**6.2 Numerical simulation**

Numerical analysis was carried out in order to examine the effects of the fin. The numerical analysis code was Star-CD and the barotropic model was used as a model for cavitation. The calculation conditions were 45 and 22 . Figure 9 shows the calculation results of calculating velocity vectors and Figure 10 shows the coordinate system according to numerical analysis. The origin is at the center of the valve stem. Velocity vectors were examined in the cross section which is near the fin in order to clarify the effects of the fin. Figs.9 (a),(c),and (e) show the velocity vectors for the cross-section which was at Y=15mm (Y=0.3Dia.). Figs.9 (b),(d),and (f) show the velocity vectors for the cross section which was at X=15mm (X=0.3Dia.). The fin is not shown in figures (b),(d), or (f) because the fin is not part of the cross section in the position where X=0.3 Dia.

Fig.9(a) reveals that for the normal valve, the flow from the orifice side and the flow from the nozzle side interfered with one another about 1 Dia from the valve stem. This interference made the downstream flow of the valve swell and brought about intense vortices. These vortices is correspond to the cloud cavitation shown in Fig.7(a).

The swell of the flow for the TYPE-B valve with two fins was relatively moderate in comparison with that of the normal valve. It is clear from Fig.9(c) that the swell of the flow is the smallest for the TYPE-C valve. However, it is probable that the cavitation intensified since the fin was located in a position where the contraction flow was severe. Fig.9 (f) shows a contraction flow more severe than that of the other valves on the orifice side. Accordingly,

(a) Normal Valve (θ=45°,σ=17,ζ=8) (b) TYPE-A (θ=50°,σ=16,ζ=8)

(c) TYPE-B (θ=47.5°,σ=16,ζ=8) (d) TYPE-C (θ=50°,σ=16,ζ=8)

suppressed by the attachment of the fins. As for the TYPE-C valve with three fins, though the two fins on either side of the valve probably suppressed the cavitation as they did for the TYPE-B valve, it is appropriate to conclude from Fig.7(d) that the cavitation reduction effect was not obtained because the central fin intensified the cavitation as it did for the

The photograph in Fig.8(c) shows that cavitation was moderate for TYPE-B in comparison to the other three valve types. In TYPE-A and TYPE-B, in the vicinity of the upper wall of the position 1 Dia from the valve stem, vortex cavitation was remarkable as mentioned above.

Numerical analysis was carried out in order to examine the effects of the fin. The numerical analysis code was Star-CD and the barotropic model was used as a model for cavitation. The

calculating velocity vectors and Figure 10 shows the coordinate system according to numerical analysis. The origin is at the center of the valve stem. Velocity vectors were examined in the cross section which is near the fin in order to clarify the effects of the fin. Figs.9 (a),(c),and (e) show the velocity vectors for the cross-section which was at Y=15mm (Y=0.3Dia.). Figs.9 (b),(d),and (f) show the velocity vectors for the cross section which was at X=15mm (X=0.3Dia.). The fin is not shown in figures (b),(d), or (f) because the fin is not part

Fig.9(a) reveals that for the normal valve, the flow from the orifice side and the flow from the nozzle side interfered with one another about 1 Dia from the valve stem. This interference made the downstream flow of the valve swell and brought about intense

The swell of the flow for the TYPE-B valve with two fins was relatively moderate in comparison with that of the normal valve. It is clear from Fig.9(c) that the swell of the flow is the smallest for the TYPE-C valve. However, it is probable that the cavitation intensified since the fin was located in a position where the contraction flow was severe. Fig.9 (f) shows a contraction flow more severe than that of the other valves on the orifice side. Accordingly,

. Figure 9 shows the calculation results of

 and 22 

vortices. These vortices is correspond to the cloud cavitation shown in Fig.7(a).

Fig. 8. Side Views of Cavitation Conditions.

TYPE-A valve.

**6.2 Numerical simulation** 

calculation conditions were 45

of the cross section in the position where X=0.3 Dia.

Fig. 9. Cavitation Flow around a Butterfly Valve (θ=45°,σ=22).

in TYPE-C, the flow interference effect was suppressed by the two fins on either side of the valve, but was canceled out by the intensification of cavitation due to the central fin. Therefore, the effect wherein two fins on either side of a valve suppresses flow interference is offset by the effect wherein the fin in the center intensifies cavitation.

Fig. 10. Coordinate System in Numerical Analysis.

Noise Reduction in Butterfly Valve Cavitation

distribution of Fig.11(c) is the reverse of Fig.11 (b).

was observed at 38

observed at 58 

by Semicircular Fins and Visualization of Cavitation Flow 493

A plate with a half circle hole shown in Fig.11(d) was installed at 3D (D: pipe diameter) upstream of the valve to obtain an biased velocity distribution. The velocity distributions were analyzed by numerical simulation using a commercial code Star-CD because the measurement of the velocity distribution was difficult behind such plate. Fig.11 shows three types of velocity distributions. The vertical axis shows the location divided by the radius of the pipe and the horizontal axis shows velocity. Fig.11 (a) shows an example of a normal turbulent velocity distribution at 1D upstream of the valve. Fig.11 (b) and Fig.11 (c) show the biased velocity distributions. In each example, the average velocity was 3 / *m s* . In Fig.11 (a), the velocity distribution mostly agreed with Blasius's law. In Fig.11(b), the velocity on the orifice side of the valve was larger than that on the nozzle side of the valve. The velocity

In many practical industry plants, the flow rates are controlled by adjusting the opening of the butterfly valve and are determined by head curves of pumps and pressure losses of the whole piping systems. Accordingly, the effect of valve shape on cavitation noise should also

In the flow of the normal velocity distribution, the cavitation at the orifice side becomes severe since the contraction at the orifice side becomes intense and the pressure around the edge becomes very low. As shown in Fig.12, the cavitation inception of the Type-B valve

suppressed after inception. As for the maximum noise, the Type-B valve was lower than the

Figure 13(a) shows the cavitation noise in the flow of the biased velocity distribution (A). In this case, the flow rate of the orifice side was much larger than that in the normal velocity distribution. Accordingly, the cavitation noise of both the normal valve and Type-B valve

 to 50 

, but in contrast the cavitation inception of the normal valve was

, 12 ).

compared with the results of

. Moreover, the increase of cavitation noise of the Type-B valve was

be investigated under the condition of a constant pressure loss coefficient.

normal valve. Therefore, the effect of semi-circular fins is clear.

Fig. 12. Cavitation Noise under normal velocity distribution ( 45

were increased remarkably in the range of 30
