Design of Quadruple Offset Butterfly Valve Used in Power Plants

Abstract

The elliptical disc is a specific characteristic of a triple offset butterfly valve (TOBV), but the structure of the elliptical disc leads to a non-uniform distribution of contact at the closing position with non-uniform wear. On the other hand, the continuous processing of elliptical sections to finish the entire disc contour is expensive compared to the processing of a circular section. In the present study, to improve the above disadvantages of an elliptical disc, an elliptic cone function was established to describe the quadruple butterfly valve (QOBV) disc shape, which has a circular section disc. In addition, some designs of a fixed thickness (7 mm) laminated seal for good sealing were proved. Sealing performance simulations were carried out using the thermal–structural coupling analysis. The best design of the QOBV showed that the maximum contact pressure was 37.2% lower than the TOBV using the seven-way laminated seal layers (thickness of A240–316 stainless and graphite was 1.3 mm and 0.6 mm, respectively). Compared to TOBV, the QOBV disc operation during the opening and closing process was faster and smoother. The disc can be applied widely to various industrial fields, engine exhaust systems, and turbine and power plants under severe environments.

1. Introduction

The eccentric butterfly valve has been used in industrial transportation systems, engine exhaust systems, turbines, and power plants to control the flow rate at the required temperature and pressure of working fluid [1,2,3,4] for a long time. The flow rate is controlled by a disc that rotates from 0 degrees (closed position) to 90 degrees (opened position). Such rotation values depend upon torque value [5,6,7]. The unreasonable torque value, corrosion, and high temperature would significantly affect valve sealing performance. The seal must maintain for their integrity, which is critical for high-risk energy production and oil processing applications. Therefore, the seal part must be considered by different working environments.

The different types of butterfly eccentric valves use different seals; at present, three eccentric butterfly valve types are accepted, single offset, double offset, and triple offset [8,9]. The single and double offset butterfly valves use a soft seal (rubber seal) at low temperature working environments, although the rubber seal has interference with the disc for good sealing and the needless interference leads to more wear [10,11]. The triple offset butterfly valve (TOBV) uses a metal seal and can work in high temperature working environments. Almost no contact happens between the disc and seat (seal). However, the disc of the TOBV has an ellipse section, which leads to nonuniform contact pressures distribution in the seat (seal) compared with the circular section, and the manufacturing cost is improved. On the other hand, the disc section area is less than a single or double offset butterfly valve in the same contacted pipe size, which means the flow rate is reduced. To solve the disadvantage of the TOBV, the quadruple offset butterfly valve (QOBV) was developed as a new concept, even though it is not accepted in standards such as API-609 standard. Additionally, its sealing problem would be considered at same time.

In previous studies, Kim reports a numerical study of the flow behavior through TOBV. The effect of the turbulence model was examined with consequent flow behavior analysis around the valve with turbulence variables such as turbulence kinetic energy and Reynolds stress [12]. Wang studied the high-pressure and high-temperature steam-induced vibration in a triple-butterfly valve. The steam pressure fluctuation characteristics of the disc were studied based on the numerical simulation results. The natural frequency of the disc was tested through the experimental modal analysis method to determine the vibration frequency characteristics of various orders [13]. Naragund found three important torques, seating/unseating, bearing friction, and hydrodynamic torque, for discs with double offset by carrying out the experiment on a butterfly valve, and the total torque obtained was compared using CFD simulation software and was validated. Further, the double offset disc was changed to single, triple, bi-lattice, and tri-lattice. A correlation is established between the experimental value and the CFD values [14]. Zhang developed an experimental test system for cryogenic high-speed hydrodynamic non-contact mechanical seals. The performances of seals under different working conditions were studied. The main performances of the seals were obtained, and the relationships of the performances with the inlet fluid pressure, the closing force, and the rotational speed were discussed [15]. Yang improved the structure of radial metal seal of the second-generation ICV, and the related theoretical model of its contact mechanical behaviors was also established. Theoretical and numerical analyses were carried out, which provides methods and means for the selection of application types and the improvement of structural parameters in intelligent completion operation [16]. Inose investigated changes in the sealability of a greased metal-to-metal seal due to elevated temperature by fundamental gas tightness tests. Changes in grease property at high temperature were investigated. It was found that the causes of the loss of the sealability of the metal-to-metal seal at a high temperature was the decrease in grease viscous resistance against high pressure due to the thermal degradation of grease [17].

In this present work, to improve the above disadvantages of TOBV, an elliptical disc, an elliptic cone function was established to describe the quadruple butterfly valve (QOBV) disc shape which has a circular section disc. On the other hand, the laminated seal used in the QOBV is suggested for a good sealing, and some designs of laminated seal had a fixed entire thickness (7 mm). Additionally, sealing performance simulations were carried out using thermal-structural coupling analysis. As the results show, the best design of the QOBV shows that the maximum contact pressure is 37.2% lower than the TOBV using 7-way laminated seal layers which have a thickness of A240–316 stainless 1.3 mm and graphite 0.6 mm, and shown in Figure 1.

2. Design of QOBV

2.1. Structural of QOBV

The QOBV is composed of a body, disc, seal, and shaft, as shown in Figure 2. The shaft is located a little to one side of the sealing center, above the sealing plane. Its rotation center also deviates from the cone axis, extending from the sealing surface. When the valve is closed, the seal and the surface of the valve disc completely contact. Opening the valve will immediately move the valve disc away from the sealing part at all points, thus minimizing the seal wear caused by the non-friction sealing surface, and thus extending the service life and closing tightly. The fourth offset characteristic in QOBV is shown in Figure 3.

First offset: The shaft deviates from the sealing plan to offer an uninterrupted sealing surface.

Second offset: The center line of the valve disc is away from the shaft’s center line, which allows the valve to leave the seal freely during the opening process.

Third offset: The center line of the conical surface is offset from the center line of the valve body (γ°) to remove the elliptical contact surface of the disc. This enables non-contact and no interference during opening and closing.

Fourth offset: The circular cone of TOBV is changed to an elliptical cone of QOBV to obtain a circular disc.

2.2. Cone Surface Equation of QOBV

To understand QOBV characteristics and differences with TOBV, two kinds of cone surface functions were used: the elliptical cone function used in QOBV and the circular cone function used in TOBV [18], shown in Equations (1) and (2). ‘a’ and ‘b’ are major and minor axes of the elliptical cone, respectively, and ‘c’ is the height of a cone.

$$\frac{x^2}{a^2}+\frac{y^2}{b^2}=\frac{z^2}{c^2}$$

(1)

$$x^2+y^2=k^2{z}^2,\mathrm{k}=\frac{\mathrm{a}}{\mathrm{c}}$$

(2)

When the ellipse cone function of QOVB and the circular cone function of TOBV overlap in the xz plane (y = 0), as shown in Figure 4, the new ellipse cone function, Equation (3), can be defined by substituting Equation (2) into Equation (1).

$$x^2+\frac{a^2}{b^2}{y}^2=k^2{z}^2$$

(3)

The sections of the QOBV disc ($S_i$) in Equations (4)–(6) depend on $H_i$ ($\mathrm{i}=1,2,\mathrm{and}3$), as shown in Figure 5. ‘$S_1$’ and ‘$S_2$’ are the upper and lower sections of the disc, and $S_0$ is a middle section of the disc center. ‘$H_0$’ is a displacement of the center of the disc from the origin of coordinates. ‘$H_1$’ and ‘$H_2$’ are displacements of the center of the disc top (BC) and bottom (AD) from the origin of coordinates. ‘$N_i$’ is the intersection between the cone center, and ‘$S_i$’, ‘$k_0$’ is the substituted for tan γ, and ‘γ ‘is the offset angle of the axis (z) from the center line of the disc. ‘$S_i$’ of TOBV makes an elliptical disc shape, which has a radius of the major axis ($R_1$) and radius of the minor axis ($R_2$) in the middle section ($S_0$), and these are defined by Equations (7) and (8). ‘α’ is half of the conical angle in the cone surface, and ‘$k$’ is substituted for tan α.

$$S_0=k_{0}x-H_0$$

(4)

$$S_1=k_{0}x-H_1$$

(5)

$$S_2=k_{0}x-H_2$$

(6)

$$\left|2R_1\right|=\left|\sqrt{\left(k_0^2+1\right)}(\frac{2k{H}_0}{1-k^2{k}_0^2})\right|$$

(7)

$$\left|2R_2\right|=\left|2k{H}_0\frac{b}{a}\right|$$

(8)

To obtain a circular disc of QOBV in the ‘$S_0$’, the ratio of radius (a/b) must be 1, as shown in Equation (9), and ‘$\eta$’ is defined by Equation (10):

$$\frac{a}{b}=\frac{1-k^2{k}_0^2}{\sqrt{k_0^2+1}}=\frac{R_1}{R_2}=1$$

(9)

$$\frac{a^2}{b^2}=\frac{{\left(1-k^2{k}_0^2\right)}^2}{k_0^2+1}=\eta$$

(10)

After ‘a’ and ‘b’ were found, the elliptical cone surface equation of QOBV is defined by Equation (11):

$$x^2+\eta y^2=k^2{z}^2$$

(11)

The sections of the circular discs ($S_i$) of QOBV are given by Equations (12)–(14):

$$x^2+\eta y^2={\left(k{k}_{0}y+H_0\right)}^2$$

(12)

$$x^2+\eta y^2={\left(k{k}_{0}y+H_1\right)}^2$$

(13)

$$x^2+\eta y^2={\left(k{k}_{0}y+H_2\right)}^2$$

(14)

When the sealing surface is truncated by a plane, which is parallel to the xoy plane with the x coordinate of y₀, the hyperbola formulation (ajbckd) illustrated in Figure 6 is obtained by Equation (15). The coordinates of the two intersection points (x_i, y, z_i) between the circular disk surface ($S_i$) and hyperbola is calculated by Equation (16):

$$\frac{z^2}{{\left(\frac{y_0}{k}\right)}^2}-\frac{x^2}{y_0{}^2}=1$$

(15)

$$x_i=\frac{-k^2{k}_0{H}_i\pm \sqrt{k^2{k}_0^2{x}_0^2+k^2{H}_i^2-x_0^2}}{k^2{k}_0^2-1},y=y_0,z_i=k_0{y}_i+H_i$$

(16)

The intersections between the vertical lines of $N_0$, $N_1$, and $N_2$ with the plane z = z0 are the center points of jk, bc, and ad, and the apexes $N_i$ of the minor axis of the intersection constitute the line$\overline{N_{i}O}$. During the valve opening and closing, point P(y, z), which is on the hyperbola ab, rotates about O₀ with a speed of $\overline{PN}$perpendicular to $\overline{O_{o}P}$, as illustrated in Figure 7. The angle of θ in Equation (17) is the friction angle between the sealing pair ($\overline{PM}$, $\overline{PN}$). The angle of β in Equation (18) is between $\overline{PM}$ and the axial line crossing the seal point and parallel with the x-axis. The angle of ${\theta }_1$ in Equation (19) is between $\overline{O_{o}P}$ and the vertical line crossing the rotary center ($O_0$). The vertical line intersects the rotation center, so the disc can quickly get out from the seal, and the friction moment reduces [18,19].

$$\theta ={\theta }_1-\beta$$

(17)

$$\beta =\arctan \left(\left|\frac{x}{k·\sqrt{y_0^2+x^2}}\right|\right)$$

(18)

$${\theta }_1=\arctan \left(\left|\frac{k·x-k·y_0}{k·z_0-\sqrt{y_0^2+x}}\right|\right)$$

(19)

2.3. Design of QOBV Disc

Based on the dimensions of the TOBV disc, as shown in

Table 1. Dimensions of QOBV and TOBV.

	TOBV	QOBV
Major diameter of the disc, R1 (mm)	190.1	190.1
Minor diameter of the disc, R2 (mm)	187	190.1
Thickness of disc, T (mm)	15	15
First offset (mm)	2.5	2.5
Second offset (mm)	23	23
Third offset (°)	10	10
A half of conical angle, α (°)	13.9	13.9
Displacement of the centrum of the disc, (mm)	369.9	369.9
	362.3	362.3
Displacement of the centrum of the disc top side, (mm)	377.5	377.5
	-	0.9698

, first offset: 2.5 mm, second offset: 10 mm, third offset: 10°, α: 13.9°, and $H_0$: 369.9 mm [18], ‘$\eta$’ from Equation (10) was calculated by 0.9698, and then the elliptical cone of QOBV was generated using mathematical software, GEOGEBR, as shown in Figure 8. The middle section of the disc ($S_0$) passing through both cones was determined by the major diameter of the disc, $R_1$ (190.1 mm), and the lower and upper sections ($S_1,S_2$) were created based on the thickness of the disc, T (15 mm). The minor diameters of TOBV and QOBV discs ($R_2$) were 187.0 mm and 190.1 mm, respectively.

The elliptical seal cone of QOBV has a circular disc with a diameter of 190.1 mm, as shown in Figure 9. The disc area of QOBV is 28,382.7 mm², which is larger than that of TOBV, 27,924.4 mm². This means the flow rate of QOBV is improved compared to that of TOBV.

3. Finite Element Analysis

Thermal–structural coupled analysis was performed to compute contact pressure at an inner temperature of 350 °C and working pressure of 1 MPa by using ANSYS workbench 2021 R2 [20].

3.1. Design of Number of Laminated Seal Layers

When the valve is closed, the sealing performance of the valve depends on the seal mechanism and type. The valve must maintain inner pressure and no leakage, regardless of temperature and pressure. In order to obtain a good sealing performance of QOBV, the laminated seal was adopted, and the design of the laminated seal should not only ensure sealing performance, but also the smooth rotation of the valve disc at a high temperature (350 °C).

The analysis models of QOBV were created using SpaceClaim Software, and the 2D geometry is shown in Figure 10. The disc’s contact slope angle range is 66.1°~86.1°. The properties of A240–316 stainless steel (retainer, seal layer, and disc) and A216-WCB (body) at 25 °C and 350 °C are listed in

Table 2. Material properties of QOBV.

	Poisson’s Ratio		Thermal Conductivity (W/m·°C)		Thermal Expansion (10−6 m/m·°C)		Elastic Modulus (GPa)
	25 °C	350 °C	25 °C	350 °C	25 °C	350 °C	25 °C	350 °C
A216 WCB	0.29	0.3	11.5	15.4	0	4.5	202	179
A240–316 stainless steel	0.29	0.29	20.1	21.2	0	5.9	195	172
Graphite	0.31	0.31	108.4	88.9	0	1.8	9.83	9.619

[18,21].

To find an appropriate number of laminated seal layers, a series of the laminated seal, which has different numbers of seal layers, was generated, as shown in Figure 11. SS and G mean the layer of A240–316 and graphite, respectively. The full thickness of the seal was fixed by 7 mm, and four kinds of seal layers with different number were created: one-layer (SS), three-layers (SS/G/SS), five-layers (SS/G/SS/G/SS), and seven-layers (SS/G/SS/G/SS/G/SS). To obtain a regular mesh quality, the stainless steel (SS) thickness was fixed by 1 mm, and the thickness of each graphite layer was 5 mm, 2.5 mm, and 1 mm, respectively. It is assumed that there is no void in the cross section surface, which means no sliding on each surface.

3.1.1. Analysis Condition

In the steady state thermal analysis, five surfaces contacting the working fluid are exposed at 350 °C. The convection coefficient of air at 25 °C (5 × 10⁻⁶ W/mm²C°) was applied to the body, the shaft, and the retainer, as shown in Figure 12a. The 3D model and result data of the thermal analysis are inputted to the static structural analysis, as shown in Figure 12b. The temperature distribution was imported to the boundary condition of structural analysis. Friction coefficients were set to 0.1 between the disc and the seal. The working pressure (1 MPa), seating torque (−5 × 10⁵ N·mm) [22], and fixed condition were applied as shown in Figure 12c.

3.1.2. Generation of Mesh

The tetrahedron element was adopted to all parts. The body size of the disc was set to 1.5 mm, that of the seal was set to 0.3 mm, and the other parts were given as 5 mm. The total nodes and elements were 964,727 and 228,610, respectively, as shown in Figure 13a. The generated mesh’s element quality, skewness, and orthogonal quality were checked, as shown in Figure 13b–d. The average element quality is 0.80963, where 1 is the best and 0 is the worst. The maximum value for skewness, which should not exceed 0.98, was 0.662. The minimum value for orthogonal quality, which should not fall below 0.05, was 0.3552, and almost elements were close to 0.9~1. Therefore, it is considered that the generated mesh is proper.

3.1.3. Result of the Thermal–Structural Coupled Analysis

In this present study, the contact pressure is calculated by Equation (20) [16], and a leakage criterion was used, which is contact pressure compared with the working pressure and estimated leakage by Equation (21) [12], shown in Figure 14.

$$P_{contact}=0.798\sqrt{\frac{F}{l\left(\frac{1-v_1^2}{E_1}\right)+(\frac{1-v_2^2}{E_2})}},\mathrm{F}=P_{working}\times cos\phi$$

(20)

When contact stress is higher than the working pressure, it could guarantee airtightness.

$$\begin{gathered} P_{contact}>P_{working}:\mathrm{no}\mathrm{leakage} \\ {P}_{contact}<P_{working}:\mathrm{leakage} \end{gathered}$$

(21)

The proper contact pressure generated by the seal provides good sealing performance and does not affect the disc’s opening and closing. Therefore, a trade-off between smooth driving and sealing is conducted. If contact pressure is a little higher than the working pressure (1 MPa), it ensures both the sealing performance and smooth driving of disc.

Table 3. The contact pressure of the laminated seal.

	Angle 86.1° (MPa)		Angle 66.1° (MPa)		Max Contact- Pressure (MPa)
	25 °C	350 °C	25 °C	350 °C	25 °C	350 °C
1-layers	17.46	98.2	12.51	0.04	27.78	282.5
3-layers	2.92	42.48	8.51	154.25	24.23	154.25
5-layers	2.87	14.37	8.25	89.8	24.16	114
7-layers	1.59	74.4	13.9	64.5	13.9	93.5

represents the contact pressures results at the angles of 88.1°and 66.1° and maximum values for the different numbers of laminated seal layers (one-layer, three-layers, five-layers, seven-layers) and temperatures (25 °C and 350 °C).

The contact pressures of the one-layer model at 25 °C were 17.46 MPa at 86.1° and 12.51 MPa at the angle of 66.1°, respectively, whereas some regions showed 0 MPa, so leakage may occur near the shaft. At 350 °C, the contact pressures were 98.2 MPa and 0.039 MPa at the angles of 86.1° and 66.1°, respectively. The maximum value at 350 °C was 282.5 MPa, but the leakage was produced near the shaft, as shown in Figure 15.

The contact pressures of the three-layers at the angles of 86.1° and 66.1° were 2.92 MPa and 8.51 MPa at 25 °C, and 42.9 MPa and 154.25 MPa at 350 °C, as shown in Figure 16. Leakage did not occur, and compared with the one-layer model at 350 °C, the contact pressures were reduced compared with one-way, but they were still too much larger than the working pressure (1 MPa).

The contact pressures of the five-layers at the angles of 86.1° and 66.1 were 2.87 MPa and 8.25 MPa at 25 °C, and 14.37 MP and 89.8 MPa at 350 °C, as shown in Figure 17. The maximum contact pressure was 114 MPa at 350 °C, leakage was not found, and compared with the three-layer model at 350 °C, the max contact pressures were reduced by 40 MPa.

The contact pressures of the seven-layers at the angles of 86.1° and 66.1° were 13.9 MPa and 1.59 MPa at 25 °C, and 93.5 MPa (maximum) and 27.184 MPa at 350 °C, respectively. The maximum contact stress (93.5 MPa) of the seven-layers at 350 °C was lower than that of one-layer (282.5 MPa), three-layers (154.25 MPa), and five-layers (114 MPa). Therefore, this suggested adopting a seven-layer model to ensure that the valve disc is good at sealing and can rotate rapidly. The contact stress distribution is shown in Figure 18.

3.1.4. Comparison of the Contact Pressures between QOBV and TOBV

To compare the contact pressures of the QOBV and the TOBV, a thermal–structural coupled analysis of TOBV with the same boundary and load conditions at 350 °C was performed. The contact pressures distribution of TOBV at each position are 18.96 MPa (A), 22.622 MPa(B), 46.122 MPa(C), 29.906 MPa(D), 28.429 MPa(E), 13.19 MPa(F), 148.9 MPa(G), and 13.548 MPa(H), respectively. The results of the QOBV contact pressures distribution at each position are 22.169 MPa(A), 23.241 MPa(B), 25.343 MPa(C), 16.278 MPa(D) 29.498 MPa(E), 11.627 MPa(F), 93.469 MPa(G), and 9.6218 MPa(H), respectively, as shown in Figure 19 and

Table 4. Comparison of the contact pressure of TOBV with that of QOBV.

	TOBV	QOBV
A	18.960 MPa	22.169 MPa
B	22.622 MPa	23.241 MPa
C	46.122 MPa	25.343 MPa
D	29.906 MPa	16.278 MPa
E	28.429 MPa	29.498 MPa
F	13.190 MPa	11.627 MPa
G	148.90 MPa(max)	93.469 MPa (max)
H	13.548 MPa (min)	9.6218 MPa (min)

. The maximum contact pressure of the TOBV was 148.9 MPa, and that of the QOBV was 93.47 MPa, and they occurred at an angle of 86.1° (G position). The maximum contact pressure was reduced by 55.431 MPa (37.2%).

3.2. Design of Laminated Seal Thickness of Seven-Layers Model

To increase the performance of the seven-layers laminated seal, the thickness design of stainless steel and graphite was conducted. With the same arrangement of the seven-layer model suggested above (SS/G/SS/G/SS/G/SS), the thickness of stainless steel varies from 0.7 mm~1.6 mm at an interval of 0.3 mm. Since the total thickness of the seal was fixed at 7 mm, the thickness of graphite varies from 1.4 mm~0.2 mm at the interval of 0.4 mm, and four analysis models were generated, as shown in Figure 20. Designs 1~4 were indicated to the thickness of 0.7 mm (SS) and 1.4 mm (G), 1 mm (SS) and 1 mm (G), 1.3 mm (SS) and 0.6 mm (G), and 1.6 mm (SS) and 0.2 mm, respectively.

The thermal–structural coupled analyses were implemented, and the results are listed in

Table 5. The results of the optimization.

Thickness		Maximum Contact Pressure (MPa)		Working Pressure (MPa)	Leakage
SS (mm)	G (mm)	SS (MPa)	G (MPa)
0.7	1.4 mm	98.9	0.87	1	Yes
1 mm	1 mm	94.69	1.4	1	No
1.3 mm	0.6 mm	86.8	3.68	1	No
1.6 mm	0.2 mm	76.15	8.55	1	No

. By increasing the SS thickness and reducing the graphite, the contact pressure from the SS surface was reduced, and the G was increased. The results of the description were the following:

Design 1 produced the maximum contact pressure of 98.9 MPa (SS) and 0.87 MPa (G). Compared with working pressure, the stainless steel (SS) prevents leakage, but the contact pressure of the graphite (G) was less than the work pressure (1 MPa), which leads to leakage problems.

Design 2 produced the maximum contact pressure of 94.69 MPa (SS) and 1.4 MPa (G), and it exceeds the work pressure and provided good performance in the simulation section, but the value (1.4 MPa) of existent contact pressure was close to work pressure (1 MPa). This leads to a dangerous actual sealing state, which may lead to leakage problems in the experimentation.

Design 3 produced the maximum contact pressure of 86.8 MPa (SS) and 3.68 MPa (G), shown in Figure 21. Both seal parts of the stainless steel (SS) and the graphite (G) not only produced a good sealing, but also produced a safe contact pressure value graphite (G) compared with design 2.

Design 4 produced the maximum contact pressure of 76.15 MPa (SS) and 8.55 MPa (G), although the contact pressures of the stainless steel (SS) and the graphite (G) exceeded the working pressure (1 MPa) shown in Figure 22. The most common form of compressed expanded graphite (CEG) thickness is from 0.3 to 5 mm [23], so the graphite thickness of 0.2 mm would lead to a high cost in manufacturing.

For these reasons, design 3 was suggested as the best design, whose contact pressures were not only sealed satisfyingly, but it is also less costly than design 4 (G: 0.2 mm) in its manufacturing.

3.3. Structural Safety of the Suggested QOBV

The structural safety of QOBV with the above design 3 was verified. The maximum equivalent stress of the main elements in Figure 23 was compared with the allowable strength in

Table 6. The structural safety of the proposed QOBV.

Part	Allowable Strength (MPa) [16]	Maximum Equivalent Stress (MPa)	Yielding
Body (A216 WCB)	184.25	128.83	No
Shaft	385.25	273.14	No
Disc (A210–316)	184.25	172.37	No
SS (A210–316)	137.35	86.8	No
Graphite	4.69	3.68	No

. It is expected to work normally and not yield at 350 °C.

3.4. Discussion

At present, the QOBV is being developed, and its experiments are being conducted confidentially, making it difficult to carry out sealing tests. Consequently, the most effective approach is to use a simulation program to estimate the sealing performance. The authors compared the contact pressure results of the QOBV with those of the TOBV and verified the reduction in contact pressure of the QOBV using the ANSYS simulation program. In the previous study by Kwak (reference 18), contact stresses of a 2D model of the triple offset butterfly valve (TOBV) were derived under symmetry conditions, specifically only at the minimum and maximum slope angles of the seal contact surface (66.1° with 97.15 MPa and 86.1° with 66.81 MPa), as illustrated in Figure 24. This study conducted a 3D analysis of the TOBV and computed contact stresses for slope angles ranging from 66.1° to 86.1°. The 3D simulation provides more realistic results than the symmetrical 2D simulation, thereby improving accuracy. Therefore, comparing our findings with those of Kwak’s previous work would be inappropriate. We evaluated the contact stress of the TOBV through 3D simulation and contrasted it with that of the quadruple offset butterfly valve (QOBV) in chapter 3.1.4. The maximum contact pressure of the TOBV was 148.9 MPa, whereas that of the QOBV was 93.47 MPa at an 86.1° angle. This represents a 55.431 MPa (37.2%) reduction in the maximum contact pressure.

4. Conclusions

This study developed the QOBV to solve the problems of the previous butterfly valves. A summary is presented in the following.

(1)

The QOBV developed in this study has advantages as follows:

-

The QOBV improved the flowrate (27,924.4 mm² to 28,382.7 mm²) compared with TOBV.

-

The thermal structure coupling analysis of the seven-layer laminated seal composed of SS and G in QOBV showed that the maximum contact pressure (93.45 MPa) was 37.2% lower than the TOBV (148.9 MPa) of the same design. Compared to TOBV, the QOBV disc operation during the opening and closing process was faster and smoother.

-

The disk with a circular cross section to finish the entire disc contour was easier than that with elliptical cross section, so that it can reduce the processing time and cost.

(2)

The surface function of the QOBV disc with a circular cross section was established from an ellipse cone function.

(3)

The thickness design of the laminated seal was conducted. The best design with SS and graphite thickness of 1.3 mm and 0.6 mm, respectively, was suggested.

The disc can be applied widely to various industrial fields, such as engine exhaust systems and turbine and power plants, under severe environments.