This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

Comprehensive Analysis to Specify a Static Var

Compensator for an Electric Arc Furnace Upgrade
Sam Morello, PE Thomas J. Dionise, PE Thomas L. Mank, PE
ARM EnerTech Associates, LLC Eaton Electrical Group Standard Steel, LLC
Johnstown, PA Power Systems Engineering Burnham, PA
Senior Member, IEEE Warrendale, PA
Senior Member, IEEE

Abstract—This steel maker is upgrading the Electric

Arc Furnace (EAF) Melt Shop to increase production and
improve metallurgical quality of their product. The increased
electrical load and harmonic generation for the new furnace
operation necessitated the installation of a static var compensator
(SVC) to meet utility interconnect requirements. This paper
describes a sequence of comprehensive analysis performed to
specify the SVC including Load Flow Analysis, Flicker Analysis
and Harmonic Analysis Studies. The analysis presented in this
paper sized and evaluated the proposed SVC reactive
compensation to mitigate flicker, harmonics and power factor for
the new EAF and LMF installations to comply with the utility
interconnect requirements.

Keywords-static var compensator (SVC), Electric Arc Furnace

(EAF), Ladle Melt Furnace (LMF), harmonic analysis, harmonic
filtering, IEEE Std 519-1992, IEEE Std 18-1992, flicker

I. INTRODUCTION
Sumitomo Metals / Standard Steel are upgrading the
Electric Furnace Melt Shop at their Burnham, PA plant to
increase production and improve metallurgical quality of their Figure 0. Simplified One-line Diagram for the Standard Steel Existing
product. The proposed solution is the revamp of the existing EAF#4 and Existing EAF#3
EAF #4 shown in Fig. 0 and the supply of a new swing type
LF/VD Station. Along with furnace process improvements, the The specific objectives of the studies were to follow IEEE
EAF #4 furnace transformer will be upgraded from 30 MVA to standards and guidelines for harmonic mitigation [2], filter
a 50/56 MVA. The new swing type design LF/VD station will bank duty evaluation [3] and to perform the flicker analysis [4].
allow two ladles of steel to be processed at the same time and In addition, the specification of basic ratings for the size of a
also allow for arc reheating after vacuum degassing if required. SVC and the application of harmonic filters. All results will
The new LF/VD station would be located in the existing EAF need to comply with the utility interconnect requirements at the
#3 position shown in Fig. 0. The existing 15 MVA furnace defined 230 kV point of common coupling (PCC) for
transformer will be reused along with the step down harmonics [2], flicker [4], unity power factor and voltage
transformer. EAF#2 will be removed. The LF/VD would be regulation.
installed after the revamp of EAF #4 and will allow the plant to
operate with one (1) EAF compared to three (3) currently. To perform all studies effectively, accurate harmonic
profiles along with loading will be developed. The data will be
To validate and design for the increased electrical load and generated from current electric furnace operation along with
harmonic generation for the new furnace operation, a sequence new furnace data. In addition to sizing and evaluating the SVC
of analysis was required. Load Flow Analysis, Flicker filter bank, the study will evaluate the physical location of the
Analysis and Harmonic Analysis Studies for the new EAF #4 SVC. The SVC will be installed a distance away from the
and LMF installations at the Standard Steel facility in incoming utility service since the majority of the plant is built
Burnham, PA were performed. The intent of the studies was to on a flood plain with limited open space.
size and evaluate the proposed SVC reactive compensation in
accordance with [1] to mitigate flicker, harmonics and power II. SCOPE OF STUDY
factor for the new EAF and LMF installations to comply with A. Define PCC at Utility 230kV at Yeagertown Sub-Station
the utility interconnect requirements.
Point of Common Coupling (PCC), as defined in [2], is a
point of metering, or any point as long as both the utility and

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

the consumer can either access the point for direct D. Utility Requirements
measurement of the harmonic indices meaningful to both or The utility electricity provider for Standard Steel have
can estimate the harmonic indices at point of interference specific requirements that are required to be followed with an
(POI). Within an industrial plant, the PCC is the point between upgrade of electric furnace usage of this magnitude. Electric
the nonlinear load and the other loads. The PCC with the Arc Furnaces are major contributors to voltage unbalance,
consumer/utility interface is the closest point on the utility side harmonics and flicker.
of the customer's service where another utility customer is or
could be supplied. Incoming Voltage Regulation
Voltage regulation of the utility 230 kV and 345 kV sub-
The PCC for the purpose of this study is the 230 kV transmission system is required to be maintained between
primary of the two (2) 46 kV step down transformers. These 92.0%-105% of nominal voltage under normal and single
transformers detailed in Table I are normally operated in transmission element outage conditions.
parallel and supply the Standard Steel plant thru two dedicated
lines (1YSS and 2YSS) to Standard Steel Sub 4. Power Factor Correction
The utility requirement for close to unity power factor is
TABLE I. UTILITY STEP DOWN TRANSFORMER that the power factor at the point of interconnection shall be
Transformer power rating : 45/60/70 MVA OA/FA/FOA controlled to be within the range of 0.97 lagging to 0.99
Primary voltage : 230,000 Volts leading. Customers connected to the sub-transmission system
Secondary voltage : 46,000 Volts design their systems to operate at close to unity power factor.
Secondary current : 941 Amps
Short Circuit Impedance : 6.18% Harmonics and Flicker
Sub 4 is a dedicated plant Sub-Station for the arc furnace The utility is aware certain electrical equipment served will
circuits. Other plant circuits are supplied by another Standard generate flicker and harmonics that can negatively impact the
Steel Sub-station and an utility feed that is not associated with utility power system. The utility requires the connected facility
the Yeagertown 230kV. Therefore, for the purpose of this will comply with harmonic voltage and current limits specified
study, only the Yeagertown supplied circuit to the Standard in [2]. Flicker will be defined as described in [4]. Pst is a
Steel plant will be evaluated. This definition was discussed measure of short-term perception of flicker obtained for a ten-
with the utility representative and approved as the PCC by the minute interval. Plt is a measure of long-term perception of
utility company. flicker obtained for a two-hour period calculated from 12
consecutive Pst values. The connected facility shall be
B. Upgrade Arc Furnace Transformer designed and operated such that Pst does not exceed 0.8 and Plt
A new Indoor Electric Arc Furnace Transformer of Table II does not exceed 0.6 for more than 1% of the time (99%
rated 50/56 MVA, continuous duty, 3 phase, 60 Hertz, FOW, probability level) using a minimum assessment period of one
55/65°C rise will replace the existing 30 MVA transformer. week. These flicker and harmonics limits must be met during
normal (N-0) system configurations including a generating unit
TABLE II. EF#4 TRANSFORMER NAMEPLATE outage.
Transformer power rating : 50 MVA E. SVC Requirements
(12% Overload 56 MVA)
Primary voltage : 46,000 Volts The large and varying loads of an Electric Arc Furnace will
Secondary voltage : (605/510/ 415) cause disturbing effects to the electric supply system. The
Secondary current : 56,600 Amps disturbances are mainly caused by fluctuations in the reactive
Tap changer : On-load, motor
operated
power and/or unsymmetrical load of the Electric Arc Furnace.
Short Circuit Impedance : 5% The following influences in the electric supply system can be
noted such as large and quick changes of active and reactive
C. Reuse Existing 15MVA Transformer Pair For New Ladle power, harmonic currents, unbalance currents, poor power
Furnace factor, voltage fluctuations and flicker.
The existing 15 MVA step-down transformer of Table III The requirements put forth by the utility necessitate the
will be re-used along with the existing ladle furnace application of a Static Var Compensator (SVC). Given the size
transformer of Table III for the new Ladle Furnace operation. and nature of the reactive power characteristics, the application
of individual filters banks alone is not practical. Instead,
TABLE III. STEP-DOWN TRANSFORMER NAMEPLATE
dynamic control of the reactive power is needed using an SVC.
Transformer power rating : 15/20/22.4 MVA The typical SVC system consists of a Thyristor Controlled
Primary voltage : 46,000 Volts Reactor (TCR) and filter circuits as illustrated in Fig. 1.
Secondary voltage : 6,900 Volts
Secondary current : 1,875 Amps
Tap changer : Off Load
Short Circuit Impedance : 6.73%

TABLE IV. LADLE FURNACE TRANSFORMER

Transformer power rating : 15/16.8 MVA
Primary voltage : 6,900 Volts
Secondary voltage : 334 Volts (Top Tap)
Secondary current : 29,050 Amps
Tap changer : Off Load
Short Circuit Impedance : 5.68%

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

c
b

e
Figure 1. Typical SVC One Line Diagram

A Filter Circuit is a tuned circuit, which consists of

capacitor banks (capacitance) and tuned reactor coils Figure 2. Aerial View of Major Plant Power Distribution for Arc Furnace
(inductance) connected in series. Filter circuits will reduce the Loads: a) Standard Steel Sub 19 46kV, b) Standard Steel Sub 4 to Sub 19
negative influence of the loads like harmonic distortion and 46kV Line, c) Standard Steel Sub 4 46kV, and d) Utility 1YSS and 2YSS
poor power factor. Results of this study will specify the tuning 46kV Lines, and e) Utility Yeagertown Sub 230kV/46kV
frequencies of the filter circuits with respect to the generated
harmonic frequencies and amplitudes to reduce the harmonic station throughout the plant for Arc Furnace service. Other
levels at the utility source. plant loads are supplied by separate utility and Standard Steel
Sub-stations. Fig. 2 shows an aerial view of major plant power
At the fundamental frequency (60Hz), the filter circuits distribution for Arc Furnace loads
behave like capacitors and provide power factor improvement.
The TCR consists of thyristor devices in series with reactors A typical SVC physical size, appropriate for the application
directly connected to the medium voltage system / SVC busbar. at Standard Steel, is illustrated in Fig. 3. This is the minimum
The TCR will reduce the negative influence of the loads like clear space required for installation.
flicker, voltage fluctuation, and voltage unbalance.
III. MODEL DEVELOPMENT
A model of the Standard Steel power system was developed
to produce a design for the new furnace additions. Elements of
the model included electrical load development thru design
data and measurements, plant location of the SVC and
harmonic models.
A. Data Collection Methodology
Power Quality Meter
A high-end Power Quality Meter was selected to monitor
each of the existing furnaces. Features of the instrument are
the ability to discern THD/harmonic spectrum and
Figure 3. Typical SVC Arrangement and Physical Size
TID/interharmonics spectrum to the 63rd harmonic and to
measures voltage flicker. In addition, the meter was configured
Two plant sites were considered for the location of the
to record all power parameters (volts, amps, MW, MVAR,
SVC. Sub 4 was first reviewed, based on it being the main
MVA, PF).
distribution sub-station for the plant arc furnace supply. This
Metering Locations location was rejected due to lack of sufficient open space in the
Two (2) High Power Quality Meters were utilized to collect vicinity of Sub 4 as well as its location in a designated flood
data. Sub 4 (Plant Incoming Sub-Station) did not have plain as shown in Fig. 4.
adequate protection and control signals for the monitoring. It
was decided to place the meters on the primary of each
operating Arc Furnace where PTs and CTs were readily
available. Meters were connected at furnaces EAF#4, EAF#2
and EAF#3. High speed and long term data was collected
during various times of the furnace cycle. Special
consideration was given to melt downs and refining operations.
These conditions represent the extreme electrical operations of
the operation.
B. SVC Location
Standard Steel is supplied by two 46 kV lines (1YSS and
2YSS) from the utility Yeagertown Substation. The lines enter

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

b
a
c c
b

Figure 4. Aerial and Ground View of Sub-Station 4: a) Standard Steel Figure 6. Aerial and Ground View of Sub-Station 19: a) Standard Steel Sub
Sub 4 46kV, b) View of Sub 4 from Left, and c) View of Sub 4 from Right 19, b) Sub 19 Proposed Site for SVC, c) Sub 4 to Sub 19 46kV Line, and d)
Standard Steel Sub 4 46kV
The second site considered was Sub 19, which supplies 46
kV directly to EAF4 as shown in Figs. 5 and 6. This sub- The power system was modeled based on the following
station is directly connected to Sub 4 via a dedicated line. Sub ones lines provided by Standard Steel. The plant is served by
19 is above the flood plain and an available track of open space the utility’s Yeagertown 230/46 kV Substation as shown in Fig.
is present. It was decided, for purposes of this study, to locate 7. Two (2) 45 MVA, 230/46 kV transformers are operated in
the sub-station in this area. In addition, the plant is planning to parallel with primary and secondary tie breakers normally
resize the conductors to Sub 19 and feeders to EAF 4. This closed. One 46 kV overhead line called 1YSS runs
study incorporates these future changes into the model. approximately 2100 feet to the Standard Steel 46 kV Substation
4-1 which serves the new EAF4 circuit. A second 46 kV
overhead line called 2YSS runs approximately 1700 feet to the
Standard Steel 46 kV Substation 4-3 which is tied thru a switch
c a to Substation 4-2 which serves the new LMF circuit.
EAF4 Circuit
The EAF4 circuit fed from 46kV Sub 4-1, consists of a SF6
b circuit breaker, 3300 feet of 46 kV overhead line, 785 feet of
two (2) 500 MCM cables per phase, EAF transformer rated 50
MVA, 46/.605 kV, 5%Z as well as the secondary EAF cables
and the EAF. The EAF will be upgraded in total with new
equipment.
LMF Circuit
The LMF circuit fed from 46kV Sub 4-2, consists of a
switch, 15MVA, 46/6.9 kV, 6.41%Z transformer, 825 feet
three (3) 1000 MCM cables per phase, LMF transformer rated
15 MVA, 6.9/.334 kV, 5.68%Z as well as the secondary LMF
cables and the LMF. The LMF will be supplied as new
Figure 5. Aerial View of Plant Showing Location of Proposed SVC at equipment.
Sub19: a) Standard Steel Sub 19, b) View of Sub 19 from Right and c) Sub 19
Proposed Site for SVC D. Utility (utility) Short Circuit
C. Model One Line The utility short circuit information for this report was
provided at Yeagertown Sub-station on the 230 kV level of
The study is processed on a digital computer using power 3543.1 MVA with X/R=8.417.
simulation software, both load flow and harmonic analysis
software, which is designed to solve the multiple simultaneous E. Electrode Regulator
non-linear equations that describe systems with real and The Electrode Regulator will replace the existing
reactive power loading. mechanical regulation with hydraulic regulation using the
Digital Regulator with Harmonics control mitigation.
Operating principle of the digital regulator is based on the
"constant impedance" and enhanced with an electrode speed
regulation loop. The effects of digital regulation on furnace
performance are monitored by the furnace operator through a
real-time display of key parameters as shown in Fig. 8.

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

a
b

Figure 7. Simplified One-line Diagram for the Standard Steel Future EAF4
and Future LMF

Figure 8. Real-Time Display of Key Parameters Indicating Furnace

Performance Controlled by the Digital Electrode Regulator

Feedback electrode voltage and current signals coming

from the secondary CT's and PT's are sent to a fast sampling Figure 9. Power Curve for EAF4: a) 50MVA Loading with SVC, and b)
analog input card and processed by a dedicated microcomputer. 50MVA Loading without SVC
In addition to voltage and current, power factor is also
computed. All the calculations between feedback AC current Heat Profile
and voltage and output signals to the servo-valves are via high Heat profile for EAF#4 for reference as shown in Fig. 10.
speed digital communication. The high speed nature of the This illustrates the electrical loading in MW from the
regulator and electrode positioning system will lend itself to beginning to end of a heat. This operation is based on three
less voltage disruption and harmonic current generation. buckets of scrap being charged into the furnace for melting.
The beginning of each bucket has an electrical characteristic of
F. 50 MVA Arc Furnace Load Definition
high voltage fluctuation and harmonic generation. As the scrap
Fig. 9 presents Load Flow Case 1 and 2 plotted on power melts to the liquid state the electrical disturbances are greatly
curves as described in [5], [6]. The points show modeled reduced.
furnace operating points without and with SVC compensation.
Note that point with SVC compensation is very close to the Furnace Secondary Impedance
power curve line due to the voltage support of the SVC. Table V contains values of furnace secondary impedance,
which will be utilized to calculate short circuit current for the
flicker calculation. This data was provided by the furnace
manufacturer.

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

b c
a

Bucket
End melt 1 Bucket

nd
st

End melt 2

Meltdown
Meltdown

Meltdown

Refine
Figure 10. Heat Profile for EAF4: a) 1st Bucket Meltdown, b) 2nd Bucket Meltdown, and c) 3rd Bucket Meltdown & Refine

TABLE V. FURNACE SECONDARY IMPEDANCE TABLE

a
RED = exceeds limit, GREEN = acceptable and YELLOW = excessive
b
G. 15 MVA Ladle Furnace Load Definition
Power Curve
Fig. 11 presents Load Flow Cases 1 and 2 (see Section 4.2)
plotted on power curves. The points show modeled furnace
operating points without and with SVC compensation. Note
that point with SVC compensation is very close to the power
curve line due to the voltage support of the SVC.
IV. LOAD FLOW ANALYSIS
A. Case 1 – Base Load (EAF4 and LMF at rated) without
SVC
Refer to Table VI for the following evaluation of Power
Factor and Voltage at the utility’s Yeagertown Sub-Station per
utility requirements.

Figure 11. Power Curve for Ladle Furnace: a) 15MVA Loading with SVC,
and b) 15MVA loading without SVC

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

TABLE VI. SUMMARY OF LOAD FLOW CASE 1 D. Power Factor Correction at PCC
The results in Table IX, taken from Case 2, shows the SVC
contributing 41.5 MVARs (Yellow) to the power system will
satisfy the utility power factor requirement. This would also
indicate the minimum size of the SVC, without concern for
flicker reduction.

TABLE IX. SVC CONTRIBUTIONS IN LOAD FLOW CASE 2

RED = exceeds limit, GREEN = acceptable and YELLOW = excessive MW Mvar MVA PF
Yeagertown Sub 56.4 0.7 56.4 1.00
Note: Base Load (EAF and LMF at rated operating point) SVC -0.7 41.5 41.5 -0.02
without SVC – both FCE loads at .8 PF- EAF4@50MVA Note: Illustrating Required MVARs for Compliance with
LMF@15MVA utility Specification.
Other Observations
Lines 1YSS and 2YSS are not equally loaded. Voltage E. Conclusions
drop at EAF4 and LMF degraded, especially LMF. This will In all cases studied (with or without the SVC filter bank),
result in loss of power to the furnace. the voltage regulation was within the utility requirements.
Although, operation of the EAF4 and LMF without a SVC will
B. Case 2 – Base Load (EAF4 and LMF at rated) with SVC violate the utility requirement for power factor without Var
Refer to Table VII for the following evaluation of Power compensation at the 230 kV PCC. Power factor correction will
Factor and Voltage at the utility’s Yeagertown Sub-Station per be required. A SVC will be a solution that can provide variable
utility requirements. MVAR support to the power system to maintain the utility
requirement. The minimum size will be 41.5 MVAR plus a
TABLE VII. SUMMARY OF LOAD FLOW CASE 1 20% factor for a total of 50 MVARs. Application of a SVC will
MW Mvar MVA PF V% support utility power factor requirement with the SVC located
Yeagertown Sub 56.4 0.7 56.4 1.00 99.8 at Sub 19. The case studies indicate that line 1YSS and 2YSS
1YSS 43.4 -12.1 45.1 0.96 will be unevenly loaded with the current Standard Steel
2YSS 12.9 10.3 16.5 0.78
SVC -0.7 41.5 41.5 -0.02
practice of SW-12 being open.
EF4 42.6 26.4 50.1 0.85 99.8
LMF 12.8 8.9 15.6 0.82 94.4
V. FLICKER ESTIMATION
Fluctuations in the currents drawn by electric arc furnaces
RED = exceeds limit, GREEN = acceptable and YELLOW = excessive may be large enough to cause voltage fluctuations in the
electric distribution system causing lamp flicker. Such voltage
Note: Base Load with SVC Compensation at Sub 19 – both fluctuations may also effect sensitive electronic equipment
FCE loads at .85 PF- EAF4@50MVA LMF@15MVA such as televisions and computers. Such voltage fluctuations
must be kept below the “threshold of irritation” to minimize the
Other Observations effects.
Lines 1YSS and 2YSS are not equally loaded. Voltage
Electric utilities use different methods to quantify the
drop at EAF4 and LMF are being supported by SVC. This
“threshold of irritation” due to flicker [6], [7], [8], [9] and [10].
results in increased power to the furnaces versus Case 1.
The UIE developed formula in [7] for efficient estimation of
C. Case 3 – Base Load with SVC & Sub 4 disc SW-12 Closed flicker makes use of the characteristic emission coefficient
Refer to Table VIII for the following evaluation of Power (KST). The KST method will be used here to predict the flicker
Factor and Voltage at the utility’s Yeagertown Sub-Station per for the proposed EAF and LMF at Standard Steel.
utility requirements. A. KST Method
TABLE VIII. SUMMARY OF LOAD FLOW CASE 3
A simple prediction formula for arc furnace flicker at the
point of common coupling (PCC) recommended by UIE is
MW Mvar MVA PF V%
Yeagertown Sub 56.4 0.7 56.4 1.00 99.8 Pst_99% = Kst · (Sccf/Sccn) (1)
1YSS 25.7 -0.8 25.7 1.00
2YSS 30.6 -1.0 30.6 1.00 where Sccf is the short-circuit level of the arc furnace at the
SVC -0.7 41.5 41.5 -0.02 PCC. Sccn is the fault level of the network at the PCC, and Kst
EF4 42.5 26.3 50.0 0.85 99.6 is the characteristic emission coefficient for Pst, ranging from
LMF 12.8 8.9 15.6 0.82 94.4
48 to 85 with a mean value of 60 [9] and [10].
RED = exceeds limit, GREEN = acceptable and YELLOW = excessive Evaluation of Utility Requirements for Flicker
] Refer to Table X for a prediction of the flicker at utility’s
Note: Load with SVC Compensation at Sub 19–Tie Bus at Sub Yeagertown 230kV Sub-Station per utility requirements for
4 - both FCE loads at .85 PF- EAF4@50MVA LMF@15MVA connecting party facility on the sub-transmission system. The
table illustrates calculated values of flicker given the industry
Other Observations range of Kst of 48 (Green) to 85 (Red), with 65 (Yellow) as a
Lines 1YSS and 2YSS are nearly equally loaded. Voltage mid-point. The application of a traditional SVC will yield a
drop at EAF4 and LMF are being supported by the SVC. This reduction factor of 2 for the Pst value derived without the SVC
is resulting in increased power to the furnaces versus Case 1. [9]. Short circuit currents were developed from the power
model and calculation short circuit at the tip of the electrode.

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

For the purpose of this study, the EAF impedance was utilized The present operation with an analog regulator and hoist
for the LMF. drives yielded a Pst value of about 1.2 as shown in Figs. 13 and
14 at the Yeagertown 230 kV PCC without SVC compensation.
Flicker as defined in [4], the Pst shall not exceed .8: This is a conservative approximation given the data in Figs. 13
 Without SVC – Pst is greater than .8 for entire range of and 14, based off of the average peaks. With a SVC reduction
Kst (Indicated in Red Text in Table X) factor of 2, this would yield a Pst value of .6, below the
required limit of .8.
 With SVC – For a Pst < .8 the furnace Kst range of 48 to
75 will be within the acceptable range. (Indicated in C. Calculated Kst
Green Text in Table X) Using measured values obtained at the point of common
coupling (PCC), the Kst of can be calculated. Rearranging (1)
TABLE X. FLICKER PREDICTION TABLE as defined in Section 5.1gives
Kst = Pst_99% (Sccn/Sccf) (2)
and using the measured values from Section 5.2, the current
plant arc furnace operation Kst is calculated to be 41. This
value is below the industry lows. Table X compares this versus
the predicted industry standard values.
Flicker PST
GREEN = lowest, YELLOW = middle, BLUE = maximum, RED = highest 1.0

0.5
B. Measured Value of Flicker – Present Operation
0.0
The present operation with an analog regulator and hoist
1.0
drives yielded a Pst value of about 5 as shown in Fig. 12 at the
0.5
46 kV furnace transformer primary. This compares with peak
flicker data below from the utility. An estimate, based on the 0.0

voltage drop due to a short circuit at the electrode tips of EAF4, 1.0
indicates a prediction of 25% of the current measured EAF4. 0.5
This would yield a Pst value of 1.25 without SVC 0.0
compensation. With a SVC reduction factor of 2, this would
yield a Pst value of .63, below the required limit of .8. These 18:00 00:00 06:00 12:00 18:00 00:00 6:00
Tue. Wed. Thu.
values are in magnitude of the analysis in Section 5.1. This Time (hours: minutes)
would indicate that the current Standard Steel furnace Kst
operation is lower than 65. Figure 13. Flicker Measured Values for Current Operation at Yeagertown 230
kV PCC (1/18/13 – 1/22/13)

Flicker PST and PLT

Flicker PST
1.5
20
1.0
15 0.5
0.0
10 1.5
1.0
15
0.5
0 0.0
1.5
20 1.0
0.5
15 0.0

10 12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00 12:00

Fri. Sat. Su. Mon. Tue.
15 Time (hours: minutes)

0
Figure 14. Flicker Measured Values for Current Operation at Yeagertown 230
20 kV PCC (1/22/13 – 1/24/13)

15
D. SVC Sizing for Flicker Reduction
10
A formula for sizing the SVC based on flicker reduction in
15 [9] is
0 Rscv~= 1 + .75(Sscv/Sf) (3)
where Sscv is the SVC size in MVARs and Sf is MVARs
14:00 14:15 14:30 14:45 15:00 15:15 15:30 15:45
Time (hours: minutes) required by the furnaces. Typical traditional SVC design is
limited to a Rscv of 2. Table XI illustrates the calculation of the
Figure 12. Flicker Measured Values for Current Operation of EAF4 10/17/12 SVC size in MVAR to achieve a Pst reduction factor of 2.

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

TABLE XI. SVC SIZING TABLE Harmonic

Order Magnitude
Phase
Angle
1 100.0 298.82
2 6.3 47.12 EAF4 Harmonic Current - borein
SVC Sizing 3 5.9 118.82
4 3.5 7.478 150
Power Factor Correction (From Case 2) MVAR 5
6
1.7
0.2
317.8
210.21
7 0.5 174.11 100
SVC output to maintain unity PF at PCC 41.5 8 1.1 330.6
9 0.3 212.97 50

Current (Percent)
SVC with 20% increase above that for PFC 50 10
11
0.5
0.1
177.74
128.42
12 0.2 62.26 0
Flicker Mitigation with SVC reduction factor of 2 13 0.2 109.23 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035

-50
SVC sized per formula above for flicker 66
-100
SVC with 20% increase above that for flicker 80 % THD 9.50%
RMS 100.45
-150
TOTAL 120.33
Time (Seconds)
E. Conclusions
Note: Above current in percent of fundamental current of 53,162A.
Operation of the EAF4 and LMF will violate the utility
requirement for flicker value of less than Pst of .8 without SVC Figure 15. 2 EAF Harmonic Current Spectrum for Bore-In
compensation. Application of a SVC will support utility flicker Harmonic Phase

requirement with furnace Kst less than 56. It is anticipated that Order
1
Magnitude
100.0
Angle
339.8
EAF4 Harmonic Current - refining
with the updated regulator and mast systems for both furnaces, 2
3
1.2
1.5
20.877
177.05
4 0.7 88.26
Kst will be lower than the current operation which has been 5
6
1.3
0.5
354.6
341.2
150

calculated at 41. SVC will also be required to be sized for 7

8
1.1
0.1
137.92
160.59
100

flicker reduction. The minimum size will be 66.67 MVARs 9 0.2 80.17 50

Current (Percent)
10 0.2 276.44

plus a 20% factor for a total of 80 MVARs. 11

12
0.1
0.1
128.45
99.42 0
13 0.1 323.6 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035

-50
VI. HARMONIC ANALYSIS
-100

The intent of the Harmonic Analysis was to size and % THD

RMS
2.74%
100.04
-150
evaluate the proposed SVC reactive compensation to mitigate TOTAL 107.20
Time (Seconds)

harmonics and power factor for the new EAF and LMF. Note: Above current in percent of fundamental current of 30,853A.
A. Approach Figure 16. LMF Harmonic Current Spectrum
The approach taken in the Harmonic Analysis was to size
the SVC filter bank to comply with the utility interconnect as shown in Fig. 17. The sources included three EAF
requirements at the 230 kV point of common coupling (PCC) manufacturers and EAF harmonic data compiled in [7]. The
for harmonics (IEEE Std 519), flicker (IEEE Std 1453) and measured values agree with those from the other sources. This
unity power factor. The harmonic analysis considered two (2) gave a high level of confidence in the harmonic model used for
possible system configurations for worst-case harmonic EAF4 in this study and used in sizing the SVC filter
production by the EAF and LMF: 1) No compensation, 2) SVC components.
with TCR and filter bank. The analysis followed guidelines in 9
IEEE Standards such as IEEE Std 519-1992 and IEEE Std 18- Harmonic Currents - EAF Bore-In
8
2006 in performing the analysis.
7
The harmonic analysis provided conceptual specifications
6 Vendor1
and basic ratings for the harmonic filters and SVC. In addition
to sizing and evaluating the SVC filter bank, existing problems 5 Miller
or areas of concern, and make engineering recommendations, if 4 Vendor2
needed. 3 Vendor3
Measured
B. Harmonic Sources 2
The source harmonics listed below were added to the Load 1
Flow model of Section III. These sources were developed from 0
harmonic measurements taken during a typical heat of the 2 3 4 5 6 7
existing EAF#4 in October 2012. For EAF4, the first bucket
bore-in was chosen due to the high power used and the high Figure 17. Comparison of EAF Harmonic Currents (Bore-In)
harmonic content due to the nature of the arcing load. For
LMF, the refining period of the heat was also chosen to use as In a similar way, the measured harmonic spectrums for the
part of this study because of the large amount of time devoted EAF at refining were compared to those from other recognized
to refining during a typical heat. The snapshots shown below sources as shown in Fig. 18. The measured values agree with
illustrate the harmonics produced by the furnaces: EAF4 first those from the other sources. This gave a high level of
bucket bore-in in Fig. 15 and LMF refining in Fig. 16. confidence in the harmonic model used for EAF4 in this study
and used in sizing the SVC filter components.
The measured harmonic spectrums for the EAF during
bore-in were compared to those from other recognized sources

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

4 The individual harmonic voltage and currents at the PCC

Harmonic Currents - EAF Refining with the SVC is compared to that without the SVC in Tables
3.5
VIII and XIV. Table VIV shows the reduction in the ITHD
3 due to the SVC filters. Table VIV also shows the reduction in
2.5
Vendor1 the individual harmonic currents due to the SVC filters.
Miller Plot1- DistortionWaveform
2 Study1 - Case1 - PCC - Bus Distortion Voltage WaveForm
Vendor2

Bus Distortion Voltage WaveForm (Volts)

300000
1.5 Vendor3 200000

100000
1 Measured
0

0.5 -100000

-200000
0 -300000
2 3 4 5 6 7 0 50 100 150 200 250 300 350
T ime (1 Cycle) or 0-360 degrees
Plot1- DistortionWaveform
Figure 18. Comparison of EAF Harmonic Currents (Refining) Study1 - Case2 - PCC - Bus Distortion Voltage WaveForm

Bus Distortion Voltage WaveForm (Volts)

300000

C. Case Descriptions 200000

100000
The performance of the proposed SVC harmonic filter bank 0
consisting of a 2nd, 3rd, 4th and 5th harmonic filter stages is -100000
analyzed for the worst-case harmonic condition. The worst- -200000

case harmonic condition consists of the EAF4 in first bucket -300000

bore-in and the LMF in refining, with both furnaces at rated 0 50 100 150 200
T ime (1 Cycle) or 0-360 degrees
250 300 350

operating point simultaneously. Harmonic analysis at the 230 Figure 19. Voltage Waveform at 230kV PCC Case 1 vs. Case 2
kV point of common coupling (PCC) was performed to verify
compliance with [2]; frequency scans were conducted to Plot1- DistortionWaveform
Study1 - Case2 - CBL-PCC - Branch Distortion Current Waveform
determine the parallel and series resonance points of proposed
Branch Distortion Current Waveform (Amps)

200
filter bank; and load flow analysis was conducted to verify
100
system voltage and kVAR output of the SVC as well as
compliance with utility PF requirements at the PCC. 0

Two (2) harmonic analysis study cases were performed and -100

evaluated on the basis of [2]: -200

0 50 100 150 200 250 300 350
 Case 1 – Both EAF4 and LMF4 at rated operating T ime (1 Cycle) or 0-360 degrees
Plot1- DistortionWaveform
points without SVC Study1 - Case1 - CBL-PCC - Branch Distortion Current Waveform
Branch Distortion Current Waveform (Amps)

200
 Case 2 – Both EAF4 and LMF4 at rated operating 100
points with SVC 0

D. IEEE Std 519 Harmonic Analysis at PCC -100

Harmonic results are provided for Cases 1 and 2 with and -200

without the SVC respectively in Figs. 19 and 20 and Tables XII 0 50 100 150 200 250 300 350
T ime (1 Cycle) or 0-360 degrees
through XIV. Table XII shows that in both cases studied (with
or without the SVC filter bank), the voltage total harmonic Figure 20. Current Waveform at 230kV PCC Case 1 vs. Case 2
distortion (VTHD) is within IEEE Std 519-1992 limits. This
low VTHD can be seen in the voltage waveforms of Fig. 19. TABLE XII. COMPARISION OF TOTAL HARMONIC DISTORTION AT PCC WITH
AND WITHOUT SVC
Table XII shows without harmonic filters in Case 1, the STUDY VTHD (%) ITHD (%)
current total harmonic distortion (ITHD) exceeds IEEE Std CASE 230kV Bus IEEE Limit 230kV Bus IEEE Limit
519-1992 limits. For example, with the EAF and LMF
simultaneously at rated operating point, the ITHD is 6.87% Case 1 0.45 1.50 6.87 2.50
well above the limit of 2.5%. The 2nd, 3rd and 4th harmonic Case 2 0.13 1.50 2.72 3.80
currents exceed the respective individual harmonic limits. This RED = in excess of IEEE Std. 519-1992 Limits
high ITHD is visible in the current waveform of Fig. 20 (top).
Case 1 = EAF and LMF at rated operating point w ithout SVC
Table XII shows with the application of the recommended Case 2 = EaF and LMF at rated operating point w ith SVC
filter bank in Case 2, the current total harmonic distortion
(ITHD) is reduced to within IEEE Std 519-1992 limits. For
example, with the EAF and LMF simultaneously at maximum
load, the ITHD is 2.72% well below the limit of 3.8%. The
2nd harmonic current of 2.18% is slightly above the respective
individual limit of 0.75%. In this case, the current distortion is
not adversely affecting the voltage distortion and is not
considered a problem. This significantly reduced ITHD is
visible in the current waveform of Fig. 20 (bottom).

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

Plot1- ScanMagnitude
TABLE XIII. COMPARISON OF INDIVIDUAL HARMONIC VOLTAGE Study1 - Case2 - BUS-SVC - Bus Scan Impedance Magnitude

DISTORTION AT PCC WITH AND WITHOUT SVC

Bus Scan Impedance Magnitude (pu)

10 0
Harmonic CASE 1 CASE 2
Order Harmonic Percent Harmonic Percent IEEE
h Voltages Distortion Voltages Distortion Limit
10 -1
1 226318.0 229495.5
2 428.4 0.189 159.1 0.069 1.0
0 5 10 15 20 25 30 35 40 45 50
3 595.7 0.263 145.3 0.063 1.0 Harmonic Order
Plot1 - Scan Magnitude
4 468.1 0.207 98.3 0.043 1.0 Study1 - Case2 - BUS-SVC - Bus Scan Impedance Magnitude

Bus Scan Impedance Magnitude (pu)

5 351.9 0.156 72.2 0.031 1.0
6 57.3 0.025 22.6 0.010 1.0 10 0

7 46.4 0.021 10.8 0.005 1.0

8 320.6 0.142 126.4 0.055 1.0
10 -1
9 107.3 0.047 45.0 0.020 1.0
10 190.4 0.084 77.6 0.034 1.0
1 2 3 4 5 6 7 8 9 10
11 63.1 0.028 28.2 0.012 1.0 Harmonic Order

12 74.1 0.033 31.8 0.014 1.0 Figure 21. Frequency Scan for SVC Filter Bank (top) and Zoom (bottom)
13 89.6 0.040 40.3 0.018 1.0
TABLE XV. SUMMARY OF SERIES AND PARALLEL RESONANCE POINTS
VTHD 226320.7 0.450 229495.7 0.130 1.5 FROM FREQUENCY SCAN

RED = in excess of IEEE Std. 519-1992 Limits

Case 1 = EAF and LMF at rated operating point w ithout SVC
Case 2 = EaF and LMF at rated operating point w ith SVC

TABLE XIV. COMPARISON OF INDIVIDUAL HARMONIC CURRENT F. IEEE Std 18 Filter Evaluation for Proposed Filter Bank
DISTORTIONAT PCC WITH AND WITHOUT SVC Ratings for capacitor voltage, current and kVA which
Harmonic CASE 1 CASE 2 manufacturers must meet are defined in [3]. When specifying a
Order Harmonic Percent Harmonic Percent IEEE filter bank, it is important to evaluate the predicted harmonic
h Currents Distortion Currents Distortion Limit currents flowing into the filter bank with respect to the ratings
1 184.5 142.0 for the capacitor per [3]. The proposed filter is comprised of
2 8.3 4.511 3.1 2.177 0.75
four stages: a 2nd Harmonic Filter (tuned to the 1.95th), a 3rd
Harmonic Filter (tuned to 2.9th), a 4th Harmonic Filter (tuned to
3 7.7 4.186 1.9 1.328 3.00
the 3.8th) and a 5th Harmonic Filter (tuned to the 4.7th). Table
4 4.5 2.468 1.0 0.674 0.75
XVI summarizes the IEEE Std 18 evaluation of the proposed
5 2.7 1.485 0.6 0.396 3.00 filter banks. Table XVII illustrates the details of the capacitor
6 0.4 0.201 0.1 0.103 0.75 voltage, current and kVA for the proposed harmonic filters.
7 0.3 0.140 0.1 0.042 3.00 The individual harmonic currents, voltages and kVA for each
8 1.6 0.845 0.6 0.433 0.75 stage are given and indicate the harmonic duty on each stage.
9 0.5 0.251 0.2 0.137 3.00 The capacitor voltage, current and kVA for all stages are within
10 0.7 0.402 0.3 0.213 0.75 IEEE Std 18 limits for the worst-case harmonic conditions
11 0.2 0.121 0.1 0.071 3.00 modeled for the furnaces.
12 0.2 0.130 0.1 0.073 0.75
TABLE XVI. TABLE 6-11 SUMMARY OF IEEE STD 18
13 0.3 0.145 0.1 0.085 3.00
EVALUATION OF ALL FILTER BANKS
ITHD 185.0 6.870 142 2.720 3.80 STUDY Votlage Current MVA
CASE Voltage Rated % Current Rated % MVA Rated %
RED = in excess of IEEE Std. 519-1992 Limits Filter 2nd 62325 69005 90.3 341 377 90.6 36.8 45.0 81.7
Case 1 = EAF and LMF at rated operating point w ithout SVC Filter 3rd 52123 52395 99.5 298 298 100.0 26.8 27.0 99.2
Case 2 = EaF and LMF at rated operating point w ith SVC Filter 4th 49320 49883 98.9 248 250 99.1 21.1 21.6 97.9
Filter 5th 48076 49883 96.4 161 167 96.5 13.4 14.4 92.9

IEEE Limit 110 IEEE Limit 135 IEEE Limit 135

RED = in excess of IEEE Std. 18-2006 Limits

E. Frequency Scan of Propose Filter Bank
The frequency scan for the proposed SVC 4-stage filter
bank is shown below in Fig. 21. The proposed filter bank is
parallel resonant at the 1.81, 2.60, 3.45 and 4.45 harmonics and
series resonant at the 1.95, 2.9, 3.8 and 4.7 harmonics as shown
in Table XV. These series resonant points (filter tuning points)
are by design and force the parallel resonant points away from
the 2nd, 3rd, 4th and 5th harmonic currents produced by the arc
furnaces. This is an important advantage of this filter bank
design.

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

H. Conclusions
TABLE XVII. CASE 2 IEEE STD 18 EVALUATION OF FILTER BANK –
HARMONIC DUTY ON EACH STAGE Installing the four (4) stage harmonic filter bank in the SVC
will enable compliance with the utility interconnect
Filter Spectrum Report
requirements for harmonics. The detailed specifications for the
Harm
Order 2nd Filter 3rd Filter 4th Filter 5th Filter 2nd, 3rd, 4th and 5th harmonic filters are given in Table XIX.
h A KVAR V A KVAR V A KVAR V A KVAR V
1 339.7 36640 62266 295.7 26684 52088 247.1 21106 49310 160.6 13373 48072 The advantages of the harmonic filters include: 1) reduction
2 29.3 136.1 2683 3.245 1.606 285.7 2.078 0.741 207.3 1.223 0.387 183 of harmonic distortion limits (both voltage and current) within
3 1.6 0.292 101.4 32.23 106.1 1896 5.531 3.524 367.8 2.345 0.951 234 IEEE Std 519-1992 limits at the 138 kV PCC, 2) improved
4 0.65 0.034 30.18 2.435 0.452 107.2 17.63 26.84 9E+05 4.684 2.844 350.5
5 0.27 0.005 9.785 0.785 0.038 27.65 1.867 0.241 74.5 6.827 4.833 408.6
voltage support during simultaneous operation of the EAF and
6 0.055 2E-04 1.695 0.149 0.001 4.374 0.289 0.005 9.61 0.456 0.018 22.77 LMF, and 3) power factor correction to 95% power factor at
7 0.072 2E-04 1.885 0.184 0.001 4.642 0.328 0.005 9.36 0.43 0.014 18.39 the utility 138 kV PCC.
8 0.312 0.004 7.152 0.778 0.023 17.13 1.322 0.076 32.98 1.594 0.165 59.67
9 0.094 3E-04 1.917 0.23 0..18 4.508 0.381 0.006 8.44 0.438 0.011 14.59 In either case (with or without the SVC filter bank),
10 0.155 8E-04 2.844 0.375 0.004 6E+05 0.608 0.013 12.14 0.681 0.024 20.39
11 0.046 6E-05 0.77 0.111 3E-04 1.773 0.177 1E-03 3.216 0.194 0.002 5.301
harmonic currents present has no detrimental effect upon the
12 0.042 7E-05 0.8 0.125 4E-04 1.828 0.198 0.001 3.285 0.214 0.002 50341 transformer rating as shown by the K-factor calculations. No
13 0.055 7E-05 0.779 0.131 4E-04 1.772 0.206 0.001 3.16 0.221 0.002 5.086 transformer de-rating is needed due to harmonics.
G. Harmonic K Factor Calculation Measurements of harmonics, flicker and other power
During the course of the study, the utility raised concern quality parameters should be taken at the 230kV PCC during
about the harmonic loading on the Yeagertown Substation the startup and commissioning of the EAF4, LMF and SVC.
transformers and wanted to know the K-factor imposed on Measurements should be taken while operating the furnace at
these transformers. The K-factor is a number derived from a their rated operating points to verify compliance with the utility
numerical calculation based on the summation of harmonic interconnect requirements for harmonics [2], flicker [4] and
currents generated by the non-linear load. The higher the K- unity power factor. Additional measurements at the furnaces
factor, the more significant the harmonic current content. The and SVC should be taken to supplement the verification of the
K-factor can be calculated by the following equation in [11] performance of the SVC.

TABLE XIX. SPECIFICATION OF HARMONIC FILTER BANK FOR SVC

K-factor = Σ(I2h2) (4) Specifications 2nd 3rd 4th 5th
series tuned harmonic order 1.95 2.9 3.8 4.7
where h is the harmonic number and I is the normalized per Capacitor rated KVAc (total three-phase) 45000 27000 21600 14400
unit current based on the actual RMS current. In Table XVIII, Capacitor rated kV (phase-to-phase) 69 52 50 50
the K-factor was calculated at the 46kV level based off of Capacitance (microfarads) 25 26 23 15
measured harmonics and 230/46kV transformer ratings from Effective KVAc (total three-phase) 19997 20811 18368 12245
Effective kV (phase-to-phase) 46 46 46 46
Yeagertown transformers YT1 and YT2 respectively. All
Reactance rated (ohm per phase) 27.8 12.1 8.0 7.8
values (with and without SVC compensation) results in values Reactance (millihenry per phase) 73.8 32.1 21.2 20.7
slightly above 1, concluding that the transformer loading has no Reactance rated current see below see below see below see below
issues with harmonic currents and will not need de-rated. fundamental (A) 565 446 375 250
I2 75% 35% 15% 10%
TABLE XVIII. K-FACTOR CALCULATION FOR TRANSFORMERS YT1 & YT2 I3 35% 75% 35% 15%
I4 15% 35% 75% 35%
46kV FE Yeagertown Substation Transformers
Harmonic I5 10% 15% 35% 75%
Order YT1 With SVC YT1 Without SVC YT2 With SVC YT2 Without SVC I6 10% 10% 15% 35%
h I I2h2 I I2h2 I I2h2 I I2h2 I7 10% 10% 10% 15%
1 355.5 0.999261 461.8 0.995298 354.8 0.999262 461.1 0.995307 I8 5% 10% 10% 10%
2 7.74 0.001895 20.83 0.0081 7.72 0.001892 20.796 0.008098 I9 and above 5% 5% 5% 5%
3 4.72 0.001585 19.34 0.015711 4.708 0.001584 19.27 0.015645
4 2.39 0.000723 11.4 0.009705 2.39 0.000725 11.375 0.009691
5 1.4 0.000387 6.86 0.005491 1.4 0.000389 6.847 0.005487
6 0.366 3.81E-05 0.93 0.000145 0.366 3.83E-05 0.927 0.000145
VII. CONCLUSIONS
7 0.15 8.72E-06 0.65 9.66E-05 0.149 8.64E-06 0.645 9.54E-05 This paper described a sequence of comprehensive
8 1.54 0.0012 3.9 0.004543 1.54 0.001205 3.896 0.004548
9 0.487 0.000152 1.17 0.000517 0.48 0.000148 1.157 0.000508
analytical studies performed to specify the SVC needed to
10 0.757 0.000453 1.86 0.001615 0.755 0.000452 1.85 0.001602 mitigate the increased electrical load and harmonic generation
11 0.252 6.08E-05 0.56 0.000177 0.2519 6.09E-05 0.558 0.000176 of the new EAF operation to meet utility interconnect
12 0.259 7.64E-05 0.6 0.000242 0.259 7.67E-05 0.599 0.000242 requirements. The results of the Load Flow Analysis, Flicker
13 0.302 0.000122 0.67 0.000354 0.302 0.000122 0.669 0.000354
Analysis and Harmonic Analysis Studies presented in this
K factor 1.0060 1.0420 1.0060 1.0419
paper were used to size and evaluate the proposed SVC
reactive compensation to mitigate flicker, harmonics and power
In either case (with or without the SVC filter bank), factor for the new EAF and LMF installations. The paper
the presented harmonic currents have no detrimental effect recommended installing a four (4) stage harmonic filter bank
upon the transformer rating. No transformer de-rating is integral to the SVC and provided detailed specifications for the
needed due to harmonics. 2nd, 3rd, 4th and 5th harmonic filters were given.
Measurements of harmonics, flicker and other power quality
parameters at the utility PCC during the startup and
commissioning of the EAF, LMF and SVC will allow confirm
the performance of the SVC to the design specifications.

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/TIA.2015.2451072, IEEE Transactions on Industry Applications

REFERENCES [7] T.E. Miller, “Reactive Power Control in Electric Systems”, John Wiley
& Sons, pp. 304-309, 1982.
[1] IEEE Std 1031-2000, IEEE Guide for the Functional Specification of [8] S.R. Mendis, M.T. Bishop, A.V. Do and D.M. Boyd, “Investigation
Transmission Static Var Compensators, IEEE SA Standards Board, of Transmission System Voltage Flicker due to Multiple AC and
2000. DC Furnace Operations”, IEEE Transactions on Power Delivery,
Vol. 10, No. 1, pg. 490.
[2] IEEE Std 519-1992, IEEE Recommended Practice and Requirements for
Harmonic Control in Electric Power Systems, IEEE, 1992, tables 10.2 [9] A. Robert, M. Couvreur, “Arc Furnace Flicker Assessment and
and 10.3, pp. 77-78. Prediction”, in Proc. 12 th Int. Conf. Elec. Distrib., CIRED, 1993,
vol. 2, pp. 17-21.
[3] ANSI/IEEE Std 18-2006, IEEE Standard for Shunt Powr Capacitors,
IEEE Standards Board 1992, ANSI 1993, section 5.2.3, p. 4. [10] Y.J. Hsu, K.H. Chen, P.Y. Huang and C.N. Lu, “Electric Arc
Furnace Voltage Flicker Analysis and Prediction”, IEEE
[4] IEEE Std 1453-2011, IEEE Recommended Practice - Adoption of IEC
Transactions on Instrumentation and Measurement, Vol. 60, No.
61000-4-15:2010, Electromagnetic compatibility (EMC) - Testing and 10, October 2011 .
measurementtechniques – Flickermeter - Functional and design
specifications, IEEE SA Standards Board, 2011. [11] IEEE Std 1100-1999, Powering and Grounding Electronic Equipment,
IEEE, 1999, pp. 103-105.
[5] J.A. Ciotti, D.L. Pelfrey, “Electrical Equipment and Operating Power
Characteristics”, Electric Furnace Steel Making, Chapter 3, pp. 34-38,
1985.
[6] A.R. Oltrogge, “Arc-furnace voltage can be critical”, Electrical
World, December 1974.

0093-9994 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.