Results in Materials 8 (2020) 100146

Contents lists available at ScienceDirect

Results in Materials
journal homepage: www.journals.elsevier.com/results-in-materials

Annealer curve characteristics of electrolytically refined tough pitch copper

(Cu-ETP) and oxygen free up-cast copper (Cu-OF) for electrical cable wires
C. Banganayi a, *, K. Nyembwe a, K. Mageza b
a
University of Johannesburg, Department of Metallurgy Johannesburg, South Africa
b
University of Johannesburg, Metal Casting Technology Station (MCTS) Johannesburg, South Africa

A R T I C L E I N F O A B S T R A C T

Keywords: Annealing is a critical step during the wire drawing of copper to produce wires with the required ductility and
Cu-OF conductivity for electrical applications. In this investigation, the characteristics of commercial annealers for wire
Cu-ETP drawing of electrolytically refined tough-pitch copper (Cu-ETP) and oxygen-free upcast copper (Cu-OF) wires
Percentage elongation
were assessed through annealer curves with respect to applied voltage and resulting percentage elongation of final
Resistance annealing
Annealer curve
products. Experiments were carried out on three M315 Niehoff machines of varying speeds used for medium wire
Wire drawing drawing from 2.2 mm diameter to 0.66 mm and 0.53 mm diameters. The results show similarities and differences
Dynamic recrystallisation between annealer curve characteristics of wire drawing machines regarding the required voltage for minimum
and maximum elongation of wires without oxidation depending on the type of copper raw material used. In
particular, the minimum 18% required elongation of wires was achieved with the same applied voltage irre-
spective of the machine speed and type of copper. In addition, Cu-OF reached higher maximum percentage
elongation without oxidation of the wire compared to Cu-ETP.

1. Introduction
Copper plays a crucial role in the development of any country because
it is the main material used in cables for the distribution of electrical
power to and within human settlements, factories and transportation
systems. Copper wires for electrical purposes are manufactured using
standard grade electrolytically refined tough-pitch copper (Cu-ETP) or
oxygen-free copper (Cu-OF) from an upcast process [1]. The dynamically
recrystallised grains of the Cu-ETP are equiaxed while the ones of Cu-OF
are as-cast, large columnar grains. Fig. 1 shows the typical microstruc-
tures of Cu-ETP and Cu-OF [2,18].
1.1. Copper wire drawing
The process of wire drawing of copper involves rod breaking where a
7.9- or 8-mm diameter (Ø) rod is drawn through successive tungsten
carbide dies of 6 mm Ø down to about 2.2 mm Ø or less. The dies are
submerged in a water-soluble lubricant that cools down the wire and
reduces the wear on the dies during the wire drawing process. The wire is
finally passed through an inline electric resistance annealer which uses
steam as a reducing environment to prevent oxidation of copper wires
during annealing and helps maintain the bright colour of the copper [3].
The metallurgical annealing process of the cold-drawn wires restores the
ductility and electrical conductivity.
The literature has shown that the recrystallisation temperature of Cu-
ETP is affected by different amounts of various impurities [4]. According
to the studies, sulphur (S), lead (Pb) and bismuth (Bi) increase the
recrystallisation temperature of pure copper (Cu). It was also found that a
10-ppm increase in S, for example, raises the recrystallisation tempera-
ture by 60
C. (Fig. 2).
As the residence time of wire during inline annealing is very short
(milliseconds) any increase in impurity levels of copper may be detri-
mental to the mechanical properties of the final drawn wire. Control of
the main impurities is also important in maintaining the high conduc-
tivity of copper measured by %IACS (International Annealed Copper
Standard) [5]. There has also been extensive work on the effect of im-
purities, such as phosphorous (P), sulphur (S) and oxygen (O), on the
properties, including electrical conductivity, of Cu-OF [6]. This research
found that P exists in solid solution in Cu at 300–800
C. P in solution
decreases conductivity by 0.73% for each l0 ppm to 60 ppm added [6]. In
practice P is kept at <5 ppm. (See Annexure A of results of materials used
in this experiment). Regarding S, only the existence of Cu2S, which is
thought to cause hot shortness during rolling, has been reported. The
presence of O as Cu2O and its effects on hot and cold shortness have been

* Corresponding author.
E-mail addresses: cbanganayi@gmail.com (C. Banganayi), dnyembwe@uj.ac.za (K. Nyembwe), kmageza@uj.ac.za (K. Mageza).

https://doi.org/10.1016/j.rinma.2020.100146
Received in revised form 6 September 2020; Accepted 9 September 2020
Available online 3 October 2020
2590-048X/© 2020 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
C. Banganayi et al.
Fig. 1. Microstructure of a) Cu-ETP b) Cu-OF [2,11].
Fig. 2. Effects of impurities on the recrystallisation temperature [8].
reported, especially regarding its effects on conductivity and recrystal-
lisation temperature. Cu2O is reported to reduce electrical conductivity
by 0.126% for each 0.01% of oxygen [6,7]. It is for this reason that when
producing Cu-OF, the cathodes are remelted in a highly reducing envi-
ronment by lancing carbon monoxide (CO) or using graphite or charcoal
[1].
1.2. Annealer curves
Annealer curves are graphs that show the variation in the degree of
annealing as a function of percentage elongation as a function of the
annealer applied voltage during inline resistance annealing [9]. The
graph has an S-curve, which shows a slow increment of percentage
elongation at the beginning followed by a rapid rise and flattening to-
wards the end. The S-curve is in line with the three-solid state phe-
nomenon involved during metallurgical annealing, namely recovery,
recrystallisation and grain growth in the electrical conductor [10]. The
determination of annealer curves is essential for the optimisation of
copper cable manufacturing, in order to obtain the required electrical
properties, including resistance and conductivity. Because of the differ-
ence in the microstructure of the two types of copper cable raw materials,
experience has shown that Cu-OF will anneal at higher voltages (and
therefore higher temperatures) than Cu-ETP, usually in the order of
10–15% or more [11].
Annealer curves of copper wires are very seldom published in scien-
tific literature as they are considered confidential operational parame-
ters. In this paper, the annealer curves were derived to determine the
common voltage and machine speed for processing different wire sizes of
Cu-ETP and Cu-OF without having to adjust the machine parameters each
time a different material was used. As such, the paper provides useful
data for process control during annealing of commercial copper wires. A
better understanding of copper cable response during annealing could be
gained from this investigation.
2
Results in Materials 8 (2020) 100146
2. Experimental procedure
2.1. Raw materials
A Cu-ETP rod of 7.9 mm Ø and a Cu-OF rod of 8 mm Ø with chemical
compositions, as shown in Annexure A, were used at the start of the wire
drawing process. These copper rods were initially drawn down to 2.2 mm
Ø wire size. The 2.2 mm Ø wire was further drawn down to 0.66 mm Ø
and 0.53 mm Ø using medium wire drawing machines. The wire sizes
used for the experiment were selected because they are the most common
sizes used at the South African copper cable manufacturing plant where
this investigation was carried out.
2.2. Wire drawing machines and operating parameters
The experiments were carried out on M315and MM315 Niehoff ma-
chines. The machines used for medium wire drawing were equipped with
inline resistance annealers [16]. These machines are referred to as ma-
chine 1, machine 2 and machine 3. Machine 1 was a seven-wire drawing
machine used to draw wires from 2.2 mm to 0.53 mm Ø, while machines
2 and 3 were single-wire drawing machines used to draw 2.2 mm Ø wire
to 0.66 mm Ø wire. Castrol Iloform BWS 153 lubricant was used during
the wire drawing and annealing processes. The pH and fat content of the
lubricant were kept at the operating levels prescribed by the manufac-
turer. Steam from the lubricant was used as the reducing environment.
The different machine speeds selected for wire drawing varied from 9 to
20 m/s. The speeds for each machine were kept constant throughout the
experiments. The voltage was then varied for all three machines starting
from a minimum of 13 V to a maximum of 38 V, in line with production
parameters to avoid wire tarnishing due to oxidation at high tempera-
tures. The empirical relationship between the speed of the wire and the
voltage is given in Equation (1) below.
pffiffiffi
UVolts ¼ ð vÞ:k (1)
where U ¼ voltage, v ¼ speed of wire and k ¼ annealer factor with k ¼
8.2.
During resistance annealing a current is passed through the wire at
different contact points by means of contact wheels while the wire is
being transmitted at very high speeds, sometimes up to 60 m/s. The
heating during resistance annealing is based on the adiabatic heating
which ignores heat loss during heating and only concentrates on the
temperature rise. This is based on the following adiabatic heating
equation [12].
J2 ðtÞρ ðTÞdt ¼ μCðTÞdT (2)
where: t ¼ time.
T ¼ temperature
ρ(T) ¼ resistivity
μ ¼ density.
C(T) ¼ specific heat.
C. Banganayi et al. Results in Materials 8 (2020) 100146

Fig. 3. Graph of voltage vs percentage elongation for 0.53 mm Ø, 7 strands Cu-

Fig. 4. Graph of voltage vs percentage elongation for 0.66 mm Ø Cu-ETP and
ETP and Cu-OF copper run on machine 3 at 9 m/s.
Cu-OF copper run at 11 m/s on machine 2.

J ¼ current density of conductor.

machine 2 for the two types of copper. As can be observed from both
The ohmic heat produced during a time interval, dt, causes a tem-
annealer curves, the minimum required 18% elongation for single wire
perature rise of dT in the conductor [13].
drawing was achieved at 17 V.
The above equation can be integrated and used to find the relation-
Because the wires are further processed by bunching and insulation it
ship between the cross-section area of the wire and current (I), which is
is advised that the percentage elongation be much higher than the
then used to calculate different settings for different wires on different
minimum value to avoid failures at later stages of processing. The highest
machines.
percentage elongation values of 31% were achieved at 23.5 V for both
The experimental determination of an annealer curve was carried out
Cu-ETP and Cu-OF. The wire at this voltage was tarnished due to
following a prescribed and common procedure that was found to be
oxidation. Cu-OF was observed to start tarnishing at 23.5 V while Cu-ETP
practical and to work best. Once the specific speed of the machine was
tarnished earlier at 22.6 V at an elongation of 29.45%. For machine 2,
selected, the voltage was then increased slowly while the samples in the
there is a big operating window for the adjustment of annealer voltages
wire bobbins were separated by a sample marker with the specific
while still achieving the correct elongation and colour of the wire if
voltage written on it. This was done until the wire turned dark or was
current conditions of the machine are to be used. For this machine, a
tarnished. The bobbins were then taken for rewinding while collecting
common voltage range of 21–23 V will yield the maximum percentage
the samples at each test voltage for tensile test.
elongation at the 11 m/s without tarnishing either wire.
Fig. 5 shows the annealer curve results of experiments carried out on
2.3. Mechanical testing machine 3. The minimum required elongation of 18% was achieved at
around 28 V which is much higher than that of machines 1 and 2 which
Elongation was the most important mechanical property of the wire occurred at 17 V and 18 V respectively. This could be attributed to less
to be determined as this was used to establish the annealer curves for all resistance on the annealer of machine 3 due to a higher operating wire
the machines for the different types of copper used in the medium wire drawing speed of 20 m/s compared to 9 m/s and 11 m/s for machines 1
drawing machines. For each point on the annealer curve, three samples and 2 respectively.
were used to obtain the average percentage elongation value for all three The highest percentage elongation values of 35% at 38.2 V for Cu-ETP
machines. The percentage elongations were obtained through tensile
(UTS) measurements on an Instron 3369 Tensile test machine loaded
with Bluehill software at a constant strain rate of 96 mm/min at room
temperature.

3. Results and discussions

3.1. Annealer curves

Fig. 3 shows the annealer curve results on machine 1. The minimum

required elongation of 18%, as specified by SANS 1411–1:2008 [14], was
achieved at around 18 V. 22% for the single wire and 20% for the multi
wire was achieved at 20 V, as specified by BS EN 13602:2013 [15]. The
results display an average taken from 7 wires. The highest percentage
elongation values of 29% at 25 V for Cu-ETP and 36.5% at 23 V for Cu-OF
were achieved. Cu-OF was observed to start tarnishing at 24 V and 34%
elongation while Cu-ETP tarnished earlier at 28 V at an elongation of
27%. For this machine, a common voltage range of 21–23 V will yield the
maximum percentage elongation at the optimum speed without tarn-
ishing either wire. Fig. 5. Graph of voltage vs percentage elongation for 0.66 mm Ø Cu-ETP and
Fig. 4 shows the annealer curves related to medium wire drawing Cu-OF copper run at 20 m/s on machine 3.

3
C. Banganayi et al.
Fig. 6. Applied voltage required for 18% minimum elongation of wire.
and 36.5% at 34 V for Cu-OF copper were achieved. Although higher
ductility is achieved at these voltages, the wires could not be used as they
were tarnished due to oxidation. Cu-OF was observed to start tarnishing
at 33 V and 33% elongation while Cu-ETP tarnished earlier at 31 V and
an elongation of 29%. For machine 3, there is an even larger operating
window for the adjustment of annealer voltages while still achieving the
correct elongation and colour of the wire if the present conditions of the
machine are to be used. For this machine, a common voltage range of
29–32 V yielded the maximum percentage elongation at 20 m/s without
tarnishing either wire.
Fig. 6 provides additional insight into the relationship between the
machine speed and the voltage required to achieve the minimum elon-
gation in the different materials.
In addition, there seems to be no difference between Cu-OF and Cu-
ETP regarding the required voltage to achieve the minimum percent-
age elongation of 18% irrespective of machine speed. This could arguably
be true for the wire sizes considered in the investigation. The phenom-
enon could be due to a higher percentage reduction in area from 2.2 mm
to 0.53 and 0.66 mm which represents more than 80% cold work [16].
Fig. 7 shows the comparison of maximum elongation achieved
without tarnishing of wires between Cu-ETP and Cu-OF. It appears that
Cu-OF can achieve a higher maximum percentage elongation without
oxidation of the wire compared to Cu-ETP for the different machine
speeds. This is attributed to Cu-OF having fewer impurities, such as
Cu2O, than are present in Cu-ETP [17,18].
4. Conclusions
The study assessed the annealer curve characteristics of medium wire
drawing machines for raw Cu-ETP and Cu-OF. The investigation indicates
that no adjustment of operational parameters of wire drawing machines
is required to achieve a minimum stipulated elongation when input wire
Appendix A. CheMICAL compositions of tested materials
Chemical composition of 8 mm Cu-OF and 7.9 mm Cu-ETP rods used in the experiments.
CHEMICAL ANALYSIS
Sample Number 8 mm Cu-OF
Cu BAL
Sn (ppm) <5
Pb <5
Fe 5 11
Al <5
Mn <5
Ni <5
4
Results in Materials 8 (2020) 100146
Fig. 7. Maximum elongation at maximum voltage before tarnishing of wire.
is switched from Cu-ETP to Cu-OF for the considered size reduction from
2.2 mm diameter to 0.66 mm and 0.53 mm. For this level of percentage
elongation, the required applied voltage in the annealer increases with
the wire drawing machine speed. The study also corroborated that raw
Cu-OF is superior to Cu-ETP in terms of the maximum achievable elon-
gation at the highest possible voltage in the annealer without wire
oxidation as it contains less than 10 ppm of oxygen as can be seen in the
chemical analysis provided in Annexure A. The investigation contributes
to a better understanding of copper annealing post wire drawing. It also
provides a guideline for the optimum operation of medium wire drawing
machines for the different types of copper used in the cable industry.
Future investigation will include the measurement of electrical wire
properties and could possibly explore different wire sizes.
Credit author statement
C Banganayi: Conceptualization, Methodology, Validation, Formal
analysis, Investigation, Writing - original draft, Formal analysis. K
Nyembwe.: Writing - original draft, Writing - review & editing prepara-
tion, Formal analysis, K Mageza: Investigation, Data curation
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
We would like to thank Aberdare Cables PTY Ltd for providing the
material, machines, and testing equipment for the experiment. We also
would like to thank Mr Henni Scholtz and Mr Zak Du Preez for their
technical input and leadership guidance during the investigations.
7.9 mm Cu-ETP
BAL
<5
<5
7
<5
<5
(continued on next column)
C. Banganayi et al. Results in Materials 8 (2020) 100146

(continued )
CHEMICAL ANALYSIS

Sample Number 8 mm Cu-OF 7.9 mm Cu-ETP

Sb <5 <5
As <5 <5
P <5 <5
Si <5 <5
Cd <5 <5
Zn <5 <5
Bi <5 <5
Te <5 <5
Ag 6 15
Se <5 <5
S 10 8
Mg <5 <5
Cr <5 <5
Co <5 <5
Oxygen <10 <200
Be <5 <5
Zr <5 <5
COND % IACS n/a n/a
Recovery n/a n/a

References [9] Q.D. Grearson, A Guide to Copper Wire Drawing, 1980, pp. 59–62.
[10] S.H. Avner, “Introduction to Physical Metallurgy.” International Student Edition,
Publisher McGraw-Hill, 1988.
[1] M. Nairn, Graphite crucible furnace technology as a basis of highest quality copper
[11] S. Cochrane, G. Yeoman, Cu-OF or Cu-ETP?: Some General Comparisons of Small
redraw rod, Wire J. Int. 36 (2003) 68–73.
Diameter Copper Wire Rod Products, 2006, pp. 1–17.
[2] Copper Development Association Inc, Microstructure of copper alloys [Online]
[12] D. Clark, K. Costin, D. Hutton, R. McConeghy, A. Scheidemantle, and R. Smellie;
Available at: https://www.copper.org/resources/properties/microstructure
SSC spool piece design , Suppercollider 3 pp 202-210 Edited by John Nonte.
/coppers.html, , 2017. (Accessed 7 September 2017).
[13] R. Kratz, P. Wyder, Principles of Pulsed Magnetic Design Engineering Materials,
[3] J. C Hogg, Copper Wire Production, ” Publisher International Wire & Machinery
Springer-Verlag Berlin Heidelberg, 2002, pp. 27–28.
Association, 1990, pp. 3–11.
[14] South African National Standards, Materials of insulated electric cables and flexible
[4] W H Bowyer, The effects of impurities on the properties of OFP copper specified for
cords: Part1 Conductors 2 (2008) 3–21.
the copper iron canister SKI 1999, Report 99:44 pg 1-16.
[15] BS EN 13602, Copper and Copper Alloys. Drawn, Round Copper Wire for the
[5] National Bureau of Standards Circular No, [5] Publication No. 28 of the
Manufacture of Electrical Conductors, 2013, pp. 1–26.
International Electrotechnical Commission, "International Standard of Resistance
[16] Troostwijk Auctions, 1988 Niehoff MM315 twin copper wire drawing line [Online]
for Copper (March 1914) Is Included in Appendix V of This Document, 1914, p. 31.
Available at: https://www.troostwijkauctions.com/uk/twin-copper-wire-drawing
[6] S. Smart Jr., The effects of impurities in copper, in: A. Butts (Ed.), Copper,
-line/03-24774-33798-4436305/, , 2017. (Accessed 7 September 2017).
Rheinhold Publishing Corp NY, 1954, p. 410.
[17] H. Pops, J Holloman, “Effects of oxygen concentration on the recrystallization
[7] S. Jakania, M.H. Mathona, Ph Gerbera, b M. Benyoucefb, C.H. de Noviona,
behaviour of copper wire, Wire J. Int. (1994).
T. Baudin, Influence of oxygen content on the static recrystallization of ETP copper,
[18] Hitachi Cables, Oxygen free copper (OFC) CAT.NO CE 100A 1-10 wusatowska AM
Mater. Sci. Forum 467–470 (2004) 471–476.
S., miura H., T sakai (2002), nucleation and microtexture development under
[8] F.K. Huybrechts, P. Schamp, The effects of impurities on the recrystallization
dynamic recrystallization of copper, Mater. Sci. Eng. A323 (2006) 177–186.
behaviour of tough pitch hot rolled copper rod, Wire J. Int. (1996) 40–49.