Carlos Americo
de Souza Silva
Industrial
and Systems Engineering Department, Technology Center, Federal University of
Santa Catarina, Brazil
E-mail: camericoss@gmail.com
Iracyanne
Retto Uhlmann
Industrial
and Systems Engineering Department, Federal University of Santa Catarina, Brazil
E-mail: iracyanne.uhlmann@gmail.com
Enzo
Morosini Frazzon
Industrial
and Systems Engineering Department, Technology Center, ederal University of
Santa Catarina, Brazil
E-mail: enzo.frazzon@ufsc.br
Submission: 4/15/2019
Accept: 5/17/2019
ABSTRACT
This paper analyses a new screw torque traceability control procedure designed for an operation in a car audio workstation. First the detection score at a Failure Mode and Effects Analysis (FMEA) indicated which process improvement were needed. On the sequence, data was collected to evaluate performance indicators. Finally, specific actions were executed based on the improvement plan. As result, the FMEA detection score improved from 7 to 3, the run rate increased from 12 to 16 products/h and defects caused by improper screwing were eliminated. This research contributes to disseminate a practical application of a screw torque traceability control.
Keywords: screwdriver; torque; screw; traceability; FMEA.
1.
INTRODUCTION
Screw-tightening is an important
assignment in assembly processes due to their wide use in various types of
manufactured products. For example, about three million screws are used in a
plane, and about three thousand screws are used in an ordinary car (Li et al.,
2009). According to Ogushi et al. (2015) in almost all precise electrical and
mechanical parts, hand torque screwdrivers are used for the fastening control
of screws.
In any kind of production process, not
only the task should be properly designed but also the measurement system
should be properly implemented to keep a good product quality and an efficient
flow. For Chen (2014) and Berger et al. (2016), accurate measurement of
external forces is also important in manipulation tasks involving tool-usage,
the fastening torque and the angular displacement of the screw have to be
closely monitored.
Muelaner et al. (2010) describe
"the accuracy of the measurement system is lower than that of assembly
design takes place", calling this statement as a phenomenon. Currently,
industries are still using hand torque, screwdrivers are used for the fastening
control of screws and torque traceability is almost non-existent.
Motivated by the challenges of
torque control in car audio production process and based on the literature,
this paper applies an action-research methodology for analyzing and improving
the automation of screwing and measuring torque performance processes. Also,
torque traceability control was achieved with the improvements applied,
contributing to disseminate a practical application of a screw torque
traceability control, which allows the verification and the validation of a
basic requirement in several productive processes and industries.
2.
THEORETICAL FRAME
To assure the strength connection
throughout the life length of a car, critical issues must be developed not only
to the design of the joint but also to the assembly operation (Hermansson,
2016). Screw connection is one of the most prevalent connection method in the
assembly industry, in which impact wrench plays an important role (OGAWA et
al., 2015; GANESHMURTHY; NASSAR, 2014; WOLF; LORENZ, 2011).
Associated with screw connections,
the pushing force and the rotary torque are required for screw tightening,
which can be performed by humans or automated machines. Humans tighten the
screws based on their feelings or on their experiences by feeling the reaction
torque and force through the screwdriver. Electrical screwdriver is a precise
method to control the rotary torque, "it can rotate a screw in high
angular velocity, and come to a stop at the end of screw tightening in order to
prevent applying excess rotary torque to the product" (OGAWA et al.,
2015).
Liu et al. (2015) emphasize that in
a screw-tightening performed by humans, the pre-tightening torque and
effectiveness of the assembly process are determined to the operator's
experience and improper pre-tightening torque can cause screw defects, thus
automatic screw tightening machines are essential in modern assembly industry.
Wen et al. (2016) add the great significance of the measurement and control of
the interaction force between components to maintain the quality of the final
product. Bischoff et al. (2010) wrote about the emergence of robots with joint
torque sensing and feedback control.
Persson and Roloff (2016) define the
tightening methods implemented in production for reaching a specified clamp
force, which are:
1. Torque control: when tightening
shuts off at a predefined torque level T, it is called torque control. Often it
is carried out in two steps, a high-speed rundown until mating of all parts
which occurs at a low torque, and a lower speed when aiming for the target
torque. Tightening using torque control can only be made in the elastic range
of the fastener;
2. Angle control: is when the target
value in the last step is an angle, ϕ. This method is often carried out
with a torque step as first step to make sure that all parts have mated. Angle
control can be used both in the fasteners elastic and plastic range;
3. Gradient control: when tightening
into the plastic range, gradient control can be used. In this method, the
gradient of the torque angle rate is evaluated. Target is reached when a
predefined gradient ∆T/∆ϕ is reached;
4. Clamp force or elongation control:
is based on bolt elongation measurement. Mechanical and ultrasonic elongation
are examples of methods to measure the clamp force/elongation. Clamp force
control has been investigated thoroughly; however, implementing it in production
has been slow, due to the difficulty of assuring correct measurements and
repeatability. Thus, manual measuring, before and after tightening, is costly
and labor intensive.
3.
METHODS
The need of this research was
triggered by the FMEA evaluation during a New Product Introduction (NPI)
project. According to Schneider (1996), FMEA is used to evaluate a new system,
or design, or process or service taking into consideration all possibilities of
failure occurrence. It is focused in discovering and controlling problems
before their occurrences. First, all the potential failures are identified; on
the sequence, a critical analysis is performed taking into account three risk
factors: occurrence (O), severity (S) and detection (D); finally, the Risk
Priority Number (RPN) is determinate by the multiplication of the O, S and D
values of a failure (LIU et al., 2013). The highest priority must be given to
the failure modes with highest value of RPN whose are supposed to be more
critical those having a lower RPN (SAFARI et al., 2016). Corrective actions
must be executed and the RPN should be recalculated, to check the efficiency of
these actions (LIU et al., 2013).
The methodology applied in this
study is empirical based on action-research concepts. Tripp (2005) states that
action-research is one of the many different ways of investigation and action;
he defines it as any continued, systematic and empirically justified attempt to
improve the practice. He also groups the four phases of the basic cycle of
action-research into two big phases: (1) action composed by planning and acting
stages and (2) research composed by monitoring and evaluating stages.
The empirical methodology applied in
this study follows the following steps:
1. Problem definition based on high RPN
found during the FMEA study;
2. Cause analysis performed by process
specialists;
3. Improvement actions taken by process
specialists;
4. Results, based on the RPN
recalculation and improvement of other indicators.
4.
INDUSTRIAL APLICATION
4.1.
Object of study
The presented study was applied in a
multinational industry that designs and manufactures audio products for
automotive companies, they market for more than twenty brands around the world.
Company has part of production dedicated to contract manufacturing services and
follows a lean manufacturing culture. The issue analyzed occurred in the
workstation “Display Assy 2”, where chassis, front panel and top cover are
screwed.
4.2.
Problem definition
During the plan phase of the NPI
project, to assure an effective production process in product failures
prevention, a multidisciplinary team evaluated the FMEA. The team reported that
all defects related to the screw fixing process (badly assembled, missing,
dusted, broken and wrong) had high severity in potential failure mode (8 score
points), low cause occurrence (2 score points) and high detection control (7
score points), accounted 112 score points of RPN. Based on this score, the
conclusion was that the main problem was related to detection and it should be
necessary take improvement actions to reduce the detection control score.
4.3.
Cause analysis
The high score in detection control
stemming from the manual torque process, which the detection probability is too
low. The electrical screwdriver did not have transducer and the process control
used to evaluate the torque was manual, depending on the operator’s skills.
Figure 1 shows the resources (Manual jig and electric screwdriver) used to
perform the operation in the workstation Display Assy 2. Although the
electrical screwdriver met the whole product requirements, it did not have the
ability to program different torque specifications at the same time. Therefore,
the adjustment was performed manually, using an external torque wrench to set
the exact value to be applied. In addition, it was not possible to have
individual screw torque traceability using this manual screwing process.
Figure 1: Manual jig and electric screwdriver
used in Display Assy 2 workstation
4.4.
Improvement actions
The acquisition of a
new electric screwdriver with controller, and the design of a new
semi-automatic jig (figure 2) changed the performance of the Display Assy 2
workstation, improving the detection and ensuring the torque traceability
control for each product, ensuring reliability to the customers.
Automation and
mandatory assembly sequence was taken into account during the jigs development,
considering the use of sensors to detect parts of the products and avoiding
operator’s mistakes. The electric transducer screwdriver used to control traceability,
enabled accuracy and assured error proofing. The tightening strategy for
seating detection was implemented, and information about angle and torque
parameters were considered in Manufacturing Execution System (MES) records.
The control of each
screwing operation was feasible even with material variation due to the
improvement actions applied. In addition, the screwing process was divided into
two parts: (1) thread forming and (2) tightening.
For the thread forming
phase, two options were possible:
1. Run down speed: the filament opening
is defined according to a target torque value. This new target is obtained
based on the observation of torque and angle-tightening curve in function of
time. Values are analyzed after torque undulations due to opening of filaments;
2. Angle Run Down: it uses only the
number of rotations to be performed, however, it is necessary to set the
maximum torque and the safe value to avoid the "blanking" of the
material. According to the team analysis, the “Angle Run Down” should be the
best strategy, due to the option to avoid torque peaks in the beginning of the
screwing process.
For the tightening
phase (end of speed process), other strategies could be used: torque, torque
and angle, yield point, seating detection and post seating detection.
Nevertheless, "torque and angle" was chosen, due to its use with
gradient of tightening is not feasible, allowing the best process control. In
addition, the quality standard required by the customer could be kept.
In this automated process,
the performance of the electric transducer screwdriver with controller was
measured; figure 3 shows the aspects evaluated:
· Torque (blue line): in the
beginning, a strong force is applied because the screw is opening the hole;
after that the force is decreased, performing screw tightening; at the end,
based on speed and angle information, the controller increases the force to
achieve the torque specification;
· Speed (purple line): in the
beginning, more rotations are used because the screw is opening the hole.
Afterwards, the rotation is decreased to support the torque-applying step;
· Angle (red line): due the fixed
value of the screw diameter, the angle shows a linear performance, the applied
force does not influence this parameter.
Figure 2: Semi-automatic jig and electric
screwdriver with controller used in Display Assy 2 workstation.
Figure 3: Torque curve performance during
strategy definition – Torque (ft) x Angle (ft) x Speed (ft).
4.5.
Results
The main achievements
were evidenced as follow:
1. FMEA detection score (screw
failures) was improved from 7 to 3 (this new score means that problems can be
detected in the source, i.e. detection of failure mode in the station by
automatic controls, preventing improper operation). This new detection score improved
the RPN (severity x occurrence x detection), from 112 score (8 x 2 x 7) to 48
score (8 x 2 x 3);
2. Traceability (that was lacking in
the past) was feasible. Torque performance data for each screw according to the
product serial number are stored in MES records;
3. Run rate increased from 12
products/h to 16 products/h;
4. Defects caused by improper screwing
were eliminated: the screwing processing was automated, improving defect
detection and quality control at the shop floor and, consequently, preventing failures,
as shown in figure 4:
Figure 4: Daily defect control due to improper
screwing.
5.
CONCLUSIONS
This research analyzed
a new screw torque traceability control procedure designed for an operation in
a car audio workstation. Positive impacts from FMEA execution were evidenced
with practical implementation of automated screwing process, as follow: (1) the
FMEA Risk Priority Number was improved, (2) traceability control was
established, (3) the production output rate was increased and (4) screwing defects
were eliminated. Based on that, it is possible to state that the torque-angle
strategy resulted in the best connection between the screwed parts, compared
with the others. The research was applied in a real industry, consequently,
besides the technical achievements, the involved team was recognized by the
plant staff and the customers.
For future studies, it
is important to keep focus on practical applications for other screwing
strategies, analyzing screw types and the characteristics of the materials that
will be connected. Variables of the product and the process should also be
included in order to reduce defects and manufacturing costs.
REFERENCES
BERGER, E.; GREHL, S.; VOGT, D.;
JUNG, B.; AMOR, H. B. (2016) Experience-based torque estimation for an industrial
robot. In Robotics and Automation
(ICRA), 2016 IEEE International Conference on (p. 144-149). IEEE.
BISCHOFF, R.; KURTH, J.; SCHREIBER,
G.; KOEPPE, R.; ALBU-SCHÄFFER, A.; BEYER, A.; HIRZINGER, G. (2010) The KUKA-DLR
Lightweight Robot arm-a new reference platform for robotics research and
manufacturing. In Robotics (ISR), 2010
41st international symposium on and 2010 6th German conference on robotics
(ROBOTIK) (p. 1-8). VDE.
CHEN, C. H. (2014) Fastening torque
control for robotic screw driver under uncertain environment. In Control, Automation and Systems (ICCAS),
2014 14th International Conference on
(p. 814-818). IEEE.
GANESHMURTHY, S.; NASSAR, S. A.
(2014) Finite element simulation of process control for bolt tightening in
joints with nonparallel contact. Journal
of Manufacturing Science and Engineering, v. 136, n. 2, 021018.
HERMANSSON, T. O. (2016) Quality
assured tightening of screw joints. Proceedings of the Institution of
Mechanical Engineers, Part C: Journal of
Mechanical Engineering Science, v. 230, n. 15, p. 2588-2594.
LI, S.; HAN, M.; DENG, L. (2009)
Research on CNC machining screw which is variable pitch & groove depth
& groove width. In Computer-Aided
Industrial Design & Conceptual Design, 2009. CAID & CD 2009. IEEE 10th
International Conference on (p. 710-713). IEEE.
LIU, H. C.; LIU, L.; LIU, N. (2013)
Risk evaluation approaches in failure mode and effects analysis: A literature
review. Expert systems with
applications, v. 40, n. 2, p. 828-838.
LIU, S.; GE, S. S.; QIN, G.; LI, M.
(2015) An Automatic Screw Tightening Shaft Based on Enhanced Variable Gain PID
Control. International Journal of
Simulation--Systems, Science & Technology, v. 16, n. 5.
MUELANER, J. E.; CAI, B.;
MAROPOULOS, P. G. (2010) Large-volume metrology instrument selection and
measurability analysis. Proceedings of the Institution of Mechanical Engineers,
Part B: Journal of Engineering
Manufacture, v. 224, n. 6, p. 853-868.
OGAWA, S.; SHIMONO, T.; KAWAMURA,
A.; NOZAKI, T. (2015) Position control in normal direction for the fast
screw-tightening. In Industrial
Electronics Society, IECON 2015-41st Annual Conference of the IEEE (p.
003429-003433). IEEE.
OGUSHI, K.; NISHINO, A.; MAEDA, K.;
UEDA, K. (2015) Direct calibration chain for hand torque screwdrivers from the
national torque standard. ACTA IMEKO,
v. 4, n. 2, p. 32-38.
PERSSON, E.; ROLOFF, A. (2016)
Ultrasonic tightening control of a screw joint: A comparison of the clamp force
accuracy from different tightening methods. Proceedings of the Institution of
Mechanical Engineers, Journal of
Mechanical Engineering Science, v. 230, n. 15, p. 2595-2602.
SAFARI, H.; FARAJI, Z.; MAJIDIAN, S.
(2016) Identifying and evaluating enterprise architecture risks using FMEA and
fuzzy VIKOR. Journal of Intelligent
Manufacturing, v. 27, n. 2, p. 475-486.
SCHNEIDER, H. (1996) Failure mode and effect analysis: FMEA from theory to execution. Technometrics, v. 38, n. 1, p. 80.
TRIPP, D. (2005) Pesquisa-ação: uma introdução metodológica. Educação e pesquisa, v. 31, n. 3, p. 443-466.
WEN, K.; DU, F.; ZHANG, X. (2016)
Algorithm and experiments of six-dimensional force/torque dynamic measurements
based on a Stewart platform. Chinese
Journal of Aeronautics, v. 29, n. 6, p. 1840-1851.
WOLF, C. M.; LORENZ, R. D. (2011)
Using the motor drive as a sensor to extract spatially dependent information
for motion control applications. IEEE
Transactions on Industry Applications, v. 47, n. 3, p. 1344-1351.