Journal of Energy Storage 8 (2016) 69–77

Contents lists available at ScienceDirect

Journal of Energy Storage

journal homepage: www.elsevier.com/locate/est

Detachable electrical connection of battery cells by press contacts

Martin J. Branda,* , Philipp Berga , Elisabeth I. Kolpa , Tobias Bachb , Philipp Schmidtc,
Andreas Jossena
a
Institute for Electrical Energy Storage Technology, Technical University of Munich, Arcisstr. 21, 80333 Munich, Germany
b
Fraunhofer Institute for Silicate Research, Neunerplatz 2, 97082 Wuerzburg, Germany
c
Institute for Machine Tools and Industrial Management, Technical University of Munich, Boltzmannstr. 15, 85748 Garching, Germany

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

Article history:
Received 20 June 2016 In battery applications, every electrical connection of battery cells is important because it influences
Accepted 25 September 2016 functionality, efficiency, and safety. Increased contact resistances generate more heat at the affected
Available online 5 October 2016 terminals and if contact resistances of parallel-connected battery cells differ the current divides
unequally.
Keywords: Amongst different techniques, detachable connections are easy to handle and have advantages for
Press contact service, repair and recycling. Usually, detachable electrical connections are based on the functional
Battery assembly principle of press contacts. So, press contacts are frequently used for consumer batteries in electronic
Electrical contact resistance
devices, for pouch battery cells in scientific tests, and for large battery cells with screw connections.
Lithium-ion battery cell
This paper focusses on the electrical contact resistance of press contacts for battery cells. A novel
Equivalent electric circuit
measuring method to assess the contact resistance itself is presented. Common terminal and connector
metals for battery cells and the dependence of their contact resistances on the contact pressure, surface
roughness, and contact area are investigated and compared with welding techniques. Only if these
dependencies are considered, contact resistances as low as for welded connections are achieved.
Furthermore, due to its hard passivating Al203 surface film, aluminum contacts need a certain surface
roughness. The obtained results can serve for modeling and development of battery assemblies.
ã 2016 Elsevier Ltd. All rights reserved.

1. Introduction There are many options to electrically connect battery cells,

Large battery assemblies are of particular interest, for example
for the progressing electrification of mobility or the storage of
intermittent renewable energy. But also for portable electronic
devices, for example laptops, the battery is one of the key
components. In all battery applications, every single electrical
connection of a battery cell is of importance because it influences
the functionality, efficiency, and safety. An increased electrical
contact resistance generates more heat at the affected terminal of a
battery cell [1,2]. Additionally, for battery cells connected in
parallel, the total current is divided unequally if contact resistances
differ significantly. These uneven loads may lead to inhomoge-
neous cell degradations, as discussed in [3,4]. What is more,
defective cell connections are prone to fail suddenly if imposed to
mechanical stress.
* Corresponding author.
E-mail address: martin.brand@tum.de (M.J. Brand).
such as welding, soldering, press contacts, and screwed joints.
Because soldered connections can only be detached with signifi-
cant effort, the most common detachable connection techniques
are press contacts and screwed joints. The functional principal of
the resulting electrical contact resistances for both techniques is
the same and is based on the metallic contacts that are formed
when two metal surfaces are pressed together.
Detachable connections have significant advantages when it
comes to service, repair and recycling. Another advantage is that
this connection technique does not need expensive machines and
therefore can be handled by every specialist workshop or in certain
cases also by the customer.
An illustrative selection of battery cells with terminals for press
contacts is shown in Fig. 1. A common usage of press contacts is for
consumer batteries in electronic devices, for example cell-phones
or cameras. For this electronic applications, an example with brass
springs is shown in Fig. 1 bottom left. To test battery cells of the
pouch format, a press contact fixture can be used as shown in Fig. 1
bottom right. Such fixtures are quite common for scientific series of
measurements. Furthermore, some large battery cells with high

http://dx.doi.org/10.1016/j.est.2016.09.011
2352-152X/ã 2016 Elsevier Ltd. All rights reserved.
70 M.J. Brand et al. / Journal of Energy Storage 8 (2016) 69–77
Fig. 1. Exemplary selection of battery cells with terminals for press contacts.
power capability exhibit screw connections. Especially for lead acid
battery cells but also for some large lithium-ion battery cells, screw
or clamp connections are quite common.
The most important parameter to evaluate an electrical
connection is the electrical contact resistance. Although press
contacts are frequently used to electrically connect battery cells,
few scientific publications on contact resistances of press
connections exist. As one of the few publications, Taheri et al.
[5] investigated the electrical contact resistance losses in lithium-
ion battery assemblies with screwed joints. Within that publica-
tion, the authors focused on 20 Ah lithium-ion pouch battery cells
connected by copper bars with and without an additional
interfacial electrically conductive material [5]. Taheri et al. showed
that the pressure plays a key-role to reduce electrical contact losses
and that additional conductive materials can help to reduce
contact losses at lower contact pressures [5].
The present paper’s focus is placed on the electrical contact
resistance of press contacts at battery cells in general and lithium-
ion battery cells in particular. The electrical contact resistance
occurs as a result of the joint and not in the bulk material.
Therefore, it is a criterion which may be transferred to any size of
joint partners. Within this paper, a novel measuring and
calculation method is applied to assess the electrical contact
resistance itself.
For the series of measurements presented in this paper, the
most common terminal and connector metals for battery cells are
investigated. Because of their stability, the materials of primary
current collectors within lithium-ion battery cells are aluminum at
the cathode and copper at the anode [6]. By contrast, at the
Fig. 2. Schematically (a) surface roughness including passivating surface films, (b) deformation when contact pressure pC is applied including current paths, (c) load bearing
areas AC,lb and a-spots AC,a.
terminals of battery cells, joining of other metals can be required,
for example a brass cell terminal with a nickel plated steel
conductor [7,8]. For press contacts, mostly aluminum, copper,
brass, and nickel-plated steel metals are used [7,8].
In the following sections, fundamentals of press contacts are
presented first. Subsequently, the novel measurement and
calculation method for the electrical contact resistance as well
as the pneumatic test bench to apply defined pressure on the
tested metal specimens are presented. Based on these fundamen-
tals the results of the series of measurements are explained. For the
series of measurements, the types of metals, the contact pressure,
the surface roughness, and the contact area were varied
systematically. The measured electrical contact resistances can
serve as an important input for modeling and simulating battery
assemblies.
At the end of this paper, the resulting electrical contact
resistances are compared to the contact resistances obtained for
spot, ultrasonic, and laser beam welding. Brass (CuZn37) test
samples are used for the quantitative comparison of these four
connection techniques, as this metal can be welded with all three
welding techniques. In the conclusion, the suitability of press
contacts for battery cells and its settings to achieve low electrical
contact resistances are evaluated.
2. Fundamentals on the electrical connection of press contacts
In this section, the fundamentals of stationary electrical
contacts between two solid metals are presented. Therefore, the
 M.J. Brand et al. / Journal of Energy Storage 8 (2016) 69–77
mechanical parameters of a contact surface that influence the
electrical contact resistance are explained.
All surfaces of solid metals have a certain roughness on the
microscopic scale [9]. Consequently, if two metals are pressed
together, mechanical contact occurs at spots with asperities as
illustrated in Fig. 2.
The load bearing contact area AC,lb with actual mechanical
contact is only a fraction of the apparent contact interface with the
area AC,app [9–11]. For metals, deformation occurs and increases the
load bearing areas depending on the contact pressure pC [10,12].
Because of insulating tarnish films on the metal surfaces, the
electrically conductive area AC,a is smaller than the load bearing
area AC,lb [9,10]. It is widely accepted that these electrically
conductive areas are named a-spots [10,11].
Fig. 2(a)–(c) schematically show the roughness of the surface of
both metal contact partners including the passivating surface
films, the deformation when a contact pressure pC is applied
including the possible current paths, and the load bearing areas AC,
lb and a-spots AC,a.
The contact resistance consists of the following two compo-
nents: the constriction resistance and the film resistance [10,13].
The film resistance is caused by the reduced conductivity of the
passivating films on the load bearing surface areas that are not a-
spots [10,13]. The cause for the constriction resistance is illustrated
by Fig. 2(b). Because metallic contact occurs at discrete a-spots and
not throughout the whole apparent contact interface, the electrical
current lines bundle together and therefore experience an
additional resistance at the contact area [9–11]. These are the
two most important contributions that cause the electrical contact
resistance, whereas discontinuities in the crystal latter can be
neglected [10].
From the explained fundamentals, it is clear that the electrical
contact resistance of two metals pressed together depends on the
Fig. 3. Surface roughness for brass specimens prepared with abrasive paper with (a) grain size 180 and (b) grain size 600.
71
contact pressure, the metals’ electrical conductivities, and the
surface micro-structures. The microscopic surface structure can be
described by its roughness [14,15]. In order to control the surface
roughness within this work, the surfaces of the investigated metal
specimens were prepared with abrasive paper and abrasive fleece.
For each grain size and metal, the prepared surfaces were analyzed
with the 3D laser scanning microscope VK-X200 by the Keyence
Corp. In general, using a laser microscope is, on the one hand,
accurate and, on the other hand, does not influence the surface
because of contactless measurement [15].
Two exemplary measurement results for two brass specimens
prepared with abrasive paper with grain size 180 and 600 are
shown in Fig. 3(a) and (b). The zero line of the surface depth is the
average value of the measured depth. It can be seen, that the
measured asperities for the specimens prepared with abrasive
paper with grain size 180 are higher than the asperities measured
for grain size 600.
This is also reflected in the average roughness values Ra defined
by the ISO 4287 standard [16]. The average roughness value Ra can
be calculated by arithmetically averaging the measured surface
depths dsurf along the scanned line x (see Eq. (1)).
xZ
max
Ra ¼ 1=xmax  jdsurf ðxÞjdx ð1Þ
0
The calculated average roughness values for the microscopic
surface profiles in Fig. 3(a) and (b) are 0.68 mm for the abrasive
paper with 180 grain size and 0.42 mm for grain size 600.
Furthermore, the probability densities and material ratios show
that for both profiles the measured depths are normally
distributed. Similar to the average roughness values, the variation
of the surface depth is higher for abrasive paper with grain size 180.
72 M.J. Brand et al. / Journal of Energy Storage 8 (2016) 69–77

Fig. 4. Schematic view of test sample consisting of two connected metal specimens
overlapping by 15 mm  15 mm.

3. Method to measure electrical contact resistances of press

connections

To obtain transferable results, the electrical contact resistance

itself, which only occurs as a result of the joint and not in the bulk
material of the conductors, has to be determined. In [17], the
authors presented a calculation and measurement method to
determine the electrical contact resistance between a pair of
defined metal specimens, as shown in Fig. 4. For this method, each
test sample consists of two specimens overlapping by a defined
contact area, which is by default 15 mm  15 mm. Three spots for
voltage measurements are marked on every specimen named A1,
A2, A3 and B1, B2, B3.
For every test sample, every resistance RAi,Bj (with i,j = 1 j = 1,2,3)
from one of the points A1, A2, or A3 to one of the points B1, B2, or B3 is
measured. Since the dimensions and the material electrical Fig. 5. Pneumatic measurement set-up to investigate the electrical contact
resistivity are exactly known, the resistances RAi,A0 from any resistance of press contacts.

measuring point A1, A2, or A3 to A0 can be calculated by Eq. (2). The

same applies for the specimen B.
RAi;A0 ¼ rA  lAi;A0 =ðh  bÞ
Within this equation, rA is the material electrical resistivity of
the specimen A and rB the resistivity of specimen B. The other
parameters refer to the dimensions of the specimen; h is the
height, b the width, and lAi,AO is the distance from point A0 to Ai.
The ideal resistance of the test sample consisting of the two
specimens is obtained when the electrical contact resistance is
zero. Consequently, the ideal overlap resistance RA0,B0,id between A0
and B0 can be calculated by Eq. (3).
    ESNU-32-50-P by 32.4 N [18]. Additionally, the friction reduces the
RA0;B0;id ¼ rA lA0;B0 =ðh  bÞ jj rB  lA0;B0 =ðh  bÞ ð3Þ
The difference between the measured resistance RA0,B0,meas
and the ideal resistance RA0,B0,id is defined as the electrical contact
resistance RC.
RC ¼ RA0;B0;meas  RA0;B0;id
precision manometer is installed with a scale value of 0.02 bar, a
maximum value of 4 bar, and an accuracy class of 0.6. Once the
ð2Þ pressure is set to its desired value, the single measurements can be
activated by the on/off switch.
The compressed air is directed into the pneumatic cylinder
ESNU-32-50-P by the Festo AG & Co. KG that has a return spring
integrated. In the measurement set-up shown in Fig. 5, the length
of stroke of the piston is 19 mm. To calculate the piston force Fcyl
that is applied to the specimens, the spring force Fspring,19mm for a
distance of 19 mm has to be subtracted. According to the data
sheet, the spring reduces the force of the pneumatic cylinder
total force of the piston by 8.85% and therefore a friction coefficient
cfric = 0.915 is introduced [18]. The complete calculation of the
piston force is given by Eq. (5).
 
F cyl ¼ F ideal  F spring;19mm  cf ric 
ð4Þ ¼ pcyl  p  ð32mm=2Þ2  32:4N  0:915 ð5Þ

These equations are applicable for specimens with dimensions
as shown in Fig. 4 and especially for specimens of similar metal
types.
The resistances were measured by the four-point probes
method with the BT3562 by the Hioko E.E. Corp., which provides
a 1 kHz AC current Imeas = 100 mA to the longitudinal ends of the
test samples and measures the voltage drop. To narrow the
confidence interval, every combination RAi,Bj was measured five
times and averaged. The graphs presented in the following
chapters are based on series of measurements in which, for every
measuring point, at least five test samples were tested and
averaged subsequently.
The measurement set-up for press contacts is based on a
pneumatic cylinder, the regulation of the compressed air, and a
fixture to hold the two specimens (see Fig. 5). Before each series of
tests, the pressure is adjusted to the desired value by the regulating
valve. Since the pressure adjustment is done manually, a high
The cylinder force Fcyl is homogeneously applied to the test
sample by a plunger. The plunger covers exactly the overlapping
contact area of the specimens and can be exchanged when the
contact area is varied between 10  15 mm2, 15  15 mm2,
20  15 mm2, 25  15 mm2, and 30  15 mm2.
With that measurement set-up, the contact area and the
pressure can be varied, but once the desired values are set the
measurements are highly reproducible.
4. Measurement matrix to investigate dependencies of press
contacts
In the previous sections, the measurement setup and the
physical fundamentals of electrical contact resistances of press
contacts were presented. On this basis, a measurement matrix can
be set up which delivers fundamental statements on the
dependencies of contact resistances.
 M.J. Brand et al. / Journal of Energy Storage 8 (2016) 69–77
The following aspects influence the electrical contact resistance
of press contacts the most: contact pressure, conductivity of the
metal, contact area, surface condition of the metal, and tempera-
ture. The first four aspects are listed in the measurement matrix
and investigated in the following sections. All experiments were
carried out at an ambient temperature of approximately 22  C.
For each series of measurements only one parameter was varied
while the other parameters were kept constant. Thereby, the
influence of each single parameter on the contact resistance can be
quantified.
For the series of measurements, the following metals which are
used most frequently for press connections of battery cells or
battery assemblies are chosen:
- Copper: For battery applications mostly electrolytic tough pitch
copper (CuETP) is used because of its low specific resistance of
rCu = 17.24 mV mm [19]. For the presented series of measure-
ments CuETP with material condition R360 according to the
EN1652 standard was used.
- Aluminum: Within this work AW-1050A aluminum, which
consist of 99.5% of pure aluminum at least, was used. Because of
its low specific resistance of rAl = 27.78 mV mm, this pure
aluminum is most suitable for electrical applications, especially
to distribute energy [19]. The condition of the aluminum chosen
for the experiments is H24 according to the EN485-2 standard.
- Brass: For some battery cells, the terminals are made of brass.
This is also illustrated by the cell phone battery and the large
lead acid battery cell shown in Fig. 1. Brass is an alloy of copper
and zinc, wherein zinc causes a reduction of electrical
conductivity but an increase of the metal’s rigidity and tensile
strength [20]. In the following series of measurements, the most
common CuZn37 alloy with 37% zinc and a specific resistance of
rCuZn37 = 66.67 mV mm is used. For this metal the material
condition is R360 according to the EN1652 standard.
- Nickel-plated steel: Nickel-plated steel is often used for
terminals and connectors of lithium-ion battery cells due to
its corrosion resistance, high mechanical stability, and good
weldability [7,8]. The metal used within this work consists of
steel covered by a 2 mm thick nickel layer with rHil = 126.48
mV mm [21].
As already discussed in Section 3, all specimens had the same
dimensions as shown in Fig. 4. To investigate the influence of the
contact pressure on the electrical contact resistance, the contact
pressure was varied by adapting the cylinder pressure pcyl (see
Table 1
Measurement matrix including varied parameters to investigate dependencies of press contacts.
Dependency on Investigated metal
Contact pressure and material Copper
Aluminum
Brass
Nickel-plated steel
Surface roughness and material Aluminum
Brass
Contact area Brass
73
Eq. (5)). The variation was done in the following seven steps:
0.31 N mm2, 0.38 N mm2, 0.48 N mm2, 0.63 N mm2, 0.98 N
mm2, 1.41 N mm2, and 2.04 N mm2.
The dependencies of the contact resistance on the surface
roughness and surface films is investigated by preparing the
surfaces with abrasive papers with grain sizes varying between 60,
100, 180, 320, 600, and abrasive fleece with grain size 1000. For the
tests with varying surface roughness, only brass and aluminum
were investigated. Copper has similar surface films as brass.
Nickel-plated metals are not suitable for this investigation,
because preparation with abrasive paper would destroy the nickel
layer. For aluminum and brass and the used abrasive papers,
average roughness values between 0.35 mm and 1.70 mm are
obtained. For all other measurements, test samples were prepared
with abrasive fleece by default. Nickel-plated steel was only
cleaned with a woolen sheet to keep the nickel layer undestroyed.
The contact area was varied from 150 mm2 to 450 mm2 in six
steps by varying the overlap of the test specimens and by changing
the corresponding plunger. The pressure was kept constant, for
example by applying twice the cylinder force Fcyl in case of
doubling the contact area.
A detailed overview on the three series of measurements and
the varied parameters is given by Tab. 1.
For the following series of measurements, all investigated test
samples consist of two specimen of the same metal.
5. Influence of contact pressure on electrical contact resistance
In this section the influence of the contact pressure on the
electrical contact resistance is investigated for copper, aluminum,
brass, and nickel-plated steel test samples. Fig. 6(a) depicts the
measured contact resistances over the variation of the contact
pressure for all four investigated metals.
In the graph of Fig. 6(a), the averaged values and the 90%
confidence intervals are shown. This confidence interval defines
the range in which the average value is located with a stochastic
confidence of 90% [22]. To narrow the confidence interval, the
contact resistance of every test sample was measured five times.
Each of these five measurements, is based on nine single resistance
measurements (see Section 3). Furthermore, at least five different
samples were tested and averaged for every measuring point. So, at
least 225 single resistance measurements were done to obtain one
single measuring point marked by a cross in Fig. 6.
Because of the increased scattering of measured contact
resistances at low contact pressures, the number of test samples
Contact pressure Av. roughness Contact area
0.31 N mm2 0.42 mm 15.0  15.0 mm2
0.38 N mm2 (except nickel-plated steel)
0.48 N mm2
0.63 N mm2
0.98 N mm2
1.41 N mm2
2.04 N mm2
0.63 N mm2 0.42 mm 15.0  15.0 mm2
0.52 mm
0.83 mm
0.95 mm
1.23 mm
1.67 mm
0.63 N mm2 0.42 mm 10.0  15.0 mm2
15.0  15.0 mm2
20.0  15.0 mm2
25.0  15.0 mm2
30.0  15.0 mm2
74 M.J. Brand et al. / Journal of Energy Storage 8 (2016) 69–77
Fig. 6. Dependency of (a) electrical contact resistance including the 90% confidence
intervals and (b) relative electrical contact resistance on contact pressure for
different metals.
was increased from five to ten for the contact pressures of
0.31 N mm2, 0.38 N mm2, and 0.48 N mm2. By that, the confi-
dence interval is narrowed at low contact pressures and the
information in the graph is more reliable. The ambient tempera-
ture, surface roughness and contact area were kept constant
(compare Tab. 1).
For the sake of better comparison Fig. 6(b) shows the mean
values of Fig. 6(a) normalized to the electrical contact resistance at
a contact pressure of 0.31 N mm2. For all tested metals, the
electrical contact resistances decrease with increasing contact
pressure. With a higher contact pressure, the materials are tighter
pressed together causing the a-spot areas to grow and therefore
providing broader current paths with lower constriction resistance
[5]. For the overall constriction resistance, the single constriction
resistances at each a-spot can be regarded as being connected in
parallel [14,13]. Similar to the findings of Taheri et al. [5], the drop
of the electrical contact resistance is steep at lower contact
pressures and decreases only little for contact pressures above
1 N mm2. This is because, a linear increase of the number of a-
spots results in a nonlinear drop of the constriction resistance,
which is proportional to the contact pressure by pCk with k being a
constant of proportionality [14,13].
As shown in Fig. 6(b), the contact pressure can reduce the
electrical contact resistance by about 50%. Furthermore, only at
high contact pressures a good reproducibility of the contact
resistances can be achieved. At a contact pressure of 2.04 N mm2
scattering is low and contact resistances do not vary much with the
contact pressure anymore. Here, except for aluminum, the contact
resistances are sorted in the order of the materials conductivity:
Copper with a contact resistance of 0.018 mV, brass with
0.110 mV, and nickel-plated steel with 0.186 mV. By contrast,
aluminum has a higher conductivity as brass but the measure-
ments show a far higher electrical contact resistance of 1.777 mV
for a contact pressure of 2.04 N mm2. This is because of the
insulating Al2O3 surface film. In oxygen-rich ambiance, the surface
film forms almost instantly on pure aluminum surfaces [19,13].
Furthermore, Al2O3 is a very hard ceramic with a Vickers hardness
of HAl2O3 = 2720 kg mm2 [23]. Aluminum in general is a much
softer metal [23,13]. According to the product sheet, the chosen
H24 aluminum AW1050A has a Vickers hardness of HAl = 33 kg
mm2. Consequently, it is difficult to form significant a-spot areas
and therefore the electrical contact resistance is much higher for
aluminum than for the other investigated metals, where the oxide
layer can be pierced easily by asperities. Also the confidence
intervals are larger for aluminum as for the other metals, indicating
lower reproducibility and more variation in the measurement
results.
For the series of measurements in this section, the surfaces
were treated with abrasive fleece with grain size 1000. For
aluminum this means that many load bearing areas exist where the
Al2O3 film is in between the surfaces and prevents the formation of
a-spots. Consequently, it can be supposed that for rougher surfaces
the more distinct asperities could pierce the Al2O3 film and
therefore reduce the electrical contact resistance. This supposition
is checked in the following section.
6. Influence of surface roughness on electrical contact
resistance
The measurements of the previous section showed for
aluminum an electrical contact resistance that is far higher than
for all other tested metals. Because abrasive fleece with grain size
1000 was used for the previous experiment, it is supposed that the
high contact resistance is caused by the lack of distinct asperities
that pierce through the hard Al2O3 surface film. In this section, the
surface roughness of the aluminum test samples is varied from an
average roughness value Ra of 0.42 mm to maximum 1.67 mm. For
this experiment, the contact pressure is kept at pC = 0.63 N mm2
and the apparent contact interface is always AC,app = 15  15 mm2
(compare Tab. 1). To cross-check whether a dependency on the
surface roughness also exists for other metals with softer
passivating surface films, the series of measurements was also
done with brass test samples.
Fig. 7 depicts the measured electrical contact resistances over
the surface roughness. As supposed, the contact resistance of the
aluminum samples shows a strong dependency on the surface
roughness. For an average roughness value of Ra = 1.67 mm the
electrical contact resistance is 0.072 mV and therefore lower than
the contact resistance of brass. Finally, this value perfectly fits in
the order of the tested metals when sorted according to their
electrical conductivity. For aluminum, also the reproducibility
increases and the variation of the measured values decreases with
a higher surface roughness. For the two lowest average roughness
values, 0.42 mm and 0.52 mm, ten instead of five test samples were
tested in order to narrow the confidence interval.
Fig. 7. Dependency of electrical contact resistance on surface roughness for
aluminum and brass including the 90% confidence intervals.
 M.J. Brand et al. / Journal of Energy Storage 8 (2016) 69–77
These findings can be explained by two reasons. Firstly, with
stronger abrasion the aluminum oxide film is probably better
removed. Secondly and more importantly, an increased surface
roughness obtains more pronounced micro peaks that can pierce
through the hard passivating surface film. This experiment
strengthens the arguments from Section 5 that the high electrical
contact resistance for aluminum is caused by its surface film.
Brass as well as copper also show a passivating surface film but
this film is softer and can be bridged with enough contact pressure
(see Fig. 6). Therefore, the electrical contact resistance of the brass
samples hardly depends on the surface roughness. Additionally,
the variations of the measurements are relatively low and not
significantly influenced by the surface roughness.
In another experiment, the brass and copper surfaces were not
cleaned with abrasive paper directly before the measurements but
left resting for 5 days after surface preparation. Therefore,
passivating surface films formed and electrical contact resistances
of about 0.3 mV for brass and 0.15 mV for copper were measured.
Additionally, the variations and confidence intervals increased
dramatically.
To conclude, on aluminum surfaces an insulating Al2O3 surface
film forms almost instantly in oxygen rich ambiance. This film
exhibits a much higher hardness compared to aluminum itself.
Therefore, aluminum surfaces have to be rather rough to exhibit
distinct asperities that can pierce through the Al2O3 film. So, if
aluminum materials should be connected with low electrical
contact resistance and high reproducibility, attention has to be
paid to the surface treatment.
Also for copper and brass surfaces, passivating films are a
problem for press contacts. This can be mitigated by cleaning the
surfaces before connecting them. The passivating surface films are
not as hard as passivating films on aluminum and build up far
Fig. 8. Comparison of brass samples connected by press contacts with spot, ultrasonic, and laser beam welding with regard to contact area according to [17]; Because of
different characteristics of each connection technique separate abscissas are necessary.
75
slower. Therefore, the average roughness values do not signifi-
cantly influence the electrical contact resistance.
7. Electrical contact resistance of press contacts compared to
welded connections
In this section, the electrical contact resistances of brass
samples connected by press connection are compared to the
contact resistances obtained for spot, ultrasonic, and laser beam
welding. For the comparison of the four connection techniques,
brass is chosen because this metal can also be welded with the
three named welding techniques. The results of the three different
welding techniques have been discussed in detail in [17]. For each
measurement point and each welding technique the optimal
welding parameters were used while the overlap of the brass
specimens was kept at 15  15 mm2. By contrast, for press contacts
the overlap is varied in five steps from 10  15 mm2 to 30  15 mm2.
Fig. 8 illustrates the comparison of the four connection
techniques. Separate abscissas are necessary because the contact
areas and weld geometries are not quantitatively comparable. The
areas that have a direct metal to metal contact depend on the
connection technique and the chosen connection parameters, as
for example the weld seam. On each abscissa in Fig. 8, the overlap
of the brass specimens as well as the contact areas or weld
geometries are illustrated. For these illustrations, the samples
longitudinal side as well as the current flow points from the
bottom to the top.
According to Schmidt et al. [24], the ideal geometry of the
connection would be a seam on both edges of the overlap with no
electrical contact resistance. This ideal seam can be nearly realized
with laser beam welding because this technique allows almost
arbitrary weld seams. For the measurements presented in Fig. 8,
76 M.J. Brand et al. / Journal of Energy Storage 8 (2016) 69–77
the keyhole welded seams are four small lines with each 2 mm
length and two 13 mm lines with and without beam oscillation.
The weld seams were placed as close as possible to the edges of the
specimens’ overlap.
For spot welding, the optimal seam can be approximated by
placing single weld spots on these ideal lines. The numbers of weld
spots are varied from four to eight to twelve.
For ultrasonic welding, four different sonotrodes and their
corresponding anvils were used. They vary in its areas between
5  5 mm2, 6.5  10.5 mm2, 6.5  12.5 mm2, and 13  13 mm2.
For the press connections, the overlap of the brass specimens is
varied between 10  15 mm2, 15  15 mm2, 20  15 mm2,
25  15 mm2, and 30  15 mm2, while the contact pressure is kept
at 0.63 N mm2 and the surface roughness is kept at 0.42 mm. To
keep the contact pressure constant, the force of the pneumatic
cylinder is adapted to the contact area of the plunger.
The electrical contact resistances are comparable at the
measurement points in Fig. 8 where the contact area of the press
contact is 15  15 mm2. Only at this points, the optimal connection
parameters for the 15  15 mm2 overlap of the specimens are
applied for each connection technique.
With 0.130 mV, the contact resistance of laser beam welding is
the lowest. This is because, on the one hand, the metals of both
specimens are melted and mixed together in the weld seam and, on
the other hand, nearly the optimal weld geometry can be realized.
The contact resistance for press connections is relatively low at
0.154 mV because the contact pressure is applied to the complete
overlap area.
Although both specimens’ metals are melted and intermixed
when spot welded, only twelve weld spots can be realized.
Furthermore, micrographs of some spot welded brass test samples
showed voids within the weld nuggets that are filled with zinc
vapor [17]. The melting interval of CuZn37 is between 902  C and
920  C, whereas the boiling point of zinc is at 907  C. The highest
welding temperatures occur in the center of the weld nugget,
where the boiling temperature of zinc was potentially reached.
Therefore, the contact resistance is 0.167 mV and thereby higher as
with laser beam welding.
If the optimal sonotrode area for the 15  15 mm2 overlap is
chosen, ultrasonic welding reaches a contact resistance of
0.169 mV, which is relatively high. This welding technique does
not melt the specimens’ metals but press them together while
applying longitudinal vibrations with a frequency of 20 kHz and an
amplitude of about 29 mm. To avoid cracks due to resonance in the
brass test samples, the maximum welding time has to be limited.
Micrographs of the ultrasonic welded samples reveal only small
spots of actual connection at the surface [17].
The contact area for press contacts is increased stepwise from
10  15 mm2 to 30  15 mm2 to show the approximately linear
decrease of the contact resistance. The contact area can only be
varied by the overlap of the two specimens.
8. Conclusion: characteristics of press contacts and suitability
for batteries
Press connections are frequently used to electrically connect
battery cells. A common usage of press contacts can be found
within electronic devices with exchangeable batteries. Also large
battery cells, especially lead acid batteries, exhibit screw or clamp
connections. For tests and scientific investigations with pouch
battery cells, it is quite common to use a press contact fixture.
With regard to service, repair and recycling, detachable press
connections have significant advantages. Furthermore, this con-
nection technique does not need expensive machines and is easy to
handle. But as the investigations of this paper clearly show, only
those who know about the characteristics and dependencies of
press contacts can guarantee the reliability and low electrical
contact resistance of press connections.
To evaluate the press contacts, this paper focuses on the
electrical contact resistance. Therefore, a novel measuring and
calculation method was applied to assess the electrical contact
resistance itself. Furthermore, a highly accurate pneumatic test
bench with an exchangeable plunger was developed and used for
the series of measurements.
The following most common metals for press connections at
battery cells were investigated: copper, aluminum, brass, and
nickel-plated steel. For all metals, the contact resistances depend
on the contact pressure. If the contact pressure is far below
1 N mm2, the contact resistance is increased and also the
variations of the measurements are higher due to current
constriction. For higher contact pressures, the contact resistance
does not vary much and depends mostly on the electrical
conductivity of the connected metals itself. Within battery
applications, some effort has to be put into the design of the
contacts to guarantee a sufficient contact pressure. So, probably
additional components, for example screws, are necessary that
increase the total weight, volume and complexity of the battery
system.
For all tested metals, passivating surface films have a significant
influence on the contact resistances. Before each measurement,
the copper, aluminum, and brass specimens were cleaned with
abrasive fleece with grain size 1000. Nickel-plated steel was only
cleaned with a woolen sheet to keep the nickel layer undestroyed.
By contrast to the other tested metals, an insulating Al2O3 film
forms almost instantly on aluminum surfaces in oxygen rich
ambience. Because the Al2O3 film is much harder than aluminum
itself, it is more difficult for the micro-peaks to pierce through the
surface film and form a-spots. That is why the contact resistance of
aluminum samples depends on the surface roughness. The series
of measurements shows, that an increased surface roughness with
more distinct asperities reduces the contact resistance drastically.
By contrast, for the investigated brass samples, no clear
dependency on the surface roughness can be seen. Especially for
battery applications with aluminum press contacts, the influence
of passivating surface films and the surface roughness have to be
considered.
Finally, the electrical contact resistances of press contacts were
compared to the contact resistances obtained with spot, ultrasonic,
and laser beam welding. With press connections the complete
available contact area can be used for electrical connection.
Another advantage is that principally every metal can be connected
with every other metal regardless of their melting points.
Furthermore, by contrast to press contacts, welding techniques
cause a certain heat-input into the battery cells during the
connection process.
The measured electrical contact resistances can serve as
important input when modeling and simulating battery cells
and assemblies. Furthermore, the explained characteristics and
dependencies of press contacts should be considered by everybody
applying this technique to electrically connect battery cells.
Acknowledgements
Funding from the Bavarian Ministry of Economic Affairs and
Media, Energy and Technology of the project PowerLab (
LABAY75E) is gratefully acknowledged.
The authors thank Korbinian Schmidt from the Institute for
Electrical Energy Storage Technology at the Technical University of
Munich for supporting the investigations with practical work-
manship.
The responsibility for this publication rests with the authors.
 M.J. Brand et al. / Journal of Energy Storage 8 (2016) 69–77
References
[1] B. Wu, V. Yufit, M. Marinescu, G.J. Offer, R.F. Martinez-Botas, N.P. Brandon,
Coupled thermal-electrochemical modelling of uneven heat generation in
lithium-ion battery packs, J. Power Sources 243 (2013) 544–554. http://dx.doi.
org/10.1016/j.jpowsour.2013.05.164.
[2] J. Yi, U.S. Kim, C.B. Shin, T. Han, S. Park, Three-dimensional thermal modeling of
a lithium-ion battery considering the combined effects of the electrical and
thermal contact resistances between current collecting tab and lead wire, J.
Electrochem. Soc. 160 (2013) A437–A443. http://dx.doi.org/10.1149/
2.039303jes.
[3] G.J. Offer, V. Yufit, D.A. Howey, B. Wu, N.P. Brandon, Module design and fault
diagnosis in electric vehicle batteries, J. Power Sources 206 (2012) 383–392.
http://dx.doi.org/10.1016/j.jpowsour.2012.01.087.
[4] T. Bruen, J. Marco, Modelling and experimental evaluation of parallel
connected lithium ion cells for an electric vehicle battery system, J. Power
Sources 310 (2016) 91–101. http://dx.doi.org/10.1016/j.jpowsour.2016.01.001.
[5] P. Taheri, M. Hsieh, M. Bahrami, Internal resistance matching for parallel-
connected lithium-ion cells and impacts on battery pack cycle life, J. Power
Sources 196 (2011) 6525–6533. http://dx.doi.org/10.1016/j.
jpowsour.2011.03.056.
[6] J.W. Braithwaite, A. Gonzales, G. Nagasubramanian, S.J. Lucero, D.E. Peebles, J.
A. Ohlhausen, W.R. Cieslak, Corrosion of lithium-ion battery current collectors,
J. Electrochem. Soc. (1999) 448–456. http://dx.doi.org/10.1149/1.1391627.
[7] P.A. Schmidt, M. Schweier, M.F. Zaeh, Joining of lithium-ion batteries using
laser beam welding: electrical losses of welded aluminum and copper joints,
Proc. Int. Congr. Appl. Lasers Electro-Optics (2012) 915–923.
[8] S.S. Lee, T.H. Kim, S.J. Hu, W.W. Cai, J.A. Abell, Joining technologies for
automotive lithium-ion battery manufacturing—a review, Proc. Int. Manuf. Sci.
Eng. Conf. (2010) 541–549. http://dx.doi.org/10.1115/MSEC2010-34168.
[9] S. Timsit, Electrical contact resistance: properties of stationary interfaces, Proc.
of 44th IEEE Holm Conf. on Electr. Contacts (1998) 1–19. http://dx.doi.org/
10.1109/HOLM.1998.722422.
[10] R. Holm, Electric Contacts: Theory and Application, 4th edition, Springer-
Verlag, Berlin, Germany, 1967.
77
[11] P.G. Slade, Electrical Contacts: Principles and Applications, 2nd edition, CRC
Press, Boca Raton, FL, 2013.
[12] K.L. Johnson, Contact Mechanics, Cambridge University Press, Cambridge, UK,
1987.
[13] E. Vinaricky, A. Keil, K.-H. Schröder, J. Weiser, W. Merl, Elektrische Kontakte,
Werkstoffe und Anwendungen Grundlagen, Technologien, Prüfverfahren, 2nd
edition, Springer-Verlag, Berlin, Germany, 2013.
[14] J.A. Greenwood, J.B.P. Williamson, Contact of nominally flat surfaces, Proc. R.
Soc. 295 (1966) 300–319. http://dx.doi.org/10.1098/rspa.1966.0242.
[15] M. Braunovic, N.K. Myshkin, V.V. Konchits, Electrical contacts: fundamentals,
Applications and Technology, CRC Press, Boca Raton, FL, 2006.
[16] STANDARD, Geometrical Product Specification (GPS) – Surface Texture: Profile
method – Terms, definitions and surface texture parameters, ISO 4287 (1997).
[17] M.J. Brand, P.A. Schmidt, M.F. Zaeh, A. Jossen, Welding techniques for battery
cells and resulting electrical contact resistances, J. Energy Storage 1 (2015) 7–
14. http://dx.doi.org/10.1016/j.est.2015.04.001.
[18] General information: Operating conditions and standards in pneumatics, Festo
AG Co. & KG, [Online] (available: 10.05.16). https://www.festo.com/rep/es_es/
assets/pdf/Druckluftaufbereitung_en.pdf.
[19] N. Alok, The Metals Databook, McGraw-Hill, New York, NY, 1997.
[20] Kupfer-Zink-Legierungen (Messing und Sondermessing), Deutsches
Kupferinstitut, [Online] (available: 01.05.16). https://www.kupferinstitut.de/
fileadmin/user_upload/kupferinstitut.de/de/Documents/Shop/Verlag/
Downloads/Werkstoffe/i005.pdf.
[21] HILUMIN ultra-clean nickel-plated steel, Tata Steel Europe Ltd., London, UK,
[Online] (available: 01.05.16). http://www.tatasteeleurope.com/en/products-
and-services/flat/electro-plated/hilumin.
[22] M. Hazewinkel, 18th edition, Encyclopaedia of Mathematics, Vol. 1, Kluwer
Academic Publishers, Dodrecht, Netherlands, 1995.
[23] J.F. Shackelford, Y.-H. Han, S. Kim, S.-H. Kwon, CRC Materials Science and
Engineering Handbook, 4th edition, CRC Press, Boca Raton, FL, 2016.
[24] P.A. Schmidt, T. Pauleser, M.F. Zaeh, Optimisation of weld seam configurations
using a genetic algorithm, Int. Conf. Dig. Enterpr. Tech. (2014) 393–399. http://
dx.org/10.1016/j.procir.2014.10.054.