A. O. KAPTI, I. ATAKUL: DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-BASED PRESS FOR PUNCHING ...
267–274
DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-
BASED PRESS FOR PUNCHING THIN METAL SHEETS
RAZVOJ STISKALNICE Z UMETNIMI PNEVMATSKIMI MI[ICAMI
ZA TANKOSTENSKO PREBIJANJE TANKIH KOVINSKIH
PLO^EVIN
Akin Oguz Kapti
1,2 *
, Ilker Atakul
1
1
Sakarya University, Mechanical Engineering Department, 54050, Sakarya, Türkiye
2
Sakarya University Research, Development and Application Center, 54050, Sakarya, Türkiye
Prejem rokopisa – received: 2023-02-01; sprejem za objavo – accepted for publication: 2023-04-05
doi:10.17222/mit.2023.783
The present study examines the nonlinear behavior of pneumatic, artificial muscles and investigates their availability for produc-
ing pressing forces over the experimentally determined tensile forces. It covers the design and manufacturing studies of a test
setup and a pneumatic, artificial-muscle-based press to achieve this goal. The press design consists of a single pneumatic artifi-
cial muscle to provide the main pressing force and another two to bring the press back to the neutral position. The proposed ap-
proach is considered sufficient for thin sheet-metal punching molds and fills a gap in the spectrum of pressing technologies. A
sufficient level of pressing force for thin sheet-metal punching is found to be achievable using a single 40-mm-diameter, pneu-
matic, artificial muscle. The results show that the press can produce (9.1, 23.1 and 36.9) kN pressing forces at (200, 400 and
600) kPa air pressures, respectively.
Keywords: pneumatic artificial muscle, table-top press, sheet metal, punching.
Predstavljeni so preizkusi nelinernegaobna{anja umetnih pnevmatskih mi{ic in raziskava mo`nosti izdelave stiskalnice na
osnovi eksperimentalno dolo~enih nateznih obremenitev. Za dosego `elenega cilja so oblikovali napravo in njeno postavitev,
analizirali mo`nosti za njeno izdelavo ter jo nato izdelali z ustreznimi umetnimi pnevmatskimi mi{icami. Izdelana stiskalnica je
imela eno umetno pnevmatsko mi{ico (napihljiv gumijast valj), ki je zagotavljala glavno silo stiskanja (kompresijo) in drugih
dveh, ki sta stiskalnico vra~ali nazaj v nevtralni polo`aj. Ocenili so, da predlagani pristop omogo~a prebijanje tankih kovinskih
plo~evin in s tem zapolnjuje vrzel oz. manjkajo~e vrste stiskalnic za vrsto stiskalnih tehnologij. Ugotovili so, da je
dose`enaraven prebijalne sile dovolj velika za prebijanje tankih kovinskih plo~evin pri uporabi ene umetne pnevmatske mi{ice
premera 40 mm. Rezultati so pokazali, da s stiskalnico lahko dose`ejo sile stiskanja velikosti (9,1in 23,1oz. 36,9) kN pri tlakih
zraka (200, 400 oz. 600) kPa.
Klju~ne besede: umetna pnevmatska mi{ica, namizna stiskalnica, kovinska plo~evina, prebijanje
1 INTRODUCTION
A pneumatic, artificial muscle (PAM) is a flexible
tensile actuator that allows more precise motion and po-
sition control, where larger forces can be achieved with
smaller powers than a pneumatic cylinder. It has been
named muscle because it works similarly to the principle
of the biological muscle, producing pulling force by con-
tracting and shortening. It consists of a contractible cy-
lindrical rubber diaphragm with a high force-to-weight
ratio (400:1). This diaphragm, which is made of a pres-
sure-resistant hose and strong rhomboid fibers, sur-
rounds the working environment and provides an airtight
seal. The working principle of the PAM is that the pneu-
matic pressure acting over the entire inner surface forces
the cylindrical structure to be spherical and thus creates a
tensile force in the longitudinal direction.
PAM has many outstanding features. It has the prop-
erties of a high force-to-weight ratio, structural flexibil-
ity, lightness, insensitivity to dirt (abrasive particles,
dust, etc.), a hermetically sealed structure, and ease of
handling. The pulling force produced by a PAM in its
initial position reaches up to ten times higher than that of
a conventional pneumatic cylinder of the same diameter.
It provides silent and precise positioning with pressure
regulation, including intermediate positions. It enables
large parts (side covers, plates, etc.) to be easily
clamped. Due to its adjustable spring force and flexibil-
ity, a PAM can absorb the high stresses that occur when
the films and papers are transported or wound using roll-
ers. It does not need to be constantly connected to the
compressed-air supply to function. It allows the ampli-
tude and cycle rate to be adjusted independently of each
other. No need for lubricant, as it does not contain any
moving parts. On the other hand, there are also some dis-
advantages, such as the maximum 25 % of the stroke
length being available, the reduction of the force to zero
depending on the stroke, and the absence of double-act-
ing operation availability.
PAMs have begun to attract attention again in recent
years. In the literature, there are many PAM-based stud-
ies conducted in different application areas, such as a
parallel manipulator designed using three PAMs,
1
a novel
actuator for an underwater robot retrieving nuclear
Materiali in tehnologije / Materials and technology 57 (2023) 3, 267–274 267
UDK 621.979.1 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 57(3)267(2023)
*Corresponding author's e-mail:
aokapti@sakarya.edu.tr (Akin Oguz Kapti)

waste,
2
a search-type rescue robot,
3
an active vibration
control system for a footbridge,
4
an underground ex-
plorer robot mimicking the peristaltic crawling of earth-
worms,
5
a crankshaft mechanism designed by using three
PAMs for a small rubber-wheeled vehicle,
6
and mechani-
cal actuation studies on PAMs that are driven by hydro-
gen-gas pressure provided by absorption of the metal hy-
dride.
7–9
Robotics engineers have also begun to
rediscover PAMs, and use them as robotic actuators and
prosthetic assistive devices.
10
They are also promising
for biomimetic-legged robots because of their adjustable
stiffness, high power/weight ratio, and structural flexibil-
ity.
11
A biomimetic exoskeleton powered by an antago-
nistic pair of PAM for reliable actuation in orthopedic re-
habilitation,
12
a two-link anthropomorphic arm
constructed by PAMs for the upper-limb rehabilitation
training,
13
a manipulator powered by PAMs for an
exoskeleton application,
14
a robotic lower limb consists
of a novel mechanism that is driven by a single PAM
combined with a torsion spring for therapy robots,
15
and
an assistive device for the agonist-antagonist actuation of
the human forearm
16
can be mentioned as PAM-based
studies focused on PAM behavior in robotic and pros-
thetic applications. The new type of bendable
17,18
or
twistable
19
PAMs generating grasping or torsional mo-
tions for soft actuators also exist in the literature.
Despite the variety of applications mentioned above,
a study in which PAMs are used to obtain pressing force
has not been found in the literature. This study aims to
experimentally measure the tensile force produced by
PAMs at different air pressures and contraction rates and
to examine their nonlinear behavior in a PAM-based
press over the experimentally determined tensile forces.
In this study the general structure of the PAM was inves-
tigated, and the design, manufacturing, and test studies
of a PAM-based press actuated with three PAMs were
carried out.
2 EXPERIMENTAL PART
2.1 Structure of the PAM
The basic dimensions and the force-generation model
of the PAM are given in Figure 1. Basically, a PAM op-
erates as a single-acting actuator against a pre-tensioning
load. When pressure is applied, the pre-tensioned PAM
contracts, shortens in length, increases in diameter, and
produces a tensile force that nonlinearly depends on the
contraction ratio, with optimum dynamic characteristics
and minimum air consumption. The radial force pro-
duced by air pressure is converted to the tensile force on
each non-extensible braided thread turning around the
tube with a definite initial angle. This angle has an essen-
tial role in transforming the radial force into tensional
force.
The initial length, the initial inner diameter, and the
initial volume of the PAM can be expressed by Equations
(1) – (3), respectively, as follows:
LL
00
=⋅
f
cos (1)
D
L
n
0
0
=
⋅
f
sin π
(2)
V
L
n
0
2
00
2
4
=
⋅⋅
f
3
sin cos 
π
(3)
where L
f
,  o
, and n are the length of rhomboid fibers,
initial fiber winding angle, and the number of windings,
respectively. To obtain the equation expressing the ten-
sile force generated by a PAM, the first PAM’s force
characteristic model based on the principle of energy
conservation was reconsidered by Chou and Hanna-
ford.
20
In this model, the PAM force is expressed by
Equations (4) – (6), respectively, as follows:
d
d
f
3
V L
n  
=
⋅⋅− 22
4
00
3
0
2
sin cos sin
π
(4)
d
d
f
L
L
  =− ⋅sin
0
(5)
Fp
V
L
p
L
n
p
D
pam
f
2
0
2
d
d
=− =
⋅−
=
=
−
( cos )
cos
31
4
4
31
1
2
0
2
2
0
  π
π
−
⎛ ⎝ ⎜ ⎜ ⎞ ⎠ ⎟ ⎟ cos
2
0
 (6)
The force obtained from Equation (6) is proportional
to the air pressure and the initial cross-sectional area.
The fiber-winding angle, which is 28.6° at the initial
state,
21
increases up to 54.7° in the case of the minimum
length and the maximum contraction, where the force is
zero. The error between the measured and calculated
force characteristic of the PAM is not negligible because
the effects of the membrane’s elasticity are entirely ne-
A. O. KAPTI, I. ATAKUL: DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-BASED PRESS FOR PUNCHING ...
268 Materiali in tehnologije / Materials and technology 57 (2023) 3, 267–274
Figure 1: Basic dimensions and the force-generation model of the
PAM

glected in the Schulte model. This model is insufficient
as it allowed the correct calculation of the PAM’s tensile
force only in the case of the initial fiber angle. It was
later developed by Tondu and Lopez
22
as follows:
F
pA
ce c
LL
L
pam
p
=− ⋅ +
− ⎛ ⎝ ⎜ ⎜ ⎞ ⎠ ⎟ ⎟ ⎡ ⎣ ⎢ ⎢ ⎤ ⎦ ⎥ −
3
1
0
2
2
0
12
0
0
tan
()
 ⎥ −
pA
0
2
2
0
sin  (7)
where c
1
and c
2
are constants specified experimentally.
Wickramatunge and Leephakpreeda
23
developed a new
model supposing that the force characteristic of a PAM
is equivalent to a mechanical spring with displacement
and pressure-dependent spring stiffness. In their model,
the pressure-dependent spring stiffness, shortening, and
PAM force are expressed by Equations (8) – (10) as fol-
lows:
kcpcpLcLc =+⋅++ ()
3
2
21
2
0
ΔΔ (8)
ΔLLhphph =− + + ()
2
2
10
(9)
FkL
pam
=⋅ Δ (10)
where; c
0
, c
1
, c
2
, c
3
, h
0
, h
1
and h
2
are constant coeffi-
cients specified experimentally. Considering the mem-
brane’s elasticity, Martens and Boblan
21
developed a
PAM force model using Equations (11) – (12) as fol-
lows:
Fp
D
EHL
D
pam
0
2
ru pe
d
d
=
−
−
⎛ ⎝ ⎜ ⎜ ⎞ ⎠ ⎟ ⎟ −
π
π
4
31
1
2
0
2
0
0
cos
cos
   L
EHD
−
−
ru 1
 0
π
(11)
Ec Lc Lc Lc
ru
=+++
3
3
2
2
10
(12)
where E
ru
is the modulus of elasticity of the membrane,
 pe
is the strain in the perimeter direction,  l
is the strain
in the length direction, H
0
is the wall thickness of the
membrane, and c
0
, c
1
, c
2
and c
3
are the constant polyno-
mial coefficients specified experimentally.
2.2 Test setup
Figure 2 shows the experimental setup for the PAM’s
tensile force measurement. It consists of an air compres-
sor, conditioner, directional valve, proportional air-pres-
sure valve, and load cell. The main idea of the experi-
ment is to measure the tensile force generated by the
PAM at different contraction ratios and operating pres-
sures. The proportional air-pressure valve adjusts the air
pressure in the PAM, and the directional air valve manip-
ulates the PAM connection between the compressor line
and the atmospheric environment. The PAM length is
fixed during the measurement with a length adjustment
screw. The pressure is varied between the lower and up-
A. O. KAPTI, I. ATAKUL: DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-BASED PRESS FOR PUNCHING ...
Materiali in tehnologije / Materials and technology 57 (2023) 3, 267–274 269
Figure 3: PAM-based press design: a) solid CAD model, b) schematic structure
Figure 2: Test setups: a) PAM–20–200, b) PAM–40–600

per limits with intervals of 3.9 kPa, which is the accuracy
of the proportional pressure valve. The tensile force of
the PAM is measured by a load cell. The measurement
process was repeated three times to reduce errors, and
the force graphics are drawn according to the average
values.
2.3 PAM-based press design
The solid CAD model and schematic structure of the
developed press actuated with three PAMs are given in
Figure 3. The press includes a clamping system consist-
ing of articulated arms. This mechanism converts the
pulling force produced by the single-acting PAM actua-
tor into a pressing force. A cam block connected to the
end of the PAM opens the arms of the clamping system.
The PAM force is transmitted to the moving plate with
the force-increasing reduction effect of the clamping sys-
tem.
The main selection parameters of the PAM are the di-
ameter and length. These parameters are defined as the
initial inner diameter and visible muscle length between
the connections in the non-pressurized and load-free
state, respectively. The first one of the selected PAMs
(PAM-40-600: 40 mm initial diameters and 600 mm ini-
tial lengths) provides the main press force, and the other
two (PAM-20-200: 20 mm initial diameters and 200 mm
initial lengths) return the press to the neutral position.
The weight of the cam block makes up the pre-ten-
sioning effect on the first one. The technical properties
of the PAMs used are listed in Table 1.
The schematic for the formation of the pressing force
and scaled CAD drawing are given in Figure 4. In this
figure, F
pam
is the tensile force of the PAM, F
n
is the nor-
mal force to the joint, F
r
is the perpendicular force to the
vertical axis of the clamping system, F
a
is the clamping
arm force, F
p
is the pressing force,  is the angle of the
cam block,  is the angle between the arms and the verti-
cal axis, and i is the reduction ratio of the press mecha-
nism. The mathematical formulation depending on this
notation is expressed by Equations (13) – (17) as fol-
lows:
FF
np a m
= cos  (13)
FF
rp a m
= cos sin  (14)
F
F
a
r
=
⋅+ − 2 1 180 2 ( cos( ))  (15)
F
F
p
pam
=
⋅+ −
2
2 1 180 2
cos sin cos
( cos( ))

 (16)
i =
⋅+ −
2
2 1 180 2
cos sin cos
( cos( ))

 (17)
A. O. KAPTI, I. ATAKUL: DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-BASED PRESS FOR PUNCHING ...
270 Materiali in tehnologije / Materials and technology 57 (2023) 3, 267–274
Figure 4: a) Schematic for the generation of the pressing force, b) scaled CAD drawing
Table 1: Technical properties of the PAMs used in the press design.
PAM-20 PAM-40
Inner diameter 20 mm 40 mm
Nominal length 60–9000 mm 120–9000 mm
Air pressure (max.) 600 kPa 600 kPa
Frequency (max.) 3 Hz 2 Hz
Tensile force
(max. at 600 kPa)
1500 N 6000 N
Permissible freely
suspended load
80 kg 250 kg
Permissible
pre-tensioning
0.04·Nominal
length
0.05·Nominal
length
Permissible contrac-
tion
0.25·Nominal
length
0.25·Nominal
length
Permissible working
temperature
–5 … +60 °C –5 … +60 °C
Fitting size
M10×1.25–
M20×1.5
M16×1.5–
M30×1.5
Weight 169 g + 178 g/m 675 g + 340 g/m

The press was manufactured after the optimization of
the main dimensions and the sizing calculations of the
components. The diameter of the pins and the thickness
of the fixed and moving plates were determined as 16
mm and 30 mm. The maximum deformation at the plates
was found to be 0.14 mm in the deformation analysis.
Modeling and design-validation studies according to the
sizes obtained in the calculations, material supply, manu-
facturing, and assembly works were carried out. The
general view of the developed system and pneumatic cir-
cuit diagram are shown in Figure 5. The pneumatic cir-
cuit consists of the air compressor, manometer, 5/2 le-
ver-operated directional control valve, one PAM for main
press force, and two PAMs for bringing back the press to
the neutral position. The nominal operating pressure,
tank volume, and flow rate of the air compressor are
800 kPa, 24 L, and 198 L/min, respectively.
3 RESULTS AND DISCUSSION
The force graphs obtained in the experimental studies
carried out are given in Figure 6. It is observed that the
tensile force generated by the PAM is directly propor-
tional to the air pressure and the initial cross-sectional
area. It is the function of the stroke resulting from the
change in fiber orientation. Contraction rates below 12 %
provide an effective working range, and due to that the
working efficiency increases as the contraction rate be-
comes smaller. In this operating range the force varies as
1606–682 N for PAM-20-200 and 6321–3071 N for
PAM-40-600. At the initial state, the contraction ratio is
zero, and the force has its maximum value.
Conversely, the contraction ratio reaches its maxi-
mum value, while the force decreases to zero at the final
A. O. KAPTI, I. ATAKUL: DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-BASED PRESS FOR PUNCHING ...
Materiali in tehnologije / Materials and technology 57 (2023) 3, 267–274 271
Figure 6: Tensile forces of the PAMs used: a) PAM–20–200,
b) PAM–40–600
Figure 5: PAM-based press: a) general view, b) pneumatic circuit diagram (1: air compressor, 2: manometer-filter, 3: 5/2 directional control
valve, 4: PAM–20–200, 5: PAM–40–600)

state. The decrease of the force gives smoothness to the
movement. In contrast to this, pneumatic cylinders pro-
duce the same force throughout their entire stroke and re-
quire cushioning at the end of the stroke to avoid the im-
pact effect when the speed is suddenly reset to zero. The
PAM behaves like a spring and follows the application of
force when the external force changes. Both the
pre-tensioning force and the spring stiffness of this pneu-
matic spring can be varied. Operational situations such
as a constant-pressure or constant-volume spring provide
suitable properties.
The results obtained from press tests are listed in Ta-
ble 2. Among the values in this table, the clamping-arm
angle, PAM stroke, and displacement were determined
by the scaled CAD drawing in Figure 4b. The pressing
forces and reduction ratios were determined by Equa-
tions (16) and (17) according to the angles ( ,  ) and
PAM forces.
A. O. KAPTI, I. ATAKUL: DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-BASED PRESS FOR PUNCHING ...
272 Materiali in tehnologije / Materials and technology 57 (2023) 3, 267–274
Table 2: Experimental results obtained at 200 kPa, 400 kPa and 600 kPa air pressure for PAM-40-600.
Pressure
p (kPa)
Arm angle
 (deg)
PAM stroke PAM force
Fpam
(N)
Displacement
h
s
(mm)
Reduction
ratio
Press force
Fp
(N) h
pam
(mm) h
pam
(%)
200
22 0 0 2130 0 1.07 2283
16.35 18 3 1569 9.06 1.48 2316
11.26 36 6 1221 15 2.17 2656
6.52 54 9 968 18.58 3.79 3667
2.06 72 12 757 20.2 12.04 9113
400
22 0 0 4240 0 1.07 4544
16.35 18 3 3422 9.06 1.48 5051
11.26 36 6 2832 15 2.17 6159
6.52 54 9 2340 18.58 3.79 8865
2.06 72 12 1923 20.2 12.04 23149
600
22 0 0 6321 0 1.07 6774
16.35 18 3 5248 9.06 1.48 7746
11.26 36 6 4405 15 2.17 9580
6.52 54 9 3693 18.58 3.79 13991
2.06 72 12 3071 20.2 12.04 36969
Figure 7: Tensile forces of PAM–40–600 and the pressing forces at: a) 200 kPa, b) 400 kPa, c) 600 kPa, d) reduction ratio of the press mecha-
nism

The clamping system used in the design makes up a
reduction effect that increases the pressing force. The re-
duction ratio increases as the  angle of the clamping
arms with the vertical axis increase. Figure 7a to 7c
shows the pulling force produced by the PAM-40-600
and the pressing force obtained when (200, 400 and 600)
kPa of air pressure is applied, respectively. According to
these results, it can be said that the pressing forces of
(9.1, 23.1, and 36.9) kN are obtained by using a single
40-mm-diameter PAM. When these figures are exam-
ined, the muscle produces the maximum pulling force at
the beginning, and this force decreases towards the end
of the stroke. This is in line with the characteristic way
of working of the muscle. Since the sheet-metal forming
process takes place at the end of the stroke, the force
drop towards the end of the stroke is not a desirable situ-
ation. On the other hand, the press mechanism comes
into play at this stage. There is a radical increase in the
pressing force, while the pulling force of the muscle de-
creases. The pressing-force graphs in these figures indi-
cate that the most suitable region for punching the sheet
metal is the contraction range 9–12 %. It is desirable in
the case of sheet-metal cutting and punching molds be-
cause the press moves freely at first and does the work
when it comes to the stroke end. In addition, while the
tensile force of the PAM is reduced with the stroke, this
reduction effect of the clamping system acts to balance
the force. Figure 7d shows the reduction ratio change
concerning the contraction of PAM-40-600.
In presses where pneumatic and hydraulic cylinders
are used, the pressing force and velocity at a certain pres-
sure are fixed throughout the stroke. Here, in contrast,
the pressing force and velocity are variable throughout
the stroke. Because the tensile force of the PAM is vari-
able depending on the stroke and the clamping system’s
reduction effect. In addition, the cam block geometry is
an effective parameter. In the case of a linear cam-block
geometry, the pressing characteristic can be controlled
via its angle and height. The case of a curvilinear
cam-block geometry will further enhance the control
flexibility of the system.
The desired final metal parts are obtained by apply-
ing cutting, punching, bending, and forming processes to
sheet metals in mechanical and hydraulic presses. Figure
8a shows a schematic of the sheet-metal punching proce-
dure. In punching molds, the elasticity limit of the sheet
material is exceeded by the punch force, and plastic de-
formation begins. As the punch plunges into the material
up to 0.3 times the thickness, the part starts to flow to-
ward the lower mold cavity. When 0.6 times the sheet
thickness is reached, the cutting process is completed,
and the part is pushed and dropped from the mold cavity.
The parameters that determine the pressing-force re-
quirement are the shear strength of the material, the sheet
thickness, and the length of the cutting contour. De-
pending on these parameters, the pressing force required
for a circular part is expressed by Equation (18) as fol-
lows:
Fd t Rc
pm
=⋅⋅⋅ ⋅ π (18)
where d, t, R
m
and c are the diameter of the circular part,
the thickness of the sheet metal, the tensile strength of
the material, and the processing coefficient, respec-
tively. To show the suitability of the pressing force ob-
tained by PAM-40-600 for thin sheet-metal parts (t<1
mm), pressing-force calculations were made for com-
monly used metals. Tensile strengths of the 6061–an-
nealed aluminum, C11000–hot-rolled copper,
1020–hot-rolled carbon steel, and 316–annealed stain-
less steel are considered as 124, 220, 380, and 515 MPa,
respectively.
24
The force required for sheet-metal forming is a func-
tion of the cut-to-length of the part, the thickness of the
sheet metal, and the tensile strength of the material. As
the diameter and thickness of a circular part and the ten-
sile strength of the material increase, the required form-
ing force also increases. Figure 8b shows the sheet-
forming forces needed for circular parts with a wall
thickness of 1 mm and a diameter of 10–50 mm made of
the above-mentioned materials. When the figure is exam-
ined, it is seen that the required force for the circular part
with a wall thickness of 1 mm and a diameter of 50 mm,
which is made of annealed 316 stainless steel with the
highest tensile strength among the above-mentioned ma-
terials, is 33.9 kN. The obtained results given in Figure
8b show that the pressing force obtained by a single
PAM-40-600 is sufficient for thin sheet-metal punching
molds.
A. O. KAPTI, I. ATAKUL: DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-BASED PRESS FOR PUNCHING ...
Materiali in tehnologije / Materials and technology 57 (2023) 3, 267–274 273
Figure 8: Pressing force during sheet-metal punching: a) schematic,
b) press force requirements

4 CONCLUSIONS
The design, manufacturing, and test studies of a press
actuated with three PAMs were carried out. The PAMs
are suitable for generating the pressing force, besides the
robotic, prosthetic, and biomimetic applications. At the
end of the study, a prototype of the PAM-based press
was developed as an alternative approach to hydraulic
and mechanical presses. It can be used for a more precise
determination of the areas where the PAM-based systems
can be operated and in which design changes can be
made in accordance with the requirements and adapta-
tions to the related sectors. Since the PAM is a hermeti-
cally sealed system and not equipped with any moving
components, it is possible to operate the developed sys-
tem in dirty and dusty environments and even in water.
The production cost is lower because of the relatively
cheap pneumatic circuit elements. Although the
PAM-based studies cover many areas, such as manipula-
tors, automation systems, robotic actuators, and pros-
thetic assistive devices, there is no study on presses with
PAMs in the literature.
This study shows that a sufficient level of pressing
force can be obtained with a single pneumatic artificial
muscle to form thin sheet metals and fills the gap in the
related literature. The force-enhancing effect of the press
mechanism compensates for the decrease in the tensile
force of the pneumatic artificial muscle as the contrac-
tion rate increases. It is suitable because the pressing
force increases, and the speed decreases during punching
at the end of the stroke. The results show that pressing
forces of (9.1, 23.1, and 36.9) kN are obtainable by using
a single 40-mm-diameter PAM at (200, 400, and
600) kPa air pressures, respectively.
Acknowledgment
This work was supported by TUBITAK, The Scien-
tific and Technological Research Council of Turkey
[grant number 122M619].
5 REFERENCES
1
X. Zhu, G. Tao, B. Yao, J. Cao, Adaptive robust posture control of a
parallel manipulator driven by pneumatic muscles, Automatica, 44
(2008) 9, 2248–2257, doi:10.1016/j.automatica.2008.01.015
2
K. K. K. Ku, R. S. Bradbeer, K. K. Y. Lam, L. F. Yeung, R. C. W. Li,
A novel actuator for underwater robots, IEEE J. Ocean. Eng., 34
(2009) 3, 331–342, doi:10.1109/JOE.2009.2014928
3
K. Harihara, S. Dohta, T. Akagi, F. Zhang, Development of a search
type rescue robot driven by pneumatic actuator, in: SICE Annual
Conf., Taipei, Taiwan, (2010) 1311–1317, doi:10.1299/jsmecs.
2010.48.303
4
A. Bleicher, T. Schauer, M. Valtin, J. Raisch, M. Schlaich, Active vi-
bration control of a light and flexible stress ribbon footbridge using
pneumatic muscles, in: IFAC Annual Conf., Milano, Italy, (2011),
911–916, doi:10.3182/20110828-6-IT-1002.02781
5
T. Kishi, T. Nakamura, M. Ikeuchi, Development of a peristaltic
crawling inspection robot with pneumatic artificial muscles for a
25A elbow pipe, Nature-Inspired Mobile Robotics, (2013), 301–308,
doi:10.1142/9789814525534_0039
6
G. Soy, G. Samtas, S. Korucu, Design and evaluation of pneumatic
movement mechanism supported with solar panels, J. Polytech., 17
(2014), 135–142, doi:10.2339/2014.17.3 135-142
7
T. Leephakpreeda, Mathematical modeling of pneumatic artificial
muscle actuation via hydrogen driving metal hydride-LaNi5,J .
Bionic Eng., 9 (2012) 1, 110–118, doi:10.1016/S1672-6529(11)
60103-0
8
T. Nuchkrua, T. Leephakpreeda, Actuation of pneumatic artificial
muscle via hydrogen absorption/desorption of metal hydride-LaNi5,
Adv. Mech. Eng., 7 (2015) 1, 364306, doi:10.1155/2014/364306
9
T. Nuchkrua, T. Leephakpreeda, Novel Compliant Control of a Pneu-
matic Artificial Muscle Driven by Hydrogen Pressure Under a
Varying Environment, IEEE Trans. Ind. Electron., 69 (2022)7 ,
7120–7129, doi:10.1109/TIE.2021.3102486
10
T. Nuchkrua, T. Leephakpreeda, T. Mekarporn, Development of ro-
bot hand with Pneumatic Artificial Muscle for rehabilitation applica-
tion, in: NANOMED Annual Conf., Phuket, Thailand, (2013),
55–58, doi:10.1109/NANOMED.2013.6766315
11
A. Ho{ovský, J. Pitel, K. @idek, J. Sárosi, L. Cveticanin, Dynamic
characterization and simulation of two-link soft robot arm with pneu-
matic muscles, Mech. Mach. Theory, 103 (2016), 98–116,
doi:10.1016/j.mechmachtheory.2016.04.013
12
A. Merola, D. Colacino, C. Cosentino, F. Amato, Model-based track-
ing control design, implementation of embedded digital controller
and testing of a bio-mechatronic device for robotic rehabilitation,
Mechatronics, 52 (2018), 70–77,
doi:10.1016/j.mechatronics.2018.04.006
13
T. Wang, X. Chen, W. Qin, A novel adaptive control for reaching
movements of an anthropomorphic arm driven by pneumatic artifi-
cial muscles, Appl. Soft. Comput., 83 (2019), 105623, doi:10.1016/
j.asoc.2019.105623
14
G. V. Prado, M. B. C. Sanchez, Control strategy of a pneumatic arti-
ficial muscle for an exoskeleton application, IFAC Papers OnLine, 52
(2019), 281–286, doi:10.1016/j.ifacol.2019.06.075
15
T. C. Tsai, M. H. Chiang, Design and control of a 1-DOF robotic
lower-limb system driven by novel single pneumatic artificial mus-
cle, Appl. Sci., 43 (2020) 1, 1–21, doi:10.3390/app10010043
16
B. Kalita, S. K. Dwivedy, Nonlinear dynamic response of pneumatic
artificial muscle: A theoretical and experimental study, Int. J.
Non-Linear Mech., 125 (2020), 103544, doi:10.1016/j.ijnonlinmec.
2020.103544
17
G. Zhong, Y. Hou, W. Dou, A soft pneumatic dexterous gripper with
convertible grasping modes, Int. J. Mech. Sci., 153 (2019), 445–456,
doi:10.1016/j.ijmecsci.2019.02.028
18
G. Zhong, W. Dou, X. Zhang, H. Yi, Bending analysis and contact
force modeling of soft pneumatic actuators with pleated structures,
Int. J. Mech. Sci., 193 (2021), 106150, doi:10.1016/j.ijmecsci.
2020.106150
19
W. Xiao, X. Du, W. Chen, G. Yang, D. Hu, X. Han, Cooperative col-
lapse of helical structure enables the actuation of twisting pneumatic
artificial muscle, Int. J. Mech. Sci., 201 (2021), 106483,
doi:10.1016/j.ijmecsci.2021.106483
20
C. P. Chou, B. Hannaford, Measurement and modeling of McKibben
pneumatic artificial muscles, IEEE Trans. Robot. Autom., 12 (1996)
1, 90–102, doi:10.1109/70.481753
21
M. Martens, I. Boblan, Modeling the static force of a Festo pneu-
matic muscle actuator: A new approach and a comparison to existing
models, Actuators, 6 (2017) 4, 1–11, doi:10.3390/act6040033
22
B. Tondu, P. Lopez, Modeling and control of McKibben artificial
muscle robot actuators, IEEE Control Syst. Mag., 20 (2000)2 ,
15–38, doi:10.1109/37.833638
23
K. C. Wickramatunge, T. Leephakpreeda, Study on mechanical be-
haviors of pneumatic artificial muscle, Int. J. Eng. Sci., 48 (2010) 2,
188–198, doi:10.1016/j.ijengsci.2009.08.001
24
W. D. Callister, D. G. Rethwisch, Materials science and engineering,
8
th
ed., Wiley, (2011)
A. O. KAPTI, I. ATAKUL: DEVELOPMENT OF A PNEUMATIC, ARTIFICIAL-MUSCLE-BASED PRESS FOR PUNCHING ...
274 Materiali in tehnologije / Materials and technology 57 (2023) 3, 267–274