription:

Sttic nd dynmic ck-coron chrcteristics Ginluc Bcchieg, Ivo Gllimerti, IRS srl - vi Vigonovese 81-35124 Pdov - Itly (e-mil: cchieg@irswe.it) Véronique Arrondel, Philippe Rizer, Jérôme Lecointre, Michel Hmlil EDF R&D 6, qui Wtier 78 41 Chtou - FRANCE (e-mil : veronique.rrondel@edf.fr) Astrct Voltge-current chrcteristics re sic tool to nlyse the opertionl conditions of n electrosttic precipittor, nd to detect the onset of Bck-coron, tht strongly ffects the collection performnces. In this investigtion, lortory set-up hs een rrnged, with clssicl ESP wireplte configurtion; oth sttic nd dynmic voltge-current chrcteristics hve een mesured nd nlysed with different geometries, dust chrcteristics, nd lyer thickness. Similr results hve een lso otined in n industril ESP, nd compred with the lortory results. In the lortory ESP, the sttic voltge-current chrcteristics nd the collection efficiency hve een mesured s functions of time; with high resistivity dust, ck-coron tkes plce s lyer thickness increse, with decrese of the efficiency nd shrp rise of the current. The effect of electrode type nd powder resistivity hs een nlysed. With vrying pplied voltges, the dynmic voltge-current chrcteristics show n hysteresis cycle, the recorded vlues depending on the slope of the pplied wveform. The shpe of the dynmic chrcteristics chnges significntly with the onset of ck-coron. A self-consistent numericl model hs een pplied to simulte sttic nd dynmic voltgecurrent chrcteristics in sence of ck-coron. These simultions llowed to nlyse the retrding effects of the different physicl process (cpcitive chrging, ion drift, spce chrge movements), nd to estimte the corresponding time constnts. Finlly, the dynmic effects of ck-coron hve een nlysed. A second set of experiments hve een relized in 6MW power plnt: ESP sttic nd dynmic chrcteristics hve een mesured with fly sh of low nd high resistivity: the results re very similr to those otined in the lortory ESP, nd confirm tht it is possile to define specific signture, on oth sttic nd dynmic V-I chrcteristics, of the different physicl processes ssocited to the prticle cpttion, including ck-coron. 1 Introduction The collection efficiency of n Electrosttic Precipittor depends on lrge numer of geometricl nd physicl prmeters; the most importnt of them is the current emitted from the High Voltge electrodes nd collected y the pltes [1]: for tht reson the Voltge-Current chrcteristic, nd the ctul working point on them, re used generlly used to define the ESP operting conditions [2]. It is well known tht in generl the higher is the current the etter re the collection performnces. However, with high resistivity fly sh, ckcoron dischrges my form in the lyer of deposited dust, leding to current increse, with reduction of the collection performnces. Bckcoron is series of micro-dischrges in the ir spces within the collected dust lyer [3]: it forms when n excessive field is induced in the lyer y the chrge flowing through its surfce: it is therefore relted to the locl vlues of the current density. Bckcoron forms t spots on the surfce of the collecting pltes, cusing re-entrinment of previously collected dust. The current rises fter ck coron onset, nd tends to e tunnelled through existing ck coron spots: this ccelertes the ck coron growth. Positive ions, injected y ck coron into the volume of negtively energised precipittor, hve the following effects [4]: p.1

- dischrge previously chrged dust prticles; - ttch positive chrge to dust prticles, cusing their drift ckwrds to the emitting electrodes; - reduce the spce chrge, nd therefore increse the electricl field on the emitting electrode nd the coron current; - uild-up nodule shped dust deposits on the emitting electrode, tht further increse locl field nd coron current; - form sprks t much lower voltge level. In the opertion of n industril ESP it is necessry to void ll these negtive effects of ckcoron: for this good dignostic tools re needed, sed on the modifictions tht ckcoron induces on esily mesurle prmeters. 2 Voltge-current chrcteristics in lortory ESP model 2.1 Lortory ESP description The experimentl tests hve een performed in the Test Lortory of SAMES in Grenole (Frnce) on smll scle verticl precipittor (Figure 1), mde of rectngulr trnsprent PVC tue (.3 x.3 m), with three fields, ech relized with couple of pltes (Pl.1,Pl.2 nd Pl.3), nd two emitting electrodes. In order to test the effect of the geometry, cylindricl wires five millimetres in dimeter nd rion nds with points hve een used s emitting electrodes. The pplied voltge ws generted y three DC switching power supplies (6kV, 5 µa), with low ripple (less thn 1% pek to pek), controlled y remote PC: this enled to progrm different voltge wveshpes on ech field (rmp, squre, etc.). During test opertion constnt ir flow of 27 m 3 /h, t room temperture, ws supplied to the ESP. Different types of dust were fed t the top of the precipittor duct : synthetic powders (époxy polyester, with different metl dditives to vry the resistivity) nd col fly sh collected in the hoppers of n industril ESP. The size distriution of the synthetic prticles ws concentrted elow 1 um, while tht of the col fly sh ws etween 1 nd 1 um. The prticle resistivity could vry from medium to very high vlues (1 1 to 1 12 ohm cm). An lmost constnt prticle concentrtion (1 gr/m 3 ) ws mintined t the ESP inlet during ll the test period. Voltge nd current mesurements were performed t ech genertor, with fst recording oscilloscope nd dtlogger; the ir nd powder flow rtes were systemticlly checked, to estlish the inlet concentrtions; the collecting pltes were periodiclly removed nd weighted (t specified time intervls 15, 3, 6 min, 1, 2, 4, 6 hr), to estlish the deposition rte nd the collection efficiency of ech field. p.2 Figure 1 : Lortory precipittor lyout

2.2 Current chrcteristics The coron current, under negtive DC voltge, hs specific ptterns [5,6]]: t onset level low mplitude pulses pper, with regulr repetition rte (Trichel Pulses); s the voltge is incresed, the frequency of the Trichel Pulses increses, while continuous component is developed (Glow Coron); t even higher voltge second impulse component pper (Pre-rekdown Stremers), with lrge mplitude nd rndom pttern (Figure 2 ). If the sttisticl distriution of pulse mplitude is nlysed, typicl doule popultion is oserved; in the present cse (Figure 3), with dust lod ut no ck-coron, the Trichel Pulses present nrrow distriution centred round 1 µa, while the prerekdown stremers present wider distriution with mximum round 25 µa. When ck-coron is estlished, the current pttern nd the sttisticl distriution of pulse mplitude chnge completely ( Figure 2 ): new popultion ppers of very smll pulses centred round 5 µa, with rndom repetition rte, while the two previous ones dispper (Figure 3). These pulses hve to e ssocited to the ck-coron microdischrges in the dust lyer deposited on the pltes; due to their shrp form nd rndom repetition rte they produce wide H.F. electromgnetic noise, tht mkes difficult to otin clen mesurements. 4 Current ( µa ) 2 4 Current ( µa ) 2 1 2 3 4 5 6 Time (µs) Figure 2 : current signl versus time; : with dust lod, V= 4 Kv, elow ckcoron thresold; : with dust lod, V= 5 Kv, ove ckcoron thresold 4 3 Pek numer 2 1 1 2 3 4 5 Mximum pek current (µ A) Figure 3 : Sttisticl distriution of current pulse mplitude; : with dust lod, V= 4 Kv, elow ckcoron thresold; : with dust lod, V= 5 Kv, ove ckcoron thresold p.3

2.3 Sttic Voltge-Current chrcteristics The sttic Voltge-Current chrcteristics of ech field hve een mesured with step y step slowly incresing voltge wveshpe: with no dust lod, ove the coron inception threshold, ll the chrcteristics present typicl qudrtic dependence (see curve in Figure 4). When high resistivity powder is injected t top of the duct, it is possile to see tht the voltge-current chrcteristics chnge in time: this vrition is fster for the first plte nd slower for the third one. The chrcteristics of Pl.3 re presented in Figure 4, fter deposition times of 2, 4, 6 minutes): s the dust lyer deposited on the pltes increses in thickness, ck-coron dischrge forms within the lyer nd rises the dischrge current. Consequently the mesured curves move to higher currents, nd chnge shpe from qudrtic to liner, ove the ck-coron threshold. In norml opertion, the injection of low resistivity prticles lowers the current vlues, moving the voltge-current chrcteristics to the right of the reference curve, with no dust lod: this my e ttriuted to the ttchment of prt of the ionic chrge to the prticles, tht hve much lower moility; in ddition, their spce chrge reduces the electric field on the emitting electrodes, thus reducing the coron current. On the contrry, in the present cse, even in sence of ck-coron, with short deposition times the chrcteristics is moved to the left of reference curve. This hs to e ttriuted to n increse of the wire emissivity, ssocited to dust deposition on the wire surfce, tht increses the surfce roughness nd therefore the locl field distortion: in this cse, the field reduction due to the prticle spce chrge is compensted y the field increse due to locl distortion. 5 4 d Current (ma) 3 2 c 1 1 2 3 4 5 6 7 8 Figure 4 : Voltge-Current chrcteristics for different deposition times (Pl.3 - cylindricl wires; = min; =2 min; c=4 min; d=6 min) The estlishment of ck-coron lso ffects the prticle collection efficiency (see Figure 5). At the eginning of the prticle injection, ech plte hs slightly different efficiency (with n verge round 65%), depending essentilly on prticle size distriution nd concentrtion t the inlet of ech plte. As the dust lyer thickness increses nd the ck-coron forms on the pltes, the prticle collection efficiency decreses shrply nd flls down to out 2%; the overll efficiency flls from 96% to 54%. 1% Efficiency 8% 6% 4% Tot Pl. 1 Pl. 2 Pl. 3 2% % 1 2 3 4 5 6 Time (min) p.4 Figure 5 : Collection efficiency of the different pltes s function of deposition time (cylindricl wires, V=5 kv)

Sttic voltge-current chrcteristics with ck-coron re influenced y the prticle resistivity. Figure 6 show how curves chnge s powders with different resistivity nd humidity re used. Curve is given s reference without prticles; curve nd c show the results for medium nd high resistivity; curve d show the results for very high resistivity. As the resistivity increses, the field cross the deposited dust lyer increses, together with the ck-coron intensity nd the dischrge current. 5 4 3 2 d c 1 1 2 3 4 5 6 7 8 9 Figure 6 : Effect of prticle resistivity on Voltge-current chrcteristics (Pl. 1 - cylindricl wires; = no prticle lod; = medium; c = high; d = very high) The Voltge-current chrcteristics re lso influenced y the geometry of the ESP electrodes. In Figure 7 the results re shown for two types of emissive electrodes: nds with points () nd cylindricl wires of 5 millimetres in dimeter (). With no dust lod, the coron voltge threshold is lower for the shrper geometry, nd higher current vlues re otined for lower pplied voltges. A similr trend is shown y the curves with high resistivity prticles (1 nd 1): with the shrper geometry lso the ck-coron threshold ecomes lower. 5 4 1 2 3 2 1 2 1 1 2 3 4 5 6 7 8 9 Figure 7 : Effect of the geometry of emissive electrodes on Voltge-Current chrcteristics (PL: 1; : nd with points; : cylindricl wires; 1 : with dust, 2 : without dust) It is well known from the literture [7] tht the Voltge Current chrcteristics hs qudrtic form. In the present cse the form k*u*(u U )hs een dopted, where U is the ctul voltge, U the coron inception voltge, nd k field distortion coefficient, tht represents the effect of protrusions or points on smooth surfce: it hs e ssumed 1. for the wire, nd 1.43 for the nd with points. With this normliztion, in sence of ckcoron, the chrcteristics re liner, with n verge slope etween.8 nd.1 ua/kv 2 ; fter ckcoron inception, the slope suddenly chnges, nd identifies with no miguities the inception voltge (see Figure 8). This procedure my e useful in prcticl pplictions. p.5

5 4 3 2 1 1 2 2 1 1 2 3 4 5 6 k U(U-U ) (kv 2 ) Figure 8: Curve normliztion (PL: 1; : nd with points; : cylindricl wires; 1 : with dust, 2 : without dust; 2, 2 : points = experimentl, lines = simultion results) In order to verify the sttic chrcteristics on the se of the physicl processes, the ORCHIDEE simultion code [8,9,1,11] hs een pplied to the specific conditions of the lortory ESP. The results re reported s solid lines in Figure 8, while the points represent the experimentl results: the greement is quite stisfctory. 2.4 Dynmic voltge-current chrcteristics The sttic Voltge-Current chrcteristics hve een mesured y pplying step y step slowly incresing voltge; if fst vrying voltge wveshpe is pplied, the dynmic chrcteristics show n hysteresis cycle, in oth cses with nd without dust lod: n exmple of this ehviour is presented in Figure 9, without dust lod, for nds with points s emissive electrodes, nd intermittent energistion (trpezoidl wve shpe with rise nd fll times of the order of 3 ms). The mesured current vlues re different depending on whether the pplied voltge is incresing or decresing: with rising slope the current is higher thn the sttic vlue, the opposite eing true with decresing slope. This type of dynmic hysteresis cycle is typicl of processes with delyed response, which my chrcterized y Time Constnt: the mplitude of the hysteresis cycle depends on the steepness of the pplied vrition, specificlly on the rtio etween the Time Constnt nd the rise or fll times. 5 4 3 2 1 rise sttic fll 1 2 3 4 5 6 7 8 Figure 9 : Voltge-Current chrcteristics: sttic nd dynmic (PL. 1; nds with points; no dust lod) In the present cse, without dust lod, the current is crried y the negtive ions generted in the coron dischrge: with trnsient voltge, due to the finite moility, their distriution etween wires nd pltes my respond to the voltge vrition only with the dely of their trnsport processes. Therefore, ll the physicl prmeters ssocited to the current emission (spce chrge, electric field on the wire surfce, intensity of the coron dischrge) respond to voltge trnsient with the Time Constnt of the ion trnsport. p.6

5 4 3 ms 3 2 1 sttic 3 ms 3 ms 3 ms 1 2 3 4 5 6 7 Applied Voltge (kv) Figure 1 : Experimentl (points) nd simulted (lines) sttic nd dynmic Voltge-Current chrcteristics (rise-fll time for simultions: 3 nd 3 ms) In order to descrie in detil the dynmics of the physicl processes, the ORCHIDEE simultion code [11,12] hs een pplied to the specific conditions of the lortory ESP: trpezoidl voltge wveshpe hs een pplied, with different rise nd fll times, with no dust lod. The simultion results re presented in Figure 1; the lines represent the computed dt, the points the experimentl results. It cn e clerly seen tht the mplitude of the hysteresis cycle depends very much on the voltge vrition rte; for rise nd fll time of 3 ms, the experimentl nd computed results re in quite good greement. The computed results show tht, on the rising side of the dynmic chrcteristics, the voltge is in dvnce with respect to the spce chrge distriution: in fct, the chrges generted round the wire need time to cross the gp nd fill the volume; s consequence of the reduced chocking effect of the spce chrge, the electric field on the wire surfce is higher, leding to higher current emission. The opposite is true in the flling side: the dely in the spce chrge dissiption sustins the chocking effect, nd therefore reduces the field nd the current. The sme ehviour is oserved with the dust lod, in sence of ck-coron (clen pltes): the mplitude of the hysteresis cycle is slightly higher, indicting longer Time Constnt for spce chrge trnsport; this is clerly due to the ttchment of ions to the dust prticles, nd the consequent reduction of the chrge verge moility. The reduced Time Constnt my e ssumed to result from weighted verge of the moility Time Constnts for negtive ions nd chrged prticles. A very different ehviour is oserved fter the inception of ck-coron (Figure 11): while the sttic curve steeps up rpidly, the rising side of the dynmic curve increses less rpidly thn the corresponding sttionry vlues; the dynmic vlues pss from the left to the right side of the sttic chrcteristics. This indictes dely in the estlishment of the ck-coron dischrge within the dust lyer: it my e represented with new Time Constnt, which is switched on ove the ckcoron threshold. It is worth noting tht this dely hs nothing to do with tht oserved in pr. 2.3, 5 4 3 2 1 rise sttic fll 1 2 3 4 5 6 7 8 Figure 11 : Voltge-Current chrcteristics: sttic nd dynmic (PL. 1; nds with points; dust lod) p.7

which ws essentilly ssocited to the rte of lyer deposition on the pltes. An opposite ehviour is oserved in the flling prt of the voltge wve (Figure 11): however, the dely for ck-coron extinction ppers to e quite short thn tht for its inception. The shpe of 8, with the crossing of the two sides of the dynmic chrcteristics round the ck-coron inception, seems to e typicl signture of the presence of ck-coron dischrges within the deposited dust lyer: the mplitude nd form the dynmic cycle my e ssocited to the different time constnts for chrge trnsport nd dischrge formtion. 3 Voltge-current chrcteristics in n industril ESP 3.1 Industril ESP description The tests on the industril precipittor hve een relized in 6 MWe col urning unit of the EDF Power Plnt in Le Hvre (Frnce). The ESP instlled t the oiler outlet is relized with three prllel csing: ech of them is composed y 5 fields; the first nd the lst of these fields re equipped with trditionl power supplies, while the other ones re equipped with new DC switching power supplies le to feed low ripple voltge (+/- 1% pek to pek). Ech field hs 6 gs ducts with spcing of.28 m, nd 7 collecting pltes (.46 m long nd 12.4 m high). Emissive electrodes re 2 red wires per plte, plced inside rigid frmes t distnce of.12 m ech other. Rpping equipment is relized y flling hmmers. During the tests, the electrofilter flue gs flow vried from 1.25 1 6 Nm 3 /h to 1.6 1 6 Nm 3 /h for electricl lods vrying from 41 to 53 Mwe; the fly sh concentrtion ws roughly constnt round 9 g/nm3 (6% dry O2). Four test cmpigns hve een performed etween Mrch nd April 22; four different cols hve een used, to test different fly sh resistivities nd gs flow rtes: South Africn col ( 7 1 1 Ohm cm Pe = 517 Mwe), South Africn col (5 1 1-2 1 11 Ohm cm Pe = 4 MW), Austrlin col (3 1 9 Ohm cm Pe = 52 MW), Polish col (very low resistivity Pe = 5 Mwe). 3.2 ESP sttic voltge-current chrcteristics The industril test hve een relized in order to verify tht the typicl signtures of the ckcoron formtion nd development, oserved in the lortory ESP model, cn e effectively oserved lso in full scle industril plnt. The sttic Voltge-Current chrcteristics of ech field hve een mesured with step y step slowly incresing voltge. Preliminrily, test hs een relized with the plnt urning high resistivity South Africn col, just fter the rrest of the unit. With cold precipittor (6 C) nd without dust lod, two sttic chrcteristics hve een recorded, efore nd fter the evcution of the dust lyer from the pltes y rpping (Figure 12). It cn ee clerly seen, ove 28 kv, the typicl pttern of ckcoron inception in the deposited dust lyer. 5 4 Current (ma) 3 2 1 1 2 3 4 5 Figure 12 : Sttic Voltge-Current chrcteristics with () nd without () dust lyer on the pltes p.8

Similr results hve een otined under norml operting conditions: Figure 13 shows the sttic chrcteristics otined in three different cmpigns, respectively with high (), medium () nd low (c) col resistivity. 5 Current (ma) 4 3 2 c 1 1 2 3 4 5 Figure 13 : Sttic Voltge-Current chrcteristics in three different cmpigns, respectively with high (), medium () nd low (c) col resistivity. 3.3 ESP dynmic voltge-current chrcteristics The dynmic Voltge-Current chrcteristics hve een recorded y pplying n fst intermittent voltge wveshpe: lso in the industril plnt, they show n hysteresis cycle, in oth cses with nd without dust lod: n exmple of this ehviour is presented in Figure 14, with dust lod of low resistivity of the dust prticles. As in the lortory cse, the mesured current vlues re different depending on whether the pplied voltge is incresing or decresing: with rising slope the current is higher thn the sttic vlue, the opposite eing true with decresing slope. 5 4 Current (ma) 3 2 1 rise sttic fll 1 2 3 4 5 Figure 14 : Voltge-Current chrcteristics: sttic nd dynmic (Polish col, with low resistivity). As in the lortory cse, very different ehviour is oserved with high resistivity dust (5): when the sttic curve steeps up rpidly, the rising side of the dynmic curve increses less rpidly pssing from the left to the right side of the sttic chrcteristics; the opposite is true for the flling side of the chrcteristics. Also in the industril ESP, the shpe of 8, with the crossing of the two sides of the dynmic chrcteristics round the ck-coron inception, seems to e the typicl signture of the presence of ck-coron dischrges within the deposited dust lyer. p.9

5 4 Current (ma) 3 2 1 rise sttic fll 1 2 3 4 5 Figure 15 : Voltge-Current chrcteristics: sttic nd dynmic (South Africn col, with high resistivity). Conclusions In this pper the typicl signtures, tht identify the onset of ckcoron, hve een presented nd discussed: - chnge in the impulse component of the dischrge current, which moves the sttisticl distriution of the pek mplitudes towrds lower vlues; - steep chnge of the sttic Voltge-Current chrcteristics t onset threshold, prticulrly evident on normlized curves; - the ppernce of 8 shped form in the dynmic Voltge-Current hysteresis cycle, with crossing point round the onset conditions; The nlysis of these signtures my e recommended s technique to determine the onset of ck coron, nd to nlyse the susceptiility of precipittor to ck coron, under specific operting conditions. Acknowledgements The uthors wish to thnk the Compny SAMES in Grenole, its technicl director Cryl Thomé, nd its personnel, for the use of the lortory equipment, nd for the help in running the tests. The uthors re lso indet with the personnel of the Le Hvre power plnt for the prticiption to the industril ESP tests. References [1] S. OGLESBY, G. NICHOLS: Electrosttic precipittion, Dekker, New York, 1978 [2] R.J. TRUCE, W. REIBELT: "New technologies improves electrosttic precipittor performnces, VII ICESP*, Kyongjiu (Kore), 1998 [3] S. MASUDA, A. MIZUNO: Initition condition nd mode of ck-dischrge, Journl of Electrosttics 4, 1977-78 [4] R.J. TRUCE, Bck Coron nd its Effects on the Optimistion of electrosttic Precipittor Energistion control, Seventh Interntionl Clen Air Conference, August 24-28, 1981 [5] L. B. LOEB: Electricl Corons, University of Cliforni Press: Berkeley, CA,1965 p.1

[6] A. VALAGUSSA: Prove di lortorio per lo studio dell cptzione elettrosttic, Rpp. CESI L-93/3138, 1995 [7] I. GALLIMBERTI : The mechnism of the long sprk formtion, J. de physique, colloque C7, supplement u n.7, Tome 4, Juillet 1979 [8] I. GALLIMBERTI, A. GAZZANI, U. TROMBONI, E. LAMI, F. MATTACHINI, G. TREBBI:"Physicl simultion of the prticle migrtion in ESP, Prt I - Model description", VI ICESP*, Budpest, 1996. [9] I. GALLIMBERTI, A. GAZZANI, U. TROMBONI, E. LAMI, F. MATTACHINI, G. TREBBI:"Physicl simultion of the prticle migrtion in ESP, Prt II - Appliction results", VI ICESP*, Budpest, 1996. [1] E. LAMI, F. MATTACHINI, I. GALLIMBERTI, R. TURRI, U. TROMBONI: "A Numericl Procedure for Computing the Voltge-Current Chrcteristics in ESP Configurtion", Journl of Electrosttics 34 p.385-399, 1995. [11] I. GALLIMBERTI: "Recent dvncements in the physicl modelling of electrosttic precipittors", Journl of Electrosttics 43 (1998) 219-247, 1998. [12] V. ARRONDEL, G. BACCHIEGA, I.GALLIMBERTI: "ESP modelling: from University to Industril Appliction", VIII ICESP, Birminghm (USA), 21 p.11