Coupling Capacitor Voltage Transformers as

Harmonics Distortion Monitoring Devices in

Transmission Systems
Miloje Tanaskovic, nonmember, Abdul Nabi, nonmember, Slobodan Misur , nonmember ,
Paolo Diamanti, member IEEE, Ross McTaggart, member IEEE

Abstract—The proliferation of HVDC links in high voltage output voltage signals proportional to and in the phase with the
transmission systems and the growth of large industrial customers primary line-to-ground voltage, power line carrier
(with large non-linear loads) are increasing the need for harmonic communication and transient recovery voltage (TRV)
distortion voltage measurements at transmission voltages.
mitigation for circuit breakers during short line short circuit
This paper addresses the use of a Coupling Capacitor Voltage
Transformer with built-in harmonic monitoring device faults. The low voltage output signals are most often used for
(CCVTHM) to provide frequency response and accurate the following applications: metering/instrumentation,
measurement of harmonic voltage distortions in a high voltage protection schemes and to act as a low-voltage and low-power
transmission system in an economical and effective way.
supply. CCVT's are designed with an inherent margin of safety
for the various in-service operating conditions, thus providing
Keywords: Coupling Capacitor Voltage transformers (CCVT),
Coupling Capacitor Voltage transformers with harmonic the customer with a highly reliable and economical instrument
monitoring device (CCVTHM) , Harmonic distortion voltage for the supervision and control of their power transmission
measurement, frequency response, THD (Total harmonic lines.
distortion) The main components in the CCVT are:
• Capacitor Divider
I. INTRODUCTION • Step Down or Intermediate Transformer
Harmonic measurements are typically performed at the low • Series Compensation Reactor
voltage level at industrial customers using wire wound voltage • Ferroresonance damping circuit
transformers (VT's) and current transformers (CT's). The • Carrier accessories
increase of large industrial customers directly connected to the The capacitor divider is made up of many series connected
high voltage transmission system has increased the need to capacitor elements, connected line to ground. A tap is brought
accurately measure and monitor the harmonic distortion at out at an appropriate voltage level carefully coordinated with
transmission voltage levels. It is a very important to have an the intermediate transformer to provide the required output
accurate and reliable measurement of the harmonics distortion voltages. The capacitor elements on the high voltage side of
(THD), in order to obtain parameters for functional filter the tap are called C1 and the capacitor elements on the low
design and for resolving eventual disputes between supplier voltage side of the tap are called C2. To provide the reduced
and consumer of the power regarding THD. Though existing level tap voltage there are many more C1 capacitor elements
common design of CCVT provide many benefits, these than C2 capacitor elements. The capacitor elements are housed
CCVT's do not provide the linear, flat frequency response in hollow porcelain or composite insulators filled with an
required to accurately measure harmonic voltage distortions impregnating fluid.
across the most critical range of the harmonic frequency The series reactor and primary windings of the step-down
spectrums, thus requiring in the field calibration [1] that may transformer are manufactured with taps to enable voltage ratio
have an adverse effect on the accuracy of the measurement. and phase angle adjustments. The step-down intermediate
Trench Limited has developed a "special" CCVT design and / transformer, series compensating reactor and ferroresonance
or "field installation kit" to solve this problem. damping circuit are housed in the electromagnetic unit (EMU)
and immersed in mineral oil.
II. CCVT BASIC PRINCIPLE AND DESIGN Typical schematic diagram of common design Coupling
Coupling Capacitor Voltage Transformers (CCVT's) are Capacitor Voltage Transformer (CCVT) is shown in Fig.11
designed to be applied on high voltage transmission systems to while Coupling Capacitor Voltage Transformer with Harmonic
provide lower more manageable (approximately 57-115V) Monitoring terminals (CCVTHM) is shown in Fig.12.

The authors are with Trench Ltd. Canada

Presented at the International Conference on Power Systems

Transients (IPST’05) in Montreal, Canada on June 19-23, 2005
Paper No. IPST05 - 031
 III. CCVT TEST AND MEASUREMENTS
Many papers, field and laboratory tests, digital simulations disconnected from the capacitor divider of the CCVT.
and ATP- EMTP models have been developed and performed Since total harmonics distortion (THD) is the most relevant
on common design CCVT and other voltage transducers in information to be obtained from the power systems for further
order to evaluate, errors [1], transient response, [2] and analysis, the accuracy of the monitoring device (VT/PT, CT,
frequency response characteristics [3]. To the author’s CCVT etc.) in the frequency spectrum of the interest is of the
knowledge, most of the existing research, measurements and utmost importance. For this reason the test and measurements
simulations are focused on common design CCVT. had been performed at relatively low voltage level (up to
Using a modified Trench 145kV, 60Hz nominal voltage 1kVrms at 60 Hz), while accuracy of the transfer of THD from
TEMP138 CCVT model built with tapped (115Vrms) primary side to the secondary side (harmonic monitoring
th
secondary voltage terminals and special harmonic monitoring terminals) had been measured for harmonics up to 50 (3000
terminals (200Vrms), a frequency scan test was performed Hz).
using the test circuit shown in Fig. 1. The objective was to take
advantage of the capacitor divider as an integral part of CCVT
as an "ideal" monitoring device for harmonics [4].

Fig. 1 Test circuit for frequency scan of CCVT with IV. TEST RESULTS
built-in harmonic monitoring terminals The results of the laboratory frequency scan tests are shown
Measurements were performed by using multi-input in Tables I-III and corresponding Figures 2-4. It should be
oscilloscope (PHILIPS-PM3384) to obtain and store data and noticed how in agreement are THD values of CCVT input
waveforms for post-processing using ATP Analyzer. It is a and what was measured across the harmonic monitoring
very important to use proper cables (i.e. double-shielded) for terminals. At the same time THD values measured across
cabling in order to eliminate any back ground noise. The secondary of the CCVT yields the typical frequency response
premise of the test was to inject power frequency signal with for a common CCVT design. Note the large error around 500-
harmonic content (measured as % of THD) (Input 1- Fig. 1.) 800Hz.
and simultaneously measure % of THD across harmonic Figures 5 to 10 show waveforms of different harmonics
monitoring terminals (Input 2- Fig. 1.) and secondary terminals captured during frequency scan. Comparing waveforms
(Input 3- Fig. 1.). The same test was performed with CCVT (CCVT input vs. Harmonic terminals) there was no apparent
being loaded with 200VA and without secondary load as well phase shift noticed, while (CCVT input vs. sec. terminals)
as with tap point (C1/C2) of electromagnetic unit (EMU) being there is a phase shift.
Table I
Percentage of Total Harmonics Distortion (THD) at no-load
No secondary load 18
CCVT Harm onics [%] CCVT input
Fn CCVT input 16
sec.term . m on. term . Harmonic terminals
(Hz) Osc. Input1 Osc. Input3 Osc. Input2
THD(%) THD(%) THD(%) 14
180 17.17 17.29 17.20
Secondary
300 9.96 10.32 9.65 12 terminals
420 6.63 8.47 6.73
560 10.63 3.50 10.99 10
660 7.52 5.05 7.74
780 7.30 5.33 6.75 8
900 4.82 3.88 4.82
1020 4.37 3.16 4.40 6
1140 3.57 2.75 3.10
1260 3.91 3.31 4.03 4
1500 14.25 11.72 13.60
2
1620 11.30 9.30 11.17 0 500 1000 1500 2000 2500 3000
1740 7.66 6.60 7.96
(file Noload.ADF; x-var Freqeucy)%_InHi %_H1H2 %_X1X3
2000 5.06 4.25 4.90
3000 6.80 5.60 6.70 Fig.2. THD % at no secondary load

Table II
Secondary load - 200VA Percentage of Total Harmonics Distortion (THD) at Full-load
40
CCVT Harmonics [%]
Fn CCVT input
sec.term . mon. term. 35 CCVT input
(Hz) Osc. Input1 Osc. Input3 Osc. Input2
THD(%) THD(%) THD(%) 30
180 36.76 34.02 35.70
Secondary terminals
300 20.50 18.13 19.93 25
420 13.35 11.10 12.86
560 9.15 3.20 8.83 20
660 6.84 5.50 6.51 Harmonic terminals
780 7.86 6.15 7.50 15
900 5.87 4.60 5.50
1020 3.99 2.70 3.87 10
1140 6.50 4.97 6.03
1260 5.29 4.39 5.03 5
1500 3.86 3.15 3.66
1620 3.80 3.99 3.46 0
1740 3.68 3.86 3.35 0 500 1000 1500 2000 2500 3000
2000 3.56 3.74 3.24 (file Fullload.ADF; x-var Frqucy) %_InHi %_H1H2 %_X1X3
3000 2.06 1.50 1.90
Fig.3. THD % at secondary load-200 VA
Table III
E le c .m a g . u n it ( E M U ) d is c o n n e c t e d Percentage of Total Harmonics Distortion (THD) E.M. Disconnected
70
H a r m o n ic s
Fn C C V T in p u t [%]
m o n . te r m .
(H z) O s c . In p u t1 O s c . In p u t2 60
CCVT input
T H D (% ) T H D (% )
180 6 8 .9 6 6 9 .2 8 50
300 3 8 .9 9 3 9 .4 3
420 2 3 .0 1 2 2 .8 1 40 Harmonic terminals
560 1 4 .7 4 1 4 .6 3
660 1 0 .7 5 1 0 .3 8 30
780 6 .8 9 6 .8 7
900 4 .9 3 4 .9 1 20
1020 9 .0 8 8 .8 1
1140 6 .7 5 6 .5 7 10
1260 5 .4 0 5 .0 3
1500 3 .2 0 3 .1 0 0
1620 3 .1 8 3 .0 8 0 500 1000 1500 2000 2500 3000
1740 3 .1 7 3 .0 7 (file EMdisconnected.ADF; x-var Freqeucy)%_InHi %_H1H2
2000 3 .1 5 3 .1 0
3000 2 .2 8 2 .3 5
Fig.4. THD % Tap point (C1/C2) of (EMU) disconnected
 Fundamental frequency with 3rd harmonics (No-load) Fundamental frequency with 3rd harmonics (Full-load)
3 2.0
[V] [V]
CCVT input 1.5
2
1.0
1 Harmonic terminals
0.5

0 0.0
Sec. terminals
-0.5
-1
-1.0
-2
-1.5

-3 -2.0
0 5 10 15 20 25 30 35 [ms] 40 0 5 10 15 20 25 30 35 [ms] 40
(file 3rd.ADF; x-var tn) v_TpDv v_HarT v_Sec (file 3rd.ADF; x-var tn) v_TpDv v_HarT v_Sec
factors: 1 4 1.6 1 factors: 1 4 1 1
offsets: 0 0 0 0 offsets: 0 0 0 0

th th
Fig.5. 3 (180Hz) Harmonic waveform-no load Fig.6. 3 (180Hz) Harm. waveform- sec.load-200 VA

Fundamental frequency with 50th harmonics (No-load) Fundamental frequency with 50th harmonics (Full-load)
3 1.5
[V] [V]
CCVT input
1.0
2

Harmonic terminals0.5
1
0.0
Sec. terminals
0
-0.5

-1
-1.0

-2 -1.5
0 5 10 15 20 25 30 [ms] 35 0 5 10 15 20 25 30 [ms] 35
(file 3kHz.ADF; x-var tn) v_TpDv v_HarT v_Sec (file 3kHz.ADF; x-var tn) v_TpDv v_HarT v_Sec
factors: 1 5 1.6 1 factors: 1 4 1 1
offsets: 0 0 0 0 offsets: 0 0 0 0

th th
Fig.7. 50 (3000Hz) Harmonic waveform-no load Fig.8. 50 (3000Hz)Harm. waveform- sec. load-200 VA
Fundamental frequency with 3rd harmonics (E.M. Disconnected) Fundamental frequency with 15th harmonics (E.M. Disconnected)
4 3
[V] [V]
3 CCVT input
2
2
Harmonic terminals
1
1

0 0

-1
-1
-2
-2
-3

-4 -3
5 10 15 20 25 30 35 40 [ms] 45 0 5 10 15 20 25 30 35 [ms] 40
(file 3rd.ADF; x-var tn) v_TpDv v_HarT (file 15th.ADF; x-var tn) v_TpDv v_HarT
factors: 1 4 1 factors: 1 4 1
offsets: 0 0 0 offsets: 0 0 0

th th
Fig.9. 3 (180Hz) Harm. waveform- Tap point (C1/C2) of Fig.10. 15 (900Hz) Harm. waveform- Tap point (C1/C2)
(EMU) disconnected of (EMU) disconnected
 Fig.11 Common design CCVT

Fig.12 Coupling Capacitor Voltage Transformer with Harmonic Monitoring terminals

 V. CONCLUSIONS VII. BIOGRAPHIES
As mentioned in [1], “The possibilities of using a high 1. Miloje Tanaskovic received Bachelor of Science
voltage CVT to measure harmonics depend on the design of degree in Electrical Power Engineering from University of
the CVT”. In this specific case, it is shown that using slightly Sarajevo , Bosnia and Herzegovina in 1982.
modified design of CCVT with special harmonic voltage Currently, he is the Engineering Manager at the Instrument
monitoring terminals brought out from capacitor divider of the Transformer Division of Trench Ltd. Canada.
CCVT provides the frequency response that common design 2. Abdul Nabi received Bachelor of Science degree from
CCVT in most cases cannot provide. Based on the Salahaddin University of Irbil , Iraq in 1979 , M.Sc and Ph.D
th
measurement of THD in wide frequency spectrum (up to 50 degrees from St. Petersburg State Polytechnic University , St.
harmonic), it is shown a very high accuracy of the Petersburg , Russia in 1986 and 1990 , respectively.
measurement. For example the conventional CCVT shows Currently, he is the Senior Engineer at the Instrument
errors as high as ~30% whereas the harmonic measurement Transformer Division of Trench Ltd. Canada.
output remains within 5-10% at worse, 1-3% typical. At the 3. Slobodan Misur received Bachelor of Science degree
same time there is no phase shift that may be important if there in Electrical Power Engineering from University of Sarajevo ,
is need for power measurement. Bosnia and Herzegovina in 1988.
Special considerations must be taken in the design Currently, he is the Design Engineer at the Instrument
regarding the instrument being used to monitor and measure Transformer Division of Trench Ltd. Canada.
harmonics, (i.e. high input impedance of the instrument, 4. Paolo Diamanti received diploma in Electrical Power
>200kOhm). Engineering Technologist from Ryerson Polythecnical Institute
CCVT’s (i.e.Trench product) installed in-field could be in 1976.
relatively easily retrofitted in order to be used for accurate Currently, he is the Chief Engineer at the Instrument
harmonic monitoring and measurement. Transformer Division of Trench Ltd. Canada.
The implementation of the harmonic monitoring terminals 5. Ross McTaggart, P. Eng received Bachelor of Applied
provides additional benefits to the user while maintaining all Science degree in Electrical Power Engineering from
existing features and benefits of the CCVT as a voltage University of Toronto, Canada in 1976.
measurement device such as: Currently, he is the Product Development Manager at the
• Accurate voltage transformation Instrument Transformer Division of Trench Ltd. Canada.
• Power Line Carrier Coupling
• Breaker TRV mitigation
• Accurate Harmonic Voltage measurement

VI. REFERENCES

[1] Voltage transformer frequency response. Measuring

harmonics in Norweigian 300 kV and 132 kV power systems.
CIGRE 36.05/CIRED WG 2/UIEPQ Joint Working Group
CCU2 on Voltage Quality.

[2] A Coupling Capacitor Voltage Transformer

Representation for Electromagnetic Transient Studies. D.
Fernandes Jr., W.L.A. Neves, and J.C.A. Vasconcelos. IPST
2003 technical paper.

[3] Power Quality Impacts of Series and Shunt

compensated Lines on Digital Protective Relays. Mojtaba
Khederzadeh. IPST 2003 technical paper.

[4] Problems of Voltage Transducer in Harmonic

Measurement. Yao Xiao, Jun Fu , Bin Hu, Xiaoping Li and
Chunnian Deng , IPST 2003 technical paper.