ML20005B732

Safety Evaluation Re Westinghouse Criteria for Low Pressure Nuclear Turbine Disc Insp.Westinghouse Crack Growth Rate Equations for Bore & Keyway Cracks Acceptable

ML20005B732
Person / Time
Site:	Cooper
Issue date:	08/24/1981
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20005B730	List: ML20005B729 ML20005B732
References
TAC-12589, NUDOCS 8109020025
Download: ML20005B732 (4)

Text

_-_

SAFETY EVALUATION REPORT Criteria For Low Pressure Nu. clear Turbine Dise Inspection e

Westinghouse has prepared a proprietary report covering their investigation and analysis of turbine disc cracking. This report include,s.ca. statistical analysis of all turbine disc cracks found to date and recommends criteria for scheduling disc inspections that provide a very low probabil.ity of. disc failure prior to inspection.

l We have evaluated each of the criteria presented in the report and are in agreement with either the. Westinghouse position or with one of the positions in those cases where they suggest alternatives. These criteria and our evalua :

tion of each is described below.

There are several major criteria involved in setting inspection schedules.

Basically, the approach used is to make a conservative prediction *of how fast a presumed or actual crack will grow and then schedule an inspection prior to the time the crack grows large enough to ba of concern. Analytic components of this approach are:

i A. Crack Growth Rate '

B. Critical Crack Size 5-a C. Fraction of Critical Crack Size Allowed.

The Westinghouse criterion for establishing each of t,hese factors and our evalua-tion is discussed below.

A.

Crack Growth Rate Westinghouse has performed statistical studies using the field data on crack sizes and s" apes as related to temperature of operation, location (bore or keyway), material strength, and environment. Trey have selected a conservative kD 25 e10814 0 [ h K 0500029e PDR L

^

j.

upper bound basis and develeped equations that define a conservative crack growth rate for each disc.. We have reviewed the Westinghouse methodology and find the growth-rate equations to be acceptable.

~

M

B.

Critical Crack Size

.i Westinghouse has used the usual LEFM model to calculate critical crack size, taking into consideration effects of crack shapes expected in dif-ferent locations (bore or keyEay"). The fracture toughness values used

~

~

~

in the calculations are determined from actual charpy V tests on eacn disc, using the common Rolf-Novak correl'ation. Westinghouse also pre-sented test results, obtained from both fracture mechanics specimens and i

a spin test, to show that this correlation yields over-conservatively i

low values of the toughness related to actual disc cracks because the cracks are irregular and branched. We therefore p" refer the alternative proposed; i.e., to increase the estimate of fracture toughness derived from the Rolf-Novak expressien by 20% to reflect the effect of the irreg:-

ular nature of actual service-induced disc cracks. This 20% increase is still very conservative, as all of the test data-show even larger increases.

C.

Fraction of Critical Size Allowed Westinghouse has proposed two methods for applying this factor.

One in-volves a very conservative critical crack size cticulation using the Rolf-Novak value of toughness, and then permitting operation until a crack grows to a predicted maximu,m of 75". of the critical size. An alternative approach is to use the more realistic (but still conservative) augmented toughness value { discussed in B above) that gives a larger and more realistic crack i

size and then permitting operation until a crack grows to a predicted maximum of.50% of critical. crack size.

We prefer this latte'r approach.

A growth limit of 50% of critical crack size has been the NRC criterion; consequently, an acceptable inspection schedule criter'fa is maintained as 4

follows:

1) New discs should be inspected at the first refueling outage, or be-fore any postulated crack would grow to more than 1/2 the critical depth.

2) Discs previously inspected and found to be free of cracks or that have been repaired to eliminate all indications should be reinspected using the same criterion as for new discs, calculating crack growth from the time of the last inspection.

3)

Discs operating with known and measured cracks shculd be 7einspected before 1/2 the time calculated for any crack to grow to 1/2 the criti-l cal crack death.

4) These inspection, schedules may be varied to, coincide with scheduled cutages. Westinghouse recommendations in tMs regard should be followed.

i Su=ary and Conclusions r

i A.

We agree that the Westinghouse crack growth rate equations for bore and keyway cracks are acceptable.

We agree with the alternative Westinghouse critical crack size calculational 3.

l method, using a value of fracture toughness increase of 20% above the Rolf-Novak value.

i t

I 4

I C.

We will retain a criterion of relating allowable running time before l

inspections to the time to reach 1/2 of the critical crack depth.

i O.

The NRC staff will no longer monitor each turbine inspection except l

through the normal activities of the Office of Inspection and Enforce- -

.i

ment, i

e w

a q

s k

i i

l f

e i

O j

l i

f 4

r 4

G

-g y--

--r--

,-pmm-e Wm-=

e.-.,'He,

.v,.ce.-,

4

+..w,,

, - +, -.

y y

,w,=wew e-w

--.e-.

.e,

- -g-

--r

--er-

-c-v w