Downloaded from orbit.dtu.dk on: May 27, 2020
Acoustic emission monitoring of the bending under tension test
Moghadam, Marcel; Sulaiman, Mohd Hafis Bin; Christiansen, Peter; Bay, Niels Oluf
Published in:
Procedia Engineering
Publication date:
2017
Document Version
Publisher's PDF, also known as Version of record
Link back to DTU Orbit
Citation (APA):
Moghadam, M., Sulaiman, M. H. B., Christiansen, P., & Bay, N. O. (2017). Acoustic emission monitoring of the
bending under tension test. Procedia Engineering, 207, 1421–1426.
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright
owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
You may not further distribute the material or use it for any profit-making activity or commercial gain
You may freely distribute the URL identifying the publication in the public portal
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
ScienceDirect
ScienceDirect
Available online at www.sciencedirect.com
ScienceDirect
Procedia Engineering 207 (2017) 1421–1426
International Conference on the Technology of Plasticity, ICTP 2017, 17-22 September 2017, Cambridge, United Kingdom
Acoustic emission monitoring of the bending under tension test
Marcel Moghadama*, Mohd Hafis Suleimana,b, Peter Christiansena, Niels Baya
a
Department of Mechanical Engineering, Technical University of Denmark, 2800 Kongens Lyngby, Denmark
b
Mechanical Engineering Programme, Universiti Malaysia Perlis, 02600 Arau, Malaysia
Abstract
Preliminary investigations have shown that acoustic emission has promising aspects as an online monitoring
technique for assessment of tribological conditions during metal forming as regards to determination of the onset of
galling. In the present study the acoustic emission measuring technique has been applied for online monitoring of
the frictional conditions experienced during Bending Under Tension (BUT) testing. The BUT test emulates the
forming conditions experienced when drawing sheet material over a die curvature as in deep drawing processes.
Monitoring of the developed acoustic emission in BUT testing has been found to describe the frictional conditions
during forming well and to allow for accurate assessment of the limits of lubrication.
© 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the International Conference on the Technology of Plasticity.
Keywords: Acoustic emission monitoring; Tribology; Bending under tension testing;
Nomenclature
AE
Acoustic emission
BUT
Bending under tension
FFT
Fast Fourier transform
Fb
Back tension force
Ff
Drawing force
T
Torque
*Corresponding author. Tel.: +4524244765
E-mail address: marmog@mek.dtu.dk
1877-7058 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the International Conference on the Technology of Plasticity.
10.1016/j.proeng.2017.10.907
1422
Marcel Moghadam et al. / Procedia Engineering 207 (2017) 1421–1426
1. Introduction
Galling is a wear phenomenon commonly encountered in various sheet metal forming processes, often occurring
when working with tribologically difficult materials such as stainless steel, Al- or Ti-alloys. The galling mechanism
is characterized as a localized macroscopic transfer of metallic material between contacting surfaces where surface
damage progressively increases in severity. The occurrence of galling is a major concern in sheet stamping industry,
since it leads to poor surface quality, production stops and tool deterioration. It is reported that galling wear accounts
for up to 71% of the cost of die maintenance [1][2]. Online process monitoring of sheet metal forming operations is
therefore of significant interest in order to actively minimize the occurring wear in production. Since the galling
mechanism occurs in the contacting interface between the sheet metal and the forming die, the requirement for
online process monitoring for sheet metal forming calls for an indirect measuring technique with sufficiently high
sensitivity to detect any minute differences in the process. Conventional techniques, e.g. measurement of the
forming force, have typically been found to have too limited sensitivity for assessment of the limits of lubrication
[3]. Measurement of acoustic emission (AE) has, on the other hand, been successfully applied as an online
monitoring technique for evaluation of different tribological characteristics in different metal forming operations
[4]-[7]. Behrens et al. [8] have concluded that acquisition and analysis of AE signals allows for online assessment of
production conditions and deviations in production processes. Skåre [9] similarly noted that energy analysis of the
AE signals allows for an evaluation of the quality of lubrication and detection of process defects in sheet forming
such as the stick-slip effect and cracking. Mostafavi and Pashmforoush [4] successfully applied the AE technique
for detecting the onset of galling in a slider-on-sheet test, noting a direct relation between the wear mechanism and
AE peak amplitudes.
Preliminary investigations by the present authors on detecting the onset of galling in strip reduction testing [11]
have similarly shown a correlation between the development of the signal and the generated surface roughness,
allowing for an assessment of the severity of the wear mechanism by comparing the developed AE signal with a
reference threshold value. The strip reduction test is, however, a tribologically severe sheet forming test, which due
to the large normal pressures, surface expansion and increase in temperature during testing results in severe galling
in case of insufficient lubrication. For such cases the AE technique proves sound. In case of milder tribological tests
e.g. strip drawing and bending under tension testing emulating deep drawing, where minor levels of galling may
appear, investigations are required for characterization of the correlation between the frictional conditions and the
generated AE signals. The aim of the present study is therefore to explore the possibility to apply the AE technique
for tribological tests where mild galling occurs.
2. Experimental procedure
2.1 Bending under tension test
The BUT tests are conducted on a preexisting test setup, initially described by Andreasen et. al. [10] shown in figure
1. The test replicates the tribological conditions experienced in the die curvature in a deep drawing process by
drawing plane strip material over a die shoulder with superimposed back tension.
A
Drawing force
tension
B
Die shoulder
Drawing force
tension
Tool pin with torque measurement
AE sensor
Torque
transducer
Back tension
Fig. 1: Schematic illustration of the BUT test principles (A). Experimental BUT test setup (B).
Marcel Moghadam et al. / Procedia Engineering 207 (2017) 1421–1426
1423
The BUT test was conducted by drawing 800x30x1 mm strips of 304L stainless steel over a fixed Ø10mm tool of
powder metallurgical high-speed steel DIN. W. No. 1.3344 PM, polished to a surface roughness Ra = 0.1. Constant
back tension force Fb = 6 kN was applied with a drawing length of 150mm, while the drawing speed was
approximately 300 mm/s. During the BUT test the horizontal drawing force, Ft and the torque on the tool pin was
measured with piezoelectric transducers. Two separate test series were carried out for assessment of the accumulated
wear occurring over time. They both included consecutively drawing of three steel strips on the same tool pin
surface. The strips in the two test series were lubricated with two different, plain mineral oils, CR5 Houghton
Plunger with a kinematic viscosity 𝜂𝜂 = 660 cSt at 40°C and 50/50 wt.% CR5/Sun60, respectively. Sunoco Sun60
has a viscosity 𝜂𝜂 = 10 cSt at 40°C. The mixture yields a viscosity of 60 cSt at 40°C. It was attempted to apply the
same amount of lubricant on each strip. However this may be difficult to achieve in practice and is therefore a
possible source of error.
Upon measurement of the drawing force and the torque during testing, the average friction stress and normal
pressure in the tool/workpiece contact area can be determined.
Friction stress:
𝜏𝜏 =
Normal pressure:
𝑝𝑝 =
2𝑇𝑇
𝜋𝜋𝜋𝜋𝑅𝑅2
(1)
𝐹𝐹𝑓𝑓 + 𝐹𝐹𝑏𝑏
2𝜋𝜋𝑅𝑅
(2)
2.2 Acoustic emission data acquisition
Data acquisition of the generated AE signals during BUT testing were made with a R15α piezoelectric sensor
connected to a 2/4/6 analog preamplifier with a built-in bandpass filter of 10-900 kHz. The piezoelectric AE sensor
was fixed to the pin tool holder with a clamping tool as close as possible to the tool pin. The contact area between
sensor and tool holder was provided with grease paste as an acoustic couplant. The preamplifier was connected to a
NI 9223 multipurpose data acquisition unit which was fixed to a NI 9146 4-slot expansion chassis. A sampling rate
of 1 MHz with resolution of 16-bit was selected in order to suppress aliasing of the signal and any possible loss of
signal amplitude.
3. Experimental results
3.1. Surface structure after BUT testing
The generated surface structure of the strips after BUT testing can be seen in figure 2. SEM images of the tested
workpiece strips are shown in detail in figure 3. The legends A-D in figure 3 refers to the tests and strip locations
shown in figure 2.
B
CR5/Sun60
CR5
A
Strip 1
Strip 2
Strip 3
D
C
Fig. 2. Overview of the surface structure of the strips upon bending under tension testing with the CR5/Sun60 and CR5 lubricants.
1424
Marcel Moghadam et al. / Procedia Engineering 207 (2017) 1421–1426
A
B
C
D
Fig. 3: SEM images of the surface structure upon BUT testing. (A) Minor levels of initial abrasive wear. (B) The onset of galling. (C) Galling
with increased severity. (D) Completely smooth surface structure.
The SEM images in figures 3A, 3B and 3C illustrate the development of galling mechanism during BUT testing
with the low viscous oil. Upon initial drawing of the strip with insufficient lubrication minor levels of scoring are
observed as seen in figure 3A. At the onset of galling, local pickup on the tool has occurred leading to subsequent
scoring of the strip surface upon further drawing, as seen in figure 3B. Upon further drawing of the strip material,
the local pickup on the tool increases and the strip is scored more heavily as seen in figure 3C. Figure 3D represents
an experiment with no galling obtained by running with the high viscous CR5 oil.
3.1 Drawing force and torque measurements
In figure 4 the developed drawing force and torque is shown as a function of the accumulated drawing lengths in the
BUT test for the two lubricants. Minor fluctuations are noted in the force and torque measurements in the start and
end of each dataset, due to start and stoppage of the drawing sequence [10].
Lubrication with CR5/Sun60 shows a gradual increase in the torque, which is due to the increased galling, since
this phenomenon is directly related to the friction on the tool pin as shown in eq. 1. The measured drawing force is
noted to have a significantly lower sensitivity towards changes in frictional conditions during the test than the
torque. Previous BUT tests have [10] similarly found that the torque is more sensitive to changes in friction and
enabling accurate assessment of lubricant film breakdown, while the measured drawing force generally has very low
sensitivity for assessment of galling, when this is mild. The first strip within this test series is noted to have a stable
development of the torque in the beginning, where only mild abrasive wear is observed on the strip surface.
However at a drawing length of approximately 50mm a rampant development of the torque initiates. This point
coincides with localized development of friction junctions in the surface structure, which indicates the initial stage
of galling. At further drawing progressive scoring of the strip surfaces appears. The test series lubricated with the
high viscosity CR5 lubricant is on the other hand found to have completely stable development of the drawing force
and torque, with a significantly lower torque value. This corresponds to the smooth surface structure obtained in this
test series with no sign of galling.
Marcel Moghadam et al. / Procedia Engineering 207 (2017) 1421–1426
(a)
(b)
12500
12500
7,5
7,5
Strip 2
Strip 1
Strip 3
Strip 2
Strip 1
Strip 3
5000
5,0
Force (N)
5,0
7500
7500
5000
2,5
Torque (Nm)
10000
Torque (Nm)
10000
Force (N)
1425
2,5
2500
2500
0
0
50
100
150
200
250
300
350
400
0
0,0
450
0
Accummulated drawing length (mm)
50
100
150
200
250
300
350
400
0,0
450
Accummulated drawing length (mm)
Fig. 4. Drawing force and torque experience during the BUT test lubricated with (a) CR5/Sun60, (b) CR5.
3.2 AE measurements
The instantaneous power of the acquired AE signal during the BUT test is displayed in figure 5. The red the AE
signal which represents the test on CR5 lubricant, shows a very low and stable signal. This is due to the efficient
lubrication and the resulting low frictional forces between workpiece and tool pin. The tested strip surface is smooth
and no pick-up appears on the tool surface. The much stronger and fluctuating blue AE signal represents the test on
CR5/Sun60 lubricant. The strong signal is due to thinning of the lubricant film, which causes increased friction and
mild abrasive wear at first, followed by onset of galling at approximately 50 mm drawing length as noticed in the
SEM images. Upon further drawing of the strip, the increased galling is reflected in the increased amplitude of the
AE signal, which in general follows a similar trend as the measured torque. Comparison of the two test series with
the two different lubricants thus indicates a very high signal to noise ratio. This allows for a simple way of detecting
galling by acoustic emission using simple time domain parameters.
Regarding the blue AE signal in figure 5 for the strip lubricated with Cr5/Sun60, several local minima are
observed for instance at 118, 200, 230 and 265mm drawing length with corresponding maxima in between, which
are not all reflected in the torque signal. This is explained by the stick-slip occurring due to galling, which releases
sudden signals, when local adhesion junctions between tool and workpiece are breaking. The torque measurement
may not be able to capture these sudden fluctuations. The results, however, clearly indicate a strong correlation
between galling and the transient stress waves acquired by the piezoelectric AE sensor. The limits of lubrication can
thus be determined by evaluating the development of the acquired AE signal during testing. In order to implement
the technique to manufacturing production proper choice of signal processing should be based upon the
characteristics of the monitored process and the overall objective of monitoring. A few studies [4], [11] have
however found that the galling wear occurring in metal forming production can be evaluated by assessment of
simple AE parameters e.g. RMS and count-rate.
Strip 1
Strip 2
Strip 3
Fig. 5. Power of the generated AE signal throughout the BUT test.
1426
Marcel Moghadam et al. / Procedia Engineering 207 (2017) 1421–1426
A fast Fourier transform (FFT) showing the overall frequency components of the signals is displayed in figure. 6.
From the frequency representation of the two signals, a major frequency component can be found when lubricating
with CR5 at a frequency of 40 kHz. Several minor peaks can similarly be identified, however these are likely due to
influencing noise factors, due to the low amplitudes. The frequency composition in the test setup lubricated with
CR5 therefore characterises the naturally occurring frequencies with the specific test setup without the occurrence of
galling. However, when galling occurs in the BUT test, a larger amplitude of the frequency spectra is noticed. This
is seen as a significant amplitude increase in the measured frequencies in the range of 20-160 kHz. The change in
the frequency spectrum can therefore be attributed to the occurrence of galling, since the spectrum increases
progressively in amplitude as the wear mechanism develops.
CR5/Sun60
CR5
0,005
Amplitude
0,004
0,003
0,002
0,001
0,000
0
100
200
300
400
500
Frequency (kHz)
Figur 6. Frequency domain representation of the signal generated in with the two lubricants.
4. Conclusion
The AE measuring technique has been found to be a powerful tool for online monitoring of galling during BUT
testing. By measuring the generated AE assessment of the limits of lubrication can be made and even very mild
levels of wear can be detected. The generated AE was found to be correlated to the severity of galling. This was
evaluated by visual inspection of the generated surface structure as well as by assessment of the friction stress
experienced during testing. The generated AE signal caused by galling was furthermore found to occur especially in
the low frequency range of 20-160 kHz. Assessment of the development of the captured AE signal therefore allows
for an online evaluation of galling in sheet stamping production with relatively cheap and simple measuring
equipment. This can prove valuable for process monitoring e.g. in complex forming dies where torque
measurements cannot be applied.
5. References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
E. Van der Heide, A.J. Huisin't Veld, D.J. Schipper, The effect of lubricant selection on galling in a model wear test,
Wear. 250 (2001) 973–979.
S. Hogmark, B. Podgornik, J. Viz, Improvement in galling performance through surface engineering, Surf. Eng. 22
(2006) 235–238.
C. Teymuri Sindi, M. Ahmadi Najafabadi, M. Salehi, Tribological behavior of sheet metal forming process using
acoustic emission characteristics, Tribol. Lett. 52 (2013) 67–79.
R.F.Z. S. Mostafavi, C.T. Sindi, F. Pashmforoush, Acoustic Emission Waves from the Onset of Galling between Tool
and Sheet Material, Mater. Eval. 71 (2013) 1335–1342.
T. Skåre, F. Krantz, Wear and frictional behaviour of high strength steel in stamping monitored by acoustic emission
technique, Wear. 255 (2003) 1471–1479.
A. Hase, H. Mishina, M. Wada, Correlation between features of acoustic emission signals and mechanical wear
mechanisms, Wear. 292–293 (2012) 144–150.
P. Tian, Y. Tian, L. Shan, Y. Meng, X. Zhang, A correlation analysis method for analyzing tribological states using
acoustic emission, frictional coefficient, and contact resistance signals, Friction. 3 (2015) 36–46.
C.B. Behrens, B.-a., I. El-galy, T.Huinik, Online monitoring of deep drawing process by application of acoustic
emission, Steel Res. Int. (2011) 385–389.
T. Skåre, Dynamically loaded tribosystems in plastic forming operations, dissertation, Lund University (2001)
J.L. Andreasen, D.D. Olsson, K. Chodnikiewicz, N. Bay, Bending under tension test with direct friction measurement,
Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 220 (2006) 73–80.
M. Moghadam, P. Christiansen, N. Bay, Detection of the onset of galling ins strip reduction testing using acoustic
emission, Proc. Eng. (2017).