Open Journal of Safety Science and Technology, 2011, 1, 89 -93
doi:10.4236/ojsst.2011.13009 Published Online December 2011 (http://www.SciRP.org/journal/ojsst)
Fail-Safe Desi gn b y O u te r Cover o f Hi g h Pressu re Vessel
for Food Processing
Yuichi Otsuka1, Jyunpei Fuj ii2, Masaki Takato2, Yoshiharu Mutoh3
1Top Runner Incubation Center for Academia-Industry Fusion, Nagaoka University
of Technology, Niigata, Japan
2Department of System Safety, Graduate School of Management and Technology, Nagaoka
University of Technology, Niigata, Japan
3Department of System Safety, Nagaoka University of Technology, Niigata, Japan
Received September 26, 2011; revised November 2, 2011; accepted November 10, 2011
Abstract
This paper demonstrates the effectiveness of outer-cover of pressure vessel in order to prevent risk scenario
by brittle fracture. The Fail-Safe design introducing the concept of a standby redundant system is applied to
the high-pressure vessel so that the safety of the operator can be assured in case of the vessel fractures. Based
on the limit state design (LSD) concept, a fracture mode is predicted as plastic collapse and brittle fracture.
Dimensions of a cover are determined by considering failure modes of plastic collapse and brittle fracture.
Failure experiment with pre-cracked vessel can clearly show the effectiveness of the proposed vessel for
fail-safe design.
Keywords: Food Processing, Pressure Vessel, Fail-Safe Design, Safety Engineering
1. Introduction
It is reported in a previous study [1] that the high pre-
ssure processing of foods will produce a variety of ad-
vantageous effects such as enhancement of sterilization
force and gustatory sense, and condensation of an “uma-
mi” (good) flavor. In this study, light and safe high-
pressure processing equipment is being developed, assum-
ing that the equipment will be used in small-to-medium
businesses. In most conventional high-pressure process-
ing equipment, high pressure is created directly with
such devices as a piston but they have a deficiency in
that the size of the entire equipment tend s to be large. In
order to mitigate this deficiency, indirect pressurization
and sealed systems are proposed [2] as small and light
high-pressure processing systems. In the indirect pre-
ssurization system, high pressure is created with the help
of a pressure intensifier; in the sealed system, a locking
bar type pin, such as the one shown in Figures 1 and 2,
is used. With the adoption of this sealed system, the
high-pressure vessel can be light. Since the load is sup-
ported at the corners on the locking bar guide (or lever
guide in Figure 2), however, the stress will be concen-
trated on the corners, possibly resulting in rapid fracture
in the early phase of operations. To resolve this stress
concentration issue, we optimized, from the strength re-
liability point of view, the shape of the seal section in
order to mitigate the concentration. However, it is im-
possible, only from the reliability point of view, to ra-
tionally derive countermeasures to assure the safety of
the operator against fracture of the equipment.
In this study, the Fail-Safe design introducing the con-
cept of a standby redundant system [3] is applied to the
high-pressure vessel so that th e safety of the op erator can
be assured in case of the vessel fractures, and the intrin-
sic safety feature can be implemented in the equipment.
Based on the limit state design (LSD) concept, a fracture
mode is predicted, and countermeasures against the frac-
ture mode are investigated. From the results of th e inves-
tigation, fracture testing of the pre-cracked high-pressure
vessel is carried out and the resulting effectiveness of the
concept is confirmed and reported in this paper.
2. Fail-Safe Design by Outer Cover of
Pressure Vessel
2.1. Installation of Outer Cover
As illustrated in Figure 3(a), an outer cover is installed
on the outside of the pressure vessel. The cover bears no

Y. OTSUKA ET AL. 90
Figure 1. Pin-arm sealed pressure vessel structure for food
processing.
Figure 2. Upper view of pressure vessel.
load as long as the equipment is operating normally.
Only in case of fracture of the locking bar guide (lever
guide), it bears the load and prevents fragments of the
fractured guide from being scattered. This cover can be
regarded as a sort of standby redundant system because it
does not function under the normal operation condition s.
In the following sections, the structure of the cover pre-
venting fragments from being scattered is studied in
terms of avoiding plastic collapse and rapid fracture ori-
ginating from the stress concentration area.
2.2. Calculation of Penetration Thickness
Caused by Plastic Collapse
As illustrated in Figure 3(b), the internal energy of high-
pressure liquid is converted into the dynamic energy of
objects generated by and scattered from the locking bar
Figure 3. Bullet model of fracture of protective cover.
guide fracture. The following equation is used to calcu-
late the magnitude of energy that causes the plastic de-
formation.
1B
hPAd A
2


(1)
The following figures are entered into the Equ ation (1)
to calculate the energy: α: Correction factor for energy
conversion efficiency (α = 1 assuming the worst case), P:
Internal pressure (P = 2.0 × 108 [N/m2] ) m: Sum of
arm mass and cover mass (m = 14.2 [kg]), A1: Cross sec-
tional area of cover (A1 = 0.01606 [m2]) A2: Contact area
of arm and cover (A2 = See Figure 4 [m2] ), and σB: Ten-
sile strength of SCM435 (σB = 9.32 × 108 [N/m2]). The
calculated penetration thickness h is listed in Table 1 as
a function of width w and space d.
Here we use the following equation to estimate whe-
ther impact energy can be absorbed while cracks caused
by the impact in the maximum stress concentration area
are being propagated.
5
110
8
PA w
dEl



 (2)
where, νE is the Charpy impact value (νE = 200 [J /cm2])

Y. OTSUKA ET AL.91
Figure 4. Definition of contacting area A2.
Table 1. Calculated thickness h [mm] by Equation (1) when
the failure mode “plastic collapse” occurs from the R cor-
ner of arm guide (red circle of Figure 1).
Space d [mm]
Width w [mm] 0.1 0.2 1 2 5
1 1.9 3.7 18.5 37 92.5
5 0.4 0.7 2.6 7.4 18.5
10 0.23 0.4 1.9 3.7 9.3
15 0.1 0.2 1.2 2.5 6.2
20 <0.1 0.2 0.9 1.9 4.6
and l is the length of the crack formed near the maximum
stress concentration area. The coefficient 4 is used when
the impact is applied only on one end (root) of the arm,
and the coefficient is 8 when the impact is applied on
both ends of an arm (i.e., on 8 locations (= 4 locations/
end × 2 ends)). The calculated width (w) is listed in
Table 2 as a function of the space (d) and length (l) of
the formed crack.
Dimensions are determined by the following pro cess.
1) From the fracture toughness value KIC and maxi-
mum stress value at the R corner, absorbing length l
should be less than 10 mm ( green colored columns in
the Tables 1 and 2)
2) From the constraint condition that the cover never
contacts to the top of inner pressure vessel at the maxi-
mum pressuring condition, space d should be more than
0.9 mm (blue colored columns in the Tables 1 and 2).
3) From the cost effectiveness condition, minimum va-
lue in the pink colored columns in Ta ble 2 is selected, w
is about 20 mm.
4) From the cost effectiveness condition, minimum va-
lue in the yellow colored co lumns in Table 1 is selected,
h is about 0.9 mm.
Finally selected values are emphasized by italic and
Table 2. Calculated width w [mm] by Equation (2) when the
failure mode “brittle fracture” occurs from the R corner of
arm guide (red circle of Figure 1).
Absorbing length l [mm]
Space d [mm] 1 2 5 10 20
0.1 20.1 10 4 2 1
0.2 40.2 20.1 8.3 4 2
1 200.8 100.4 40.2 20.1 10
2 401.5 200.8 80.3 40.2 20.1
5 1003.8501.9 200.8 100.4 50.2
underlined letters in the tables. Finally selected values
are emphasized by italic and underlined letters in the
tables. By considering the safety coefficient for the va-
lues of d and h, d should be more than 1.3 mm and h
should be more than 1 mm.
3. Verifications of Fail-Safe Design of
Pressure Vessel by Fracture Test
On the corners of the lever guide in both a pressure ves-
sel with the outer cover (double-wall vessel) and a vessel
without the outer cover (single-wall vessel), pre-cracks
of 5 mm in length were machined and the pre-cracked
pressure vessels were subjected to fracture testing. In the
testing, the pressure applied was increased by 25% to
250 MPa. Refer to Figure 1 for the corners (R sections),
each of which is marked with a circle. For safety precau-
tions during testing, all the equipment was enclosed in a
protective box.
As shown in Figures 5 and 6, fracture testing of the
single-wall vessel resulted in a broken guide, the inside
surface had 45 degrees of slant angle and the outside
surface was flat. We calculated the stress distribution
around the R corners in maximum pressuring by using
the finite element analyses (FEA) in ref. [2]. The direc-
tion of fractured surface is almost vertical to the direction
of the principle component of stress, and it can be esti-
mated that the fracture was brittle fracture because the
principal normal stress can normally cause the brittle fra-
cture.
Figure 7 shows the arm guide contacting condition
before test for a v essel with outer cover. Figure 8 show s
the broken arm guide for the vessel with outer cover.
Angle of cracking from the R corner of arm guide is de-
creased from 45 degree as shown in Figure 8. The de-
crease of angle means the arm guide and the seal is con-
tacting the outer cover in cracking and then the force
from pressure 250 MPa is distorted into the outer cover
that would result in the decrease of maximum stress va-
lue at the R corner. The outer cover could successfully
prevent the broken arm guide from flying.

Y. OTSUKA ET AL. 92
Figure 5. Flying parts of broken arm guide (The top part of
Figure 1). Angle of fracture surface is about 45 degree (left
and right side).
Figure 6. Remained parts of arm guide (The part shown in
Figure 1). Fracture surface shows smooth brittle features.
Figure 7. Arm guide contacting condition before test for a
vessel with outer cover.
Figure 8. Broken arm guide for the vessel with outer cover.
Angle of fracture surface is suppressed by the support from
the outer cover during fracture.
Figure 9 shows the broken roof of the protective box
by the flying parts (Figure 5) in the fracture case of the
vessel without outer cover. Figure 10 shows the pre-
served roof of the protective box in the fracture case of
the vessel with outer cover. Figure 9 can easily demon-
strate the critical risk of brittle fracture which the pene-
trated thickness of the steel plate is 6 mm. In the case of
Figure 10, which the fracture test case for the vessel
with outer cover, the plate at the roof is not damaged by
the fracture and only wetted by water flow from the
cracking part of the arm guide. The risk of wetting by
water flow is not serious and can be acceptable for users
compared with the critical damage shown in Figure 9.
Figure 11 shows change of strain during fracture pro-
pagation. In the course of measurement, fracture started
when the strain signals reaches maximum (that means
strain signal was lost), which implies that the fracture
began from the outside. From the FEA result of the arm
guide [2], maximum stress is on the inner side of the arm
guide, which starting point of brittle fracture in the ex-
periment is not corresponding. The FEA result can show
that the maximum stress point moves from inner side to
Figure 9. Broken roof of the protective box by the flying
parts (Figure 5) in the fracture case of the vessel without
outer cover.
Figure 10. Preserved roof of the protective box in the frac-
ture case of the vessel with outer cover.

Y. OTSUKA ET AL. 93
Figure 11. Change of strain during fracture propagation.
outer side when the sliding of the arm or cleara nce of th e
arm is considered. This analysis indicates the necessity
of improvement of contacting conditions between arm
and arm guide.
4. Conclusions
A double-wall high-pressure vessel in which the concept
of the standby redundant system is introduced was pro-
posed. The dimensions of the outer cover were decided
by calculations that take the fracture mode into account,
and fracture testing verifies the effectiveness of the outer
cover. This research was financially supported by JST
project “Development of Technology for Promoting
Food Quality”.
5. References
[1] Development of basic technologies available to Enhanc-
ing Value-Added to Food.
http://www.nico.or.jp/ create/gaiyou/index.html
[2] H. B. Haron, Y. Mutoh, Y. Otsuka and Y. Miyashita,
“Selection of Stress Relief Grooves On Pressure Vessels
for Food Processing,” Proceedings of JSME Hokuriku
Sinetsu Branch 48th General Conference and Lecture
Meeting CD-ROM (No. 616), Ueda, 8 March 2011.
[3] M. G. Stewart and R. E. Melchers, “Probabilistic Risk
Assessment of Engineering Systems,” Morikita Publish-
ing Inc., Tokyo, 2003.