IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 1, JANUARY 2006 273

An EMC Evaluation of the Use of Unshielded MotorCables in AC Adjustable Speed Drive ApplicationsNorbert Hanigovszki, Member, IEEE, Jørn Landkildehus, Member, IEEE, Giorgio Spiazzi, Member, IEEE, and

Frede Blaabjerg, Fellow, IEEE

Abstract—The most common solution for modern adjustablespeed drives (ASD) is the use of induction motors (IM) fed byvoltage-source inverters (VSI). The inverter generates a pulsewidthmodulated (PWM) voltage, with values of about 6 kV s oreven more. In three-leg inverters for three-phase applications theoccurrence of common-mode voltage is inherent due to asymmet-rical output pulses. As a result, for electromagnetic compatibility(EMC) reasons, in most applications shielded cables are usedbetween the inverter and the motor, implying high installationcosts. The present paper discusses the use of cheaper, unshieldedcables from an EMC point-of-view. A new method for measuringelectromagnetic interference (EMI) from unshielded cables isproposed and measurement results are presented. The level ofEMI is evaluated in different situations: without an output filter,with a classical LC output filter and with an advanced output filterwith dc link feedback. It is concluded that, from an EMC point ofview, unshielded cables can give very good performance providedthat a common-mode (CM) output filter is used.

Index Terms—Adjustable speed drives (ASD), common-mode(CM) output filter, electromagnetic compatibility (EMC), induc-tion motors (IM), pulsewidth modulated (PWM), voltage-sourceinverters (VSI).

I. INTRODUCTION

THE standard solution used in industry for ac adjustablespeed drives (ASD) remains the pulsewidth modulated

(PWM) voltage source inverter (VSI) [1] although a lot ofresearch has been done on alternative topologies [2]–[9]. TheVSI presents many advantages: high efficiency (up to 98%),open-circuit protection, small relative size, excellent regulationcapabilities, wide speed range, constant high input powerfactor, common bus regeneration, and others. To achieve highefficiency levels, fast switching insulated gate bipolar transis-tors (IGBTs) are used. As a consequence, at the output of theVSI, PWM, pulses with high (6 kV s or higher) willbe present [10]. In a three-phase inverter, PWM also generatescommon-mode (CM) voltage because of asymmetrical voltagepulses. In Fig. 1, a switching pattern is shown where it canbe observed how the CM voltage is generated. This voltage is

Manuscript received August 5, 2004; revised April 13, 2005. Recommendedby Associate Editor P. Tenti.

N. Hanigovszki and J. Landkildehus are with Danfoss Drives A/S, Graasten6300, Denmark (e-mail: norbert@danfoss.com).

G. Spiazzi is with the Electronic Engineering Department, University ofPadova, Padova 35131, Italy.

F. Blaabjerg is with the Department of Electrical Engineering, Aalborg Uni-versity, Aalborg DK-9100, Denmark.

Digital Object Identifier 10.1109/TPEL.2005.861182

Fig. 1. Common-mode voltage generation in a PWM inverter.

obtained by adding the three voltages resulting from the threeinverter legs

VV V V

(1)

Having the dc-link mid-point as a reference, the voltages V ,V , and V can either be V or V , where V isthe dc-link voltage. As a result, depending on the state of theswitches, the common mode voltage can take one of the values

V 2 (when all three legs are connected simultaneously toeither the positive or the negative bus rail) and V 6 (whentwo legs are connected to one bus rail and the third one is con-nected to the opposite rail). As a consequence, a usual three-leginverter can not have an output CM voltage equal to zero, andthis voltage is bigger ( V 2) if the zero vector is employed(all three legs in the same switching state) [11], [12].

0885-8993/$20.00 © 2006 IEEE

274 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 1, JANUARY 2006

TABLE IPRICE COMPARISON BETWEEN SHIELDED AND UNSHIELDED CABLES

The CM voltage in the motor cable of a three-phase PWMdrive with a three-phase diode rectifier on the grid side will con-sist of the following components.

• A component at 150 Hz due to the three-phase diode rec-tifier.

• A component at the switching frequency and its subse-quent harmonics.

• A broadband high-frequency component due to the steepPWM pulses.

The components due to the switching frequency and thehigh-frequency noise can cause electromagnetic interference(EMI) to neighbor equipments and installations by eitherradiation from the motor cable or by cross-talk (mainly capac-itive coupling) with other conductors. There are well-definedmethods for measuring radiated noise above 30 MHz whichare presented in various standards. Below 30 MHz the mea-surement of radiated EMI from ASD installations is ratherdifficult, as the noise depends very much on the measurementset-up, motor cable length, type and its layout. It is much easierto use a measurement method for conducted common-modenoise in the motor cable using a well designed set-up. Suchameasurement method applied for unshielded motor cables isthe scope of the present work.

The CM voltage with high and high amplitude in thecable between the inverter and the motor is a potential sourceof EMI [13]–[17], therefore usually shielded motor cables areused. Such cables are expensive and increase the overall installa-tion cost. Table I presents a price comparison between shieldedand unshielded cables. The prices are based on the catalogue ofan arbitrary European cable manufacturer. It can be observedthat unshielded cables are between 30% and 70% cheaper thanshielded cables. Unfortunately, their use is restricted by elec-tromagnetic compatibility (EMC)-related concerns. A completeinvestigation from an EMC point of view, about the use of un-shielded motor cables, has not been presented in the literatureuntil now, although the issue of motor cables for ASD applica-tions has been discussed in such papers as [18]. In the following,a new method for measuring conducted CM noise on unshieldedmotor cables will be proposed. This method will be further usedfor evaluating the EMI in the case of an inverter without output

Fig. 2. Investigated output filter topologies: (a) no filter, (b) LC filter, and (c)filter with dc link feedback.

filter [see Fig. 2(a)] and the effect of two different output filtertopologies: a classical filter [see Fig. 2(b)] and a more ad-vanced filter with dc link feedback [see Fig. 2(c)]. These filtersare often encountered in industrial applications and their pri-mary use is to mitigate the secondary effects of high andPWM modulation pattern: motor terminal overvoltages and in-sulation stress, bearing currents, acoustic switching noise, etc.Besides this, output filters are sometimes also used to enable theinstallation of unshielded cables.

II. CONDUCTED CM NOISE MEASUREMENT

ON UNSHIELDED MOTOR CABLES

The measurement of conducted CM noise in the cable be-tween the inverter and the motor is not an easy task, especiallybecause of the high levels of CM voltage at the switching fre-quency (practically the peak-to-peak value of the CM voltage isV ) and the steep voltage pulses. The use of a line impedancestabilization network (LISN) on the output of the inverter isob-viously excluded because of the above-stated reasons. Conse-quently, a nonintrusive method has to be used. Previous work[19] confirmed the possibility of using a set-up from the CISPR22/EN55022 [20] standard for measuring conducted CM noiseon shielded motor cables. Even if CISPR 22 is not a standardfor electric drives, but for IT equipment, the “in situ CDN/ISN”configuration was proved to be usable for measuring CM noiseproduced by an ASD when a shielded cable is used.

Based on previous experience with shielded cables, the con-figuration C.1.3. from the CISPR 22 standard [20] was used forCM noise measurements on unshielded cables. This set-up is a

HANIGOVSZKI et al.: EMC EVALUATION 275

Fig. 3. Set-up C.1.3. for measuring CM noise on unshielded cables, describedin CISPR 22 [20].

Fig. 4. Construction of the capacitive voltage probe [25].

combination of current probe and capacitive voltage probe (seeFig. 3) for measurements on unshielded cables. EUT representsthe ASD and AE is the induction motor. The acceptance criteriais to comply with both the current and voltage limits. This re-quirement is because of the unknown CM impedance of the aux-iliary equipment. The particularity of this measuring method isthe lack of a well defined CM impedance, unlike other set-upsthat require a 150- CM impedance. (The high CM voltagelevels due to the three-phase PWM pulse pattern and highmake the use of a well-defined CM impedance impractical.)

A first attempt to build the capacitive voltage probe was towrap an aluminum foil around the unshielded cable, on thelength of 55 cm. (The standard requires a minimum length of50 cm.) The aluminum foil was fixed with copper tape, and atap was made at the middle of the clamp, also using coppertape. An insulator keeps the cable at a constant distance fromthe ground plane (10 cm). The cable length used was 5 m forpractical reasons, in order to keep the cable straight over theground-plane. The whole set-up was installed in a shieldedroom, to avoid interference from external fields.

The CM voltage was picked up between the tap and theground plane. For visualizing the voltage a high-impedanceprobe (10 M ) and a digital oscilloscope were used. The digitaloscilloscope was also used for data acquisition for furtherprocessing [fast Fourier transform (FFT) analysis].

For noise measurements a spectrum analyzer was connectedto the capacitive clamp with a standard voltage probe, asdescribed in CISPR 16 [21]. However, because of the lowimpedance of the voltage probe (1500 ), this probe couldbe used only for measurements above 1 MHz, where theimpedance of the capacitive voltage clamp becomes lowerthan the impedance of the voltage probe. For frequenciesbelow 1 MHz the only way of observing the spectrum of the

Fig. 5. Measurement circuit with the capacitive voltage probe.

CM voltage noise is to perform a FFT analysis of the voltageacquired with the digital oscilloscope.

The method used for building the capacitive clamp presentedabove has several drawbacks: low immunity to external fields,the capacitance of the probe is not well defined since both thelength of the probe and the diameter of the cable can vary fromone test to another. Several papers [22], [23] as well as workwithin standardization groups suggest the use of a more ad-vanced voltage probe which is also commercially available andmentioned in the latest edition of the CISPR 16 standard [24].This high-impedance voltage probe consists of two electrodes,one inner electrode for picking up the signal from the cable andthe other one for shielding the first electrode from external elec-tric fields.

The signal from the capacitive clamp is applied to a trans-impedance amplifier, situated in the immediate vicinity of theclamp. The mechanical construction of such a capacitive voltageprobe is shown in Fig. 4 [25]. The voltage division depends onthe type of cable and it still requires separate calibration foreach cable type. Fig. 5 shows the equivalent measurement cir-cuit of the capacitive voltage clamp. The common-mode voltageis symbolized V and it is capacitively coupled to the innerelectrode of the voltage clamp by means of the capacitances

between the cable conductors and the inner electrode. Thisvoltage is applied to a trans-impedance amplifier characterizedby a gain , an internal capacitance and an internal resis-tance . represents the capacitance between the inner elec-trode and the outer electrode (electrostatic shield) which is con-nected to ground. At the output of the trans-impedance amplifierthe voltage is measured by means of an EMI receiveror spec-trum analyzer. From the equivalent circuit the following voltagedivision factor can be derived

V(2)

Unfortunately the commercially available capacitive voltageprobe can not measure voltage levels as high as the

276 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 1, JANUARY 2006

Fig. 6. Voltage on the capacitive clamp: (a) no filter, (b) LC filter, and (c) filter with dc link feedback.

common-mode voltage produced by a PWM inverter, there-fore this probe has been used only when experimenting withcommon-mode output filters.

With the capacitive voltage probe tests were also made witha 100-m unshielded motor cable. The cable was coiled for prac-tical reasons and the tests were done in different situations: cableat different heights over the ground plane, cable coil on theground plane and cable coil on the ground plane wrapped inaluminum.

For the current measurement, a standard high frequency cur-rent probe was used. The measurement of the CM current doesnot pose any special problems, as the value of the CM currentis relatively low.

III. MEASUREMENT RESULTS

Two sets of experiments were conducted. A first set of exper-iments aimed to evaluate CM noise on the motor cable when nooutput filter is used and when the investigated output filters areused. For this purpose, the CM voltage was measured with the

first method (the aluminum foil wrapped around the unshieldedcable). The second set of experiments focused on the use of thefilter with dc link feedback which is very effective both in DMand CM. The aim of this second set of tests was to evaluate, asaccurately as possible, whether unshielded motor cables can beused without causing unacceptable EMI levels. A commercialcapacitive voltage probe was used for these experiments.

For the tests a 3-kW induction motor at 400 V was fed bya 3-kW commercial ASD, running with 13-kHz switching fre-quency and 5-Hz fundamental output frequency. A low outputfrequency, and consequently a low output voltage, was used be-cause it is considered a worst-case situation for the CM voltagegeneration. In this situation the “zero” switching vector that pro-duces the highest CM voltage occurs for a longer period than athigher output frequencies.

A. Experimental Results With a Short Motor Cable

The first set of experiments was conducted using a 5-m un-shielded motor cable. The CM voltage was measured using a

HANIGOVSZKI et al.: EMC EVALUATION 277

Fig. 7. FFT of the voltage on the capacitive clamp: (a) no filter, (b) LC filter, and (c) filter with dc link feedback.

capacitive clamp built using aluminum foil wrapped around theunshielded cable. Fig. 6(a) shows the CM voltage measured onthe capacitive clamp. The peak-to-peak voltage is very high,over 400 V. The FFT of this voltage (up to 1 MHz) in Fig. 7(a)shows a very noisy spectrum, with high-amplitude harmonicsof the switching frequency. Between 1 MHz and 30 MHz, mea-sured with the spectrum analyzer having a resolution bandwidth(RBW) of 1 kHz and peak detector, the CM voltage is still verynoisy. This is shown in Fig. 8(a). In many applications theselevels would be prohibitively high.

Two different output filters have been fitted at the inverteroutput (see Fig. 2) to reduce the noise level. The first one,shown in Fig. 2(b), is a classical filter. It consists ofdifferential-mode (DM) inductors and DM capacitors (alsoknown as “X-caps”). The filter has a cut-off frequency belowthe switching frequency providing a sinusoidal phase-to-phasevoltage at the motor terminals. The filter used in the experi-ments has a 3.8-mH DM inductor and 1.5- F capacitors. Thesecond filter shown in Fig. 2(c) consists of a classical DM filter

followed by a CM inductor and CM capacitors (also known as“Y-caps”) connected to the dc link and rails. The valuesof the DM components are the same as the ones of the previousfilter (3.8-mH DM inductor and 1.5- F DM capacitors). TheCM inductor has 8.3 mH and the CM capacitors are 220 nFeach. Both filters were commercial products and a comparisonwas needed because of reported claims that the LC filter wouldimprove EMI performance and allow the use of unshieldedcables.

With the LC filter installed the CM voltage [see Fig. 6(b)] hasthe same high peak-to-peak value of over 400 V, as in the sit-uation without a filter. In the FFT analysis of the CM voltage[see Fig. 7(b)] no significant improvement can be observed. Inthe 1 MHz–30 MHz range the noise is reduced, but it still has avery high level [see Fig. 8(b)]. The filter with dc link feedback,due to its CM attenuation, drastically reduces the CM voltagemeasured with the capacitive clamp, as shown in Fig. 6(c). TheFFT of this voltage waveform also shows a major reduction inthe noise, as seen in Fig. 7(c). The level of the noise in the

278 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 1, JANUARY 2006

Fig. 8. CM conducted voltage noise in the 1 MHz–30 MHz range: (a) no filter, (b) LC filter, and (c) filter with dc link feedback.

1 MHz–30 MHz range shown in Fig. 8(c) is reduced very much.Just for the sake of comparison, if the CM voltage noise level inthe 1 MHz–30 MHz range is compared with the voltage limitsfrom CISPR 22 (which is a standard for telecommunicationports of IT equipments) it would comply with the class A limit(for industrial environments, 87 db V QP in the 0.5–30 MHzrange).

The current measurements show relatively low noise levelstherefore they are not shown for the tests with the 5 m cable.As in the case of CM voltage measurements, there is only littledifference between the situation without a filter and the situa-tion with the filter, especially in the lower frequency range(below 150 kHz). There is a major improvement when usingthe filter with dc link connection. Again, just for comparisonwith the limits from CISPR 22, the current measurements wouldmatch class B (for home and office environment).

B. Experimental Results With a Long Motor Cable andOutput Filter With dc Link Feedback

A second set of experiments was made with a 100 m cableusing the output filter with dc link feedback which operates both

in CM and DM. Because of this length the cable had to be coiled.Tests were performed in different situations varying the cou-pling between the cable and the ground plane and observing thebehavior of the noise. The cable coil was placed at a 20 cm dis-tance over the ground plane, after that it was directly placed onthe ground plane. We also tried to wrap the cable coil in alu-minum foil and ground it. These experiments were necessary be-cause in reality the arrangement of the cable can vary from oneinstallation to another. If the cable is placed in a metallic conduitor in a cable tray there is a higher capacitance to ground whilea cable feeding a trolley on a gantry crane has a bigger distancefrom the metallic structure and consequently there is a lower ca-pacitance to ground. The experiments were made with the motorboth connected to the ground plane and insulated from it. Sev-eral measurements were made and the results were comparedagainst the limits specified in CISPR 22, only as a reference.The CM voltage measurements were made with the commer-cial capacitive voltage probe that was calibrated for the cabletype used in the experiments. The choice of this voltage probehas two main reasons: a more accurate measurement comparedto the set-up using aluminum foil and a better shielding from

HANIGOVSZKI et al.: EMC EVALUATION 279

Fig. 9. CM conducted voltage noise in the 150 kHz–30 MHz range, 100-m cable: (a) motor and cable on GP, (b) motor insulated, cable on GP, and (c) motor andcable insulated.

external electric fields, especially cross-coupling from the bigcable coil situated in the proximity of the voltage measurementpoint.

Fig. 9 shows the measured CM voltage in three situations:both motor and cable on ground plane in Fig. 9(a), cable onground plane and motor insulated in Fig. 9(b) and both motorand cable insulated from ground plane in Fig. 9(c). It can beobserved that when the cable is placed on the ground plane thenoise level closely complies with CISPR 22 class A limits (in-dustrial environment). Only in the 150–200 kHz region the av-erage limit is slightly trespassed. The situation changes whenboth the motor and the cable are insulated from the groundplane. A resonance occurs at approximately 2.8 MHz, slightlytrespassing the limit. Experiments show a dependance of thispeak on the layout of the set-up. Therefore it is preferable toplace unshielded cables in grounded cable trays and conducts.

The CM current measurements show relatively low levels,similar to the experiments with the short cable. Fig. 10 showscurrent measurements with the motor situated on the groundplane [see Fig. 10(a)] and insulated Fig. 10(b). It can be ob-served that the CM current meets the more strict class B limits(home and office environment).

IV. CONCLUSION

In an ASD application using a PWM VSI, without any fil-tering at the output of the inverter, the CM conducted EMI levelsin an unshielded motor cable are quite high, therefore in someapplications the unshielded cable can not be used, due to EMCconcerns. A classical filter does not significantly improvethis situation. The tested output filter topology with dc link feed-back which works both in CM and DM, reduces the conducted

280 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 1, JANUARY 2006

Fig. 10. CM conducted current noise in the 150 kHz–30 MHz range, 100 m cable. (a) Motor and cable on GP. (b) Motor insulated, cable on GP.

EMI to acceptable levels even when long motor cables are used.When such a filter is used, at least in a Class A (industrial)environment, unshielded cables can be used in most situationswithout fearing EMC problems.

REFERENCES

[1] R. J. Kerkman, G. L. Skibinski, and D. W. Schlegel, “AC drives: year2000 (Y2K) and beyond,” in Proc. Applied Power Electronics Conf.Expo (APEC’99), vol. 1, 1999, pp. 28–39.

[2] F. Blaabjerg and P. Thøgersen. Adjustable speed drives—future chal-lenges and applications. presented at Int. Power Electronics Motion Con-trol Conf. (IPEMC). [CD-ROM]

[3] C. Klumpner, F. Blaabjerg, and P. Thøgersen, “Converter topologieswith low passive components usage for the next generation of integratedmotor drives,” in Proc. Power Electronics Specialists Conf. (PESC’03),vol. 2, 2003, pp. 568–573.

[4] S. Williamson and D. C. Jackson, “Integrated drives for industrialapplications,” in Proc. Power Conversion Intelligent Motion Conf.(PCIM’99), 1999, pp. 9–14.

[5] J. S. Kim and S. K. Sul, “New control scheme for AC-dc-AC converterwithout dc link electrolytic capacitor,” in Proc. Power Electronics Spe-cialists Conf. (PESC’93), 1993, pp. 300–306.

[6] J. Holtz and U. Boelkens, “Direct frequency convertor with sinusoidalline currents for speed-variable ac motors,” IEEE Trans. Ind. Electron.,vol. 36, no. 4, pp. 475–479, Nov. 1989.

[7] M. D. Manjrekar and T. A. Lipo, “A hybrid multilevel inverter topologyfor drive applications,” in Proc. Applied Power Electronics Conf. Expo(APEC’98), vol. 2, 1998, pp. 523–529.

[8] H. R. Andersen and J. Poulsen. Pulse-amplitude-modulated frequencyconverter for HVAC applications. presented at 10th Eur. Conf. PowerElectronics Applications (EPE’03). [CD-ROM]

[9] K. Taniguchi, N. Kimura, Y. Takeda, S. Morimoto, and M. Sanada, “Anovel PAM inverter system for induction motor drive,” Proc. PowerElectronics Drive Systems (PEDS’95), vol. 1, pp. 129–134, 1995.

[10] A. von Jouanne, P. Enjeti, and W. Gray, “Application issues for PWMadjustable speed ac motor drives,” IEEE Ind. Appl. Mag., vol. 2, no. 5,pp. 10–18, Sep./Oct. 1996.

[11] B. K. Bose, Ed., Power Electronics and Variable Frequency Drives.Technology and Applications. New York: Wiley, 1996.

[12] J. Holtz, “Pulsewidth modulation for electronic power conversion,”Proc. IEEE, vol. 82, no. 8, pp. 1194–1214, Aug. 1994.

[13] G. Skibinski, J. Pankau, R. Sladky, and J. Campbell, “Generation, con-trol and regulation of EMI from ac drives,” in Proc. IEEE Industry Ap-plications Conf. (IAS’97), vol. 2, 1997, pp. 1571–1583.

[14] A. L. Julian, G. Oriti, and T. A. Lipo, “Elimination of common-modevoltage in three-phase sinusoidal power converters,” IEEE Trans. Ind.Appl., vol. 14, no. 5, pp. 982–989, Sep. 1999.

[15] M. Cacciato, A. Consoli, G. Scarcella, and A. Testa, “Effects of PWMtechniques on common mode currents in induction motor drives,”in Proc. IEEE Conf. Industrial Electron. (ISIE’97), vol. 1, 1997, pp.S212–S217.

[16] L. Ran, S. Gokani, J. Clare, K. J. Bradley, and C. Christopoulos, “Con-ducted electromagnetic emissions in induction motor drive systems partI: time domain analysis and identification of dominant modes,” IEEETrans. Power Electron., vol. 13, no. 4, pp. 757–767, Jul. 1998.

[17] N. Mutoh, M. Kanesaki, J. Nakashima, and M. Ogata, “A new methodto control common-mode currents focusing on common-mode currentpaths produced in motor drive systems,” in Proc. IEEE Ind. Appl. Conf.(IAS’03), vol. 1, Sep./Oct. 1996, pp. 459–466.

[18] E. J. Bulington, S. Abney, and G. L. Skibinski, “Cable alternatives forPWM ac drive applications,” in Proc. IEEE IAS Petroleum ChemicalIndustry Conf., 1999, pp. 247–259.

[19] N. Hanigovszki, E. Petersen, J. Poulsen, and F. Blaabjerg, “Evaluation ofa method used for measuring high-frequency common-mode noise at theoutput of an adjustable speed drive,” in Proc. IEEE Int. Symp. EMC’03,vol. 2, 2003, pp. 548–552.

[20] Limits and Methods of Measurement of Radio Interference Characteris-tics of Information Technology Equipment, Eur. Std. EN55022, 1998.

[21] Specification for Radio Disturbance and Immunity Measuring Ap-paratus and Methods—Part 1: Radio Disturbance and ImmunityMeasuring Apparatus, IEC Std. CISPR 16-1, 1999.

[22] B. L. Harlacher and R. W. Stewart, “CISPR 22 conducted common modevoltage measurements along a wire using a capacitive voltage probe,” inProc. IEEE Int. Symp. EMC’02, vol. 1, 2002, pp. 1–5.

[23] Y. Hiroshima, S. Miyashita, and N. Kuwabara, “Method of measuringconducted disturbance using both capacitive voltage probe and currentprobe,” in Proc. IEEE Int. Symp. EMC, 1999, p. 798.

[24] Ammendment A1:2004: Specification for Radio Disturbance and Immu-nity Measuring Apparatus and Methods—Part 1–2: Radio Disturbanceand Immunity Measuring Apparatus—Ancillary Equipment—Con-ducted Disturbances, IEC Std. CISPR 16-1-2, 2003.

[25] Addition of a New Item Relating to Capacitive Voltage Probes, Am-mendment to CISPR 16-1, Clause 5.2.2, 2004.

HANIGOVSZKI et al.: EMC EVALUATION 281

Norbert Hanigovszki (S’02–M’06) was born inTimisoara, Romania, in 1977. He received the M.S.degree in electrical engineering from the UniversityPolitehnica, Timisoara, in 2000 and the Ph.D. degreefrom Aalborg University, Aalborg, Denmark, in2005.

During his study he spent three months as a Ph.D.Guest Student at the University of Padova, Padova,Italy, where his main focus was the research of outputfilters for ac drives with unshielded motor cables.Since 2002, he has been with Danfoss Drives A/S,

Graasten, Denmark, where he works as an EMC Engineer. His main interestsare power electronics, industrial drives, and electromagnetic compatibility.

Dr. Hanigovszki is a member of the IEEE Power Electronics, IEEE Electro-magnetic Compatibility, and IEEE Industry Applications societies.

Jørn Landkildehus (M’02) was born in Denmark in1969. He received the M.Sc.E.E. degree in electricalengineering from Aalborg University, Aalborg East,Denmark, in 1995.

Since 1995, he has been with the Research andProduct Development Department, Danfoss DrivesA/S, Graasten, Denmark, and since 2004 has beenManager of the EMC, Reliability and ApprovalsDepartment. His main research areas are powerelectronics, power converters and electrical drivesystems, including modeling, simulation, and design

with focus on EMC.Mr. Landkildehus is a member of the IEEE Electromagnetic Compatibility

Society.

Giorgio Spiazzi (M’93) was born in Legnago, Italy,in 1962. He received the M.S. (with honors) andPh.D. degrees in electronic engineering from theUniversity of Padova, Padova, Italy, in 1988 and1992, respectively.

In 1992, he spent six months at the Virginia PowerElectronics Center, Virginia Polytechnic Instituteand State University, Blacksburg. In 1993, he becamea Researcher in applied electronics. In 1997, he wasa Visiting Professor at the University of Campina,Brazil, doing research in the field of line-frequency

commutated power factor correctors. He is also co-author of the tutorialElectromagnetic Compatibility in Power Electronics, Aalborg University,Aalborg, Denmark, for the Danfoss Professor Programme in Power Electronicsand Drives. The same tutorial was also given in 2001 at different Indian IEEEChapters within the Distinguished Lecturer Program. He has published morethan 60 papers in international conferences and IEEE TRANSACTIONS. His mainactivities are in the field of power electronics and particularly on advancedcontrol techniques for dc/dc converters, single-phase and three-phase powerfactor correctors, soft-switching converters, and electromagnetic compatibilityissues in power electronics.

Frede Blaabjerg (S’86–M’88–SM’97–F’03) wasborn in Erslev, Denmark, on May 6, 1963. Hereceived the M.Sc.EE. and Ph.D. degrees fromAalborg University, Aalborg, Denmark, in 1987 and1995, respectively.

He was with ABB-Scandia, Randers, Denmark,from 1987 to 1988. He became an Assistant Pro-fessor at Aalborg University in 1992, AssociateProfessor in 1996, and a Full Professor in powerelectronics and drives in 1998. In 2000, he was aVisiting Professor with the University of Padova,

Padova, Italy, as well as part-time Programme Research Leader in wind turbinesat Research Center Risoe, Risoe, Italy. In 2002, he was a Visiting Professor atthe Curtin University of Technology, Perth, Australia. He is involved in morethan 15 research projects within the industry including the Danfoss ProfessorProgramme in Power Electronics and Drives. He is the author or co-author ofmore than 350 publications in his research fields including the book Controlin Power Electronics (New York: Academic, 2002). He is an Associate Editorfor the Journal of Power Electronics and for Elteknik. His research areas arein power electronics, static power converters, ac drives, switched reluctancedrives, modeling, characterization of power semiconductor devices and simu-lation, power quality, wind turbines, and green power inverters.

Dr. Blaabjerg received the 1995 Angelos Award for his contribution inmodulation technique and control of electric drives, the Annual Teacher Prizefrom Aalborg University in 1995, the Outstanding Young Power ElectronicsEngineer Award from the IEEE Power Electronics Society in 1998, five IEEEPrize Paper Awards during the last six years, the C.Y. O’Connor Fellowshipfrom Perth, Australia in 2002, the Statoil-Prize for his contributions in PowerElectronics in 2003, and the Grundfos Prize in acknowledgment of his in-ternational scientific research in power electronics in 2004. He has been anAssociate Editor of the IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS andof the IEEE TRANSACTIONS ON POWER ELECTRONICS. He is a member of theEuropean Power Electronics and Drives Association, the IEEE Industry Ap-plications Society Industrial Drives Committee, the Industry Power ConverterCommittee and the Power Electronics Devices and Components Committee,IEEE Industry Application Society. He has served as member of the DanishTechnical Research Council in Denmark from 1997 to 2003 and from 2001to 2003 he was Chairman. He has also been Chairman of the Danish SmallSatellite Programme and the Center Contract Committee. He became a memberof the Danish Academy of Technical Science in 2001 and in 2003 he becamea member of the Academic Council. From 2002 to 2003, he was a member ofthe Board of the Danish Research Councils. In 2004, he became Chairman ofthe Programme Committee on Energy and Environment.