ML20128B757

Discusses Pipe Support Sys Stability Concept,Including Stability Displacement Envelope at Each Support & Interdependence of Forces & Displacements Among Supports & Support Instability Modes,Per & 850314 Meeting

ML20128B757
Person / Time
Site:	Comanche Peak
Issue date:	04/30/1985
From:	Williams N CYGNA ENERGY SERVICES
To:	Beck J TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC)
References
84042.026, NUDOCS 8505280054
Download: ML20128B757 (9)

Text

c 6

SefylCf5 101 Cahfornia Street, Suite 1000, San Francisco, CA 94111-5894 415:397 5600 April 30,1985 84042.026 Mr. J.W. Beck Manager - Licensing Texas Utilities Generating Company Skyway Tower 400 North Olive Street L.B. 81 Dallas, Texas 75201

Subject:

Stability of Pipe Supports - Additional Information Comanche Peak Steam Electric Station Independent Assessment Program - Phase 3 Job No. 84042

Reference:

N.H. Williams (Cygna) letter to J.B. George (TUGCO),

" Stability of Pipe Supports," 84042.035, dated February 19, 1985

Dear Mr. Beck:

The topic of pipe support stability was discussed at a meeting between CPRT and Cygna on March 14, 1985. As a result, this letter is intended to provide an expanded discussion of the " system stability" concept and support instability modes described in the above referenced letter.

System Stability The concept of system stability for piping systems is inherent when clamp-strut arrangements (supports of the second category) are used to support the pipe.

For example, a strut, which is a two pin column member, is unstable unless the movement of the pin nearest the pipe is restrained in some manner. In the typ-ical arrangement (see Figure 1) this is done with a standard pipe clamp, which resists the motion of the pin relative to the pipe through clamping (friction) forces. As a result, the clamp maintains the position of the support relative to the pipe, and the pipe, by means of the other supports in the systen, main-tains the support within a displacement envelope in which the clamping forces will perform their function. A system designed in this manner will be stable.

8!!05280054 850430 PDR ADOCK 05000445 A PDR gM San Francisco Boston Chicago Richland , l

w. .- ... . . - - - - . . ~

. mummmmmmum upg*o_

Mr. J.W. Beck April 30,1985 Page 2 Cygna's underlying position regarding system stability is that system staMlity and individual support stability are not independent. To understand fully the concept of system stability, two fundamental and independent concepts must be related. The first is the idea of a stability displacement envelope at each support and the second is the interdependency of both forces and displacerents among all supports.

First, for all supports of the second category there exists at each suppor' a stability displacement envelope within which the pipe can displace and the sup-port remains stable (i.e., resists the applied load). This is illustrated for the single strut /U-bolt support shown in Figures 2a and 2b, where "ei" denotes the eccentricity of the initial strut force and "ef" denotes the eccentricity of the final strut force under the most severe piping system loading (i.e., that producing the largest value of the product of strut force and eccentricity).

Pipe displacements which exceed the envelope limits initiate support instability because they increase the eccentricities of the support forces to the point where the resisting (frictional) forces, which bond the support to the pipe, are overcome by the applied forces. Thus, the product of strut force and eccentri-city "ef" must be equal to or less than the resistance provided by clamping forces for stability to be maintained. This is accomplished if delta D, the support displacement for the most severe loading, is less than delta-S1, the maximum displacement for which stability can be maintained at a specified level of strut force. The limits of the stability displacement envelope are a func-tion of the specified support load and the initial eccentricities of the support forces. These limits decrease with either an increase in the specified support load or an increase in the initial eccentricities of the support forces. Thus, if the specified support load changes or the initial eccentricities change, the limits of the stability displacement envelope also change.

Second, the displacements and forces associated with each support are a function of the total number and arrangement of supports within the piping system. This is a direct result of the fact that in any continuous and externally redundant structure the influence of any imposed displacement or change in reaction is felt throughout the entire structure. A change in either the number or arrange-ment of the supports effects the displacements and forces at each support in the piping system, with some supports effected more than others.

As a consequence of basic structural behavior and of the fact that each support of the second category has a stability displacement envelope associated with it, the assessment of piping system stability cannot be separated from the stability (or removal) of an individual support. Suply stated, a piping system, which is an externally redundant structure, remains stable when an individual support becomes unstable (or is removed) if, and only if, the resulting redistribution of forces and displacements due to the loss of the support results in a stable

s Mr. J.W. Beck April 30,1985 Page 3 equilibrium configuration. The key phrase here is: "results in a stable equi-librium configuration." There is no guarantee that the instability (removal) of one support in a piping system results in a stable equilibrium configuration unless both the new displacements at each support remain within the stability displacement envelope and the new forces at each support do not exceed the cor-responding support load.

In order to demonstrate " system stability" in the presence of unstaole supports, it is not sufficient to remove an unstable support from the system and subse-quently show that each remaining support can resist the new forces. In addi-tion, it must be shown that removing the unstable support does not affect the stability of other supports. That is, overall system stability must be re-evaluated in the absence of the removed support. This evaluation can be per-formed by identifying potentially unstable supports and assessing the extent of the load / displacement redistribution due to the postulated loss of support functions.

Support Instability Modes The most obvious example of a support which has the potential to become unstable is the single strut /U-bolt support. Other less obvious examples of potentially unstable supports are the double strutted box frame and double strutted trapeze supports with U-bolts. Both of these supports will be briefly discussed for the purpose of illustrating other potential modes of support instability.

The double strut box frame cannot rotate, except as a mechanism, about the lon-gitudinal axis of the pipe and is therefore stable with respect to this mode of instability. However, since positive attachment to the pipe is not guaranteed, this support possesses the potential to become unstable by being able to move longitudinally along the pipe due to some combination of operational events such as pipe thermal displacement and vibration. As Figure 3 illustrates, a small pipe displacement, DP, can cause a rather large support slippage, DS. In its new position the support may not be able to resist any significant applied load because the support was not designed for this new configuration. What is of even greater importance is the fact that during the displacement, DP, the sup-port did not resist the applied load which would have prevented this displace-ment. Because is did not resist the applied load it is unstable. The load it should have carried was redistriouted to other supports and potentially desta-bilizing displacements at other supports were increased.

. - -.. - . - , ~. -.

o E __

Mr. J.W. Beck April 30,1985 Page 4 A double strut trapeze support with U-bolt is shown in Figure 4, position one.

During the application of a downward load the support beam CD wants to twist about the Y-axis. Stabilizing resisting forces are provided by the frictional line contact forces at E and the pipe rotational stiffness. Should the fric-tional bond be broken the support becomes momentarily unstable. However, this momentary instability Daf the rigid body type) could be tolerated, provided that it can be demonstrated t1at sufficient forces eventually develop to completely remove the instability (i.e., stop the rotation of beam CD and allow the support to function as designed). This could occur with the U-bolt cocked against the pipe as shown in position two. This second position may involve negligible vertical deflection at the pipe centerline. Thus, the support is still carrying its design load. The U-bolt and trapeze beam may be able to- resist the compo-nent of load needed to hold the support cocked, from a purely strength point of vf aw (although U-bolts have not been qualified for this type of loading). How-ever, the stiffness of the U-bolt and trapeze beam nay not be sufficient to prevent excessive deformation at the upper pin under the lateral load component at points D & C. This will have the effect of increasing the load until the pipe has displaced to a point at which the _ support no longer functions. While the concept of momentary instability is certainly viable, the qualification of a support as stable after undergoing a momentary instability may, therefore, re-quire a considerable effort.

Although the concept of system stability in the presence of potentially unstable supports is difficult to assess, the evaluation of their impact need not involve sophisticated analyses. Rather, a systematic review of each piping problem to identify potentially unstable supports would be beneficial. Then, in problems with few supports of this type, the effect of removal of all these supports on overall results could be assessed. Or, in problems with more potentially un-stable supports, modification of the supports to remove the potential instabil-ity could be proposed. In either case, the individual supports could be quanti-tatively shown to be stable within the maximum possible limits of their applied loads and displacements. In view of the many parameters which a generic study would have to consider on this issue, the systematic approach described above may be a more logical choice to pursue.

mmmmmmmmmm ewess~

Mr. J.W. Beck April 30,1985 Page 5 If you have any questions on this issue, please do not hesitate to call.

Very truly yours, e

-O N.H. Williams Project Manager NHW/ajb Enclosures cc: Mr. V. Noonan (USNRC) w/ attachments i Mr. S. Burwell (USNRC) w/ attachments Mr. S. Horin Mr. W. Treby ((USNRC) w/attachmentsBishop, Liberman, et al.) w/ attachments Mr. J. Redding (TUGCO) w/ attachments

[

Mr. J. Finneran (TUGCO) w/ attachments

! Ms. J. van Amerongen (TUGC0/EBASCO) w/ attachments l Ms. J. Ellis (CASE) w/ attachments l

Mr. D. Pigott (0rrick. Herrington & Sutcliffe) w/ attachments Mr. F. Dougherty (TENERA) w/ attachments Mr. R. Ballard (Git,bs & Hill) w/ attachments, 1

1 i

. - . _ . ~ . _ _ _ _ . _ _ . _ . . . . . . _ . . _ - . - . . . . . . _ _ . . _ . . - . _ . .. .

.. FIGURE 1 Standard Clamp-Strut Arrangement i

,\ .

,e g msr.eus ds1'z

. N 1 c c f f

x. ,

~ m ; C ,' . .&

,dr, .

I

== Q -

C- C og

. . i er fr ~ll j

(REE)

I i- .

l l

t 1

l l

t R

"'"'*-~n_

.- . . . - . . . . . . . - - . . - - - . . . ~ . . . - - ~ . . . . . . . . .. .. - .

Single Strut /U-bolt Support 0 -

( J

_as-built position I Figure 2a S

\

I Y *

' T l

0 l

l l

Stability Displacement

~

l Envelope j 4 ,  ; 5

,m 651 m -

m AS2 m

~

_ AD _ ~

eg r%

l i

Figure 2b / \

(

l Support Position for most severe

\"/

loading \

_ \e .

l I

I l

l

FIGURE 3 Box Frame and Strut Original Configuration ,

, (Front View) l  !

i 3 l (U

!l 1 4"M A")

. . r- ,

I Tb I I i-) .

/ \

Displaced Configuration (Side View) e DS :

Original Position DP y l

, i e- ....

! Displaced Position O

l .P. .

7. . . x

.. . ..~- .. .- - - -- . ..-..

FIGURE 4 Positions of Stability Double Strut Trapeze with U-bolt Elevation View A+

h

)

, r______

hC "D

. . E ,.

E2 ',

n 1

X_ -

,, .d8A . .

( '8 B

A=-

. 46" * * - 12"

• SECTION A-A SECTION A-A Position 1 Position 2

- S M y .

_____f___ pipe _____

p__ . _ _ _ __ __ _ f _

ss si k O o

C,o g- C ay

-Z l

l O

. . A,B , , ,

A,B

//////// ////Y///- l l

l l

=  ; e .

. . .