ML16131A655

White Paper Prepared in Response to a Question Raised During the Audit

ML16131A655
Person / Time
Site:	Nine Mile Point
Issue date:	05/04/2016
From:	Exelon Generation Co
To:	Office of Nuclear Reactor Regulation
References
NMP2L 2621
Download: ML16131A655 (9)

Text

Attachment 4 White Paper Prepared in Response to a Question Raised During the Audit (Nonproprietary)

This Report Does Not Contain CDI Proprietary Information CDI White Paper: 16-0SNP Significance of Steam Dryer Fluid-Structural Interaction - CDI Perspective In Response to Nine Mile Point Unit 2 Audit Revision 0 Prepared by Alexander H. Boschitsch Reviewed by Alan]. Bilanin Continuum Dynamics, Inc.

34 Lexington Avenue Ewing, NJ 08618 April 21, 2016 1of8

This Report Does Not Contain CDI Proprietary Information Significance of Fluid Structure Interaction in Steam Dryers - CDI Perspective

((

(3J))

Fluid-Structure Interaction in Steam Dryer Analysis The general phenomenon of fluid-structural interaction refers to the mutual coupling between the fluid (here, acoustic) and structural response. In the general case: (A) the fluid pressure impresses a load on the structure thus causing it to dynamically deform; and (B) simultaneously, the fluid response induces an acoustic contribution that modifies the surrounding pressure field.

Prior stress evaluations have generally accounted for the ((

(3J))

In the CDI 14-08 stress report, ((

(3J))

In 3D, the acoustic field is more complicated and additional inertial and stiffness contributions also arise. However, as with perforated plate damping, these complicating effects were conservatively accounted for ((

C3l)) However, during a technical review for Nine Mile Point Unit 2 with the NRC staff and consultants held at the CDI office on February 18, 2016, a presentation by an NRC consultant attempted to show that the effect of vibration-induced loading is even lower than this 42.5% value.

To support this assessment, two examples involving vibrating plates in unbounded domains were referenced (see also [2]). While the solutions and acoustic behaviors of those examples are well known, CDI does not believe they are representative of the confined region acoustics of steam dryers as we aim to demonstrate below.

((

(3J))

2 of8

This Report Does Not Contain CDI Proprietary Information

((

(3J))

((

(3J))

Figure 1. Pressure measurements on the Quad Cities steam dryer at the co-located pressure transducers PT20 (outside) and PT14 (inside) at 790 and 930 MWe operating conditions.

Model Problems by the NRC Consultant

((

(3)))

However, during the NRC technical review held at the CDI office on 02-18-2016, this value was challenged on the basis of two 'model' problems which would indicate that the FSI load is considerably smaller. The first model problem considers a loudspeaker in a large volume and the other examines the propagation of waves on infinite structures. Both examples are described in the 3 ofB

This Report Does Not Contain CDI Proprietary Information Structural Acoustics Tutorial published in Acoustics Today April 2007, particularly Part 2 of this tutorial. While the solutions and physics of these examples have long been understood, they are not representative of steam dryer geometries since they omit the effects of confinement in a closed 3D geometry.

In the first example, formula (5) in [2] provides the in-phase component of pressure P induced on the piston due to its own velocity u (i.e., the real part of the impedance) as p ( Ji(Zroa/ c)l

-=pell-u (roa/ c) j =pcK where p is the fluid density, c is the fluid acoustic speed, co is the circular forcing frequency, and a is the piston radius. K equates to the adjustment factor used to correct lD VIL. For the piston in an infinite domain, K depends on the Bessel function J1 and is plotted in the upper diagram of Figure 4 in [2]. Note that when coa/c=rr./2, then K=0.82. ((

(3J)]

The other model problem considers the effects of a plate (finite or infinite) that is again adjacent to an infinite space. The physics of waves radiating or decaying and the manner in which these effects are characterized in terms of sub- or super-sonic bending waves are also well understood and not in dispute. The phenomenon explains, for example, why sonic booms generated by supersonic aircraft produce a loud effect on the surface (and thus are banned for over land transport) while subsonic aircraft do not. However, the acoustics of infinite plates next to infinite steam volumes do not correspond to those of steam dryers except for higher order (and high frequency) modes.

The Effects of Confinement

((

(3)))

4of8

This Report Does Not Contain CDI Proprietary Information

((

(3J])

The high frequency case also reveals other nuances that arise in full 30 analysis. ((

(3))) This is explained by noting that the acoustic field supports different vertical modes at each frequency, resulting in different pressure distributions. Note too that the plots here show the real part of the complex pressure (i.e., in phase with the velocity); an imaginary component (not shown) also is present which, for this particular geometry, generally corresponds to a spring force (but in general could also include contributions from acceleration, jerk and other higher order rate terms).

hood supports Figure 2. Steam dryer section associated with hood section, hood supports and vane bank.

5of8

This Report Does Not Contain CDI Proprietary Information

((

(3J))

Figure 3. Geometry and boundary conditions for vibration-induced pressure example.

((

(3J))

6of8

This Report Does Not Contain CDI Proprietary Information

((

(3J))

Figure 4. Pressure field produced by moving surface. Left to right corresponds to forcing frequencies of 30 Hz, 60 Hz and 120 Hz respectively. Pressures are normalized by puc where p is the density of steam, c is the acoustic speed and u is the velocity amplitude. ((

(3)))

7 ofB

This Report Does Not Contain CDI Proprietary Information Summary The basis for ((

(3J))

References

1. Continuum Dynamics, Inc. (2014) Stress Re-Evaluation of Nine Mile Point Unit 2 Steam Dryer at 115% CLTP. C.D.I. Report No. 14-0BP (Proprietary), July.

2. Hambric, Stephen A. and John B. Fahnline, Structural acoustics tutorial, Part 2: Sound - Structure Interaction. Acoustics Today, 2007. 3(2): p. 9-27.

BofB

Attachment 4 White Paper Prepared in Response to a Question Raised During the Audit (Nonproprietary)

This Report Does Not Contain CDI Proprietary Information CDI White Paper: 16-0SNP Significance of Steam Dryer Fluid-Structural Interaction - CDI Perspective In Response to Nine Mile Point Unit 2 Audit Revision 0 Prepared by Alexander H. Boschitsch Reviewed by Alan]. Bilanin Continuum Dynamics, Inc.

34 Lexington Avenue Ewing, NJ 08618 April 21, 2016 1of8

This Report Does Not Contain CDI Proprietary Information Significance of Fluid Structure Interaction in Steam Dryers - CDI Perspective

((

(3J))

Fluid-Structure Interaction in Steam Dryer Analysis The general phenomenon of fluid-structural interaction refers to the mutual coupling between the fluid (here, acoustic) and structural response. In the general case: (A) the fluid pressure impresses a load on the structure thus causing it to dynamically deform; and (B) simultaneously, the fluid response induces an acoustic contribution that modifies the surrounding pressure field.

Prior stress evaluations have generally accounted for the ((

(3J))

In the CDI 14-08 stress report, ((

(3J))

In 3D, the acoustic field is more complicated and additional inertial and stiffness contributions also arise. However, as with perforated plate damping, these complicating effects were conservatively accounted for ((

C3l)) However, during a technical review for Nine Mile Point Unit 2 with the NRC staff and consultants held at the CDI office on February 18, 2016, a presentation by an NRC consultant attempted to show that the effect of vibration-induced loading is even lower than this 42.5% value.

To support this assessment, two examples involving vibrating plates in unbounded domains were referenced (see also [2]). While the solutions and acoustic behaviors of those examples are well known, CDI does not believe they are representative of the confined region acoustics of steam dryers as we aim to demonstrate below.

((

(3J))

2 of8

This Report Does Not Contain CDI Proprietary Information

((

(3J))

((

(3J))

Figure 1. Pressure measurements on the Quad Cities steam dryer at the co-located pressure transducers PT20 (outside) and PT14 (inside) at 790 and 930 MWe operating conditions.

Model Problems by the NRC Consultant

((

(3)))

However, during the NRC technical review held at the CDI office on 02-18-2016, this value was challenged on the basis of two 'model' problems which would indicate that the FSI load is considerably smaller. The first model problem considers a loudspeaker in a large volume and the other examines the propagation of waves on infinite structures. Both examples are described in the 3 ofB

This Report Does Not Contain CDI Proprietary Information Structural Acoustics Tutorial published in Acoustics Today April 2007, particularly Part 2 of this tutorial. While the solutions and physics of these examples have long been understood, they are not representative of steam dryer geometries since they omit the effects of confinement in a closed 3D geometry.

In the first example, formula (5) in [2] provides the in-phase component of pressure P induced on the piston due to its own velocity u (i.e., the real part of the impedance) as p ( Ji(Zroa/ c)l

-=pell-u (roa/ c) j =pcK where p is the fluid density, c is the fluid acoustic speed, co is the circular forcing frequency, and a is the piston radius. K equates to the adjustment factor used to correct lD VIL. For the piston in an infinite domain, K depends on the Bessel function J1 and is plotted in the upper diagram of Figure 4 in [2]. Note that when coa/c=rr./2, then K=0.82. ((

(3J)]

The other model problem considers the effects of a plate (finite or infinite) that is again adjacent to an infinite space. The physics of waves radiating or decaying and the manner in which these effects are characterized in terms of sub- or super-sonic bending waves are also well understood and not in dispute. The phenomenon explains, for example, why sonic booms generated by supersonic aircraft produce a loud effect on the surface (and thus are banned for over land transport) while subsonic aircraft do not. However, the acoustics of infinite plates next to infinite steam volumes do not correspond to those of steam dryers except for higher order (and high frequency) modes.

The Effects of Confinement

((

(3)))

4of8

This Report Does Not Contain CDI Proprietary Information

((

(3J])

The high frequency case also reveals other nuances that arise in full 30 analysis. ((

(3))) This is explained by noting that the acoustic field supports different vertical modes at each frequency, resulting in different pressure distributions. Note too that the plots here show the real part of the complex pressure (i.e., in phase with the velocity); an imaginary component (not shown) also is present which, for this particular geometry, generally corresponds to a spring force (but in general could also include contributions from acceleration, jerk and other higher order rate terms).

hood supports Figure 2. Steam dryer section associated with hood section, hood supports and vane bank.

5of8

This Report Does Not Contain CDI Proprietary Information

((

(3J))

Figure 3. Geometry and boundary conditions for vibration-induced pressure example.

((

(3J))

6of8

This Report Does Not Contain CDI Proprietary Information

((

(3J))

Figure 4. Pressure field produced by moving surface. Left to right corresponds to forcing frequencies of 30 Hz, 60 Hz and 120 Hz respectively. Pressures are normalized by puc where p is the density of steam, c is the acoustic speed and u is the velocity amplitude. ((

(3)))

7 ofB

This Report Does Not Contain CDI Proprietary Information Summary The basis for ((

(3J))

References

1. Continuum Dynamics, Inc. (2014) Stress Re-Evaluation of Nine Mile Point Unit 2 Steam Dryer at 115% CLTP. C.D.I. Report No. 14-0BP (Proprietary), July.

2. Hambric, Stephen A. and John B. Fahnline, Structural acoustics tutorial, Part 2: Sound - Structure Interaction. Acoustics Today, 2007. 3(2): p. 9-27.

BofB