ML20031F064
Text
e, UNITEO stall *
~*
.+
NUCLEAlt itEGULATORY COMMISSION 8
., j
.I
%"fls., o s. * /. :
WASHINGTON, D. C. 70555 APR 8 1976 l
I n
6 r.~. -
- 3. ~
> II. J. Kouts, Director C'<
- Office of Nucicar Regulatory Rescarch CONFlRNI. TORY RESEARCli FOR EXPERIMENTAL DETERMINATION OF FAILURE PRESSURES
~
OF STEAi-t GENERATOR TU13ES NRR requests'RES to experimentally evaluate the margins to failure of
(
degraded PWR steam generator tubes under normal operating and postulated 6--
cecident co.. 'iticas.
This task is proposed to be accomplished in two hrj pha sr.s. The first phase which is described in the following paragraphs I**-
~
consists of hurst and collapse tests of artificially defected tubing La a simulated steam generator environment and determination af leakage 8
rates through cracked tubes. The results from the first phase are needed expeditiously to enable an independent verification of test I
data provided by the three major U.S. FWR reactor vendors, and to resolve the apparen". inconsistiencies in their results.
NRR will
~
~
utilize this data in determining acceptable design allowances required in processing reactor licensing applications.
The second phase of the h *..
program would include tests in such areas as combined membrane and i~~
bending, fatigue properties of degraded tubes, and critical fracture
'r====
toughness data.
This part of the program may be formulated and described at a subsequent date. The results of this second phase may s
he provided over a longer term than the first phase.
Status of the Problem b
(
5 team generator tubes constitute a major portion of the reactor coolant l
pressure boundary.
Degradation of the tubes could result in either y--
l thru-wall penetration of the wall or rupture of the tube, which would g
release radioactive coolant into the secondary system and thence into l
the environment.
In the event of a Loss of Coolant Accident (LOCA) l a thru-wall pcactration or rupture could release steam from the I
secondary side into the containment or into the reactor vessel.
Therefore, it is important that the margins to failure of degraded tubes he known, and that design allowable stresses be established with y
a high degree of confidence.
g
),.
The three major steam' generator manuf acturers use four dif ferent tube L,,
.{
sizes in i; heir different designs.
Only We tinghouse and Combustion m
. Engineering have conducted collapse and burst pressure test programs
[l with artificially defected tubes. These tests were conducted in
.)
L-l
^
o
(* **.*.
1 l
rec===
i ir -
,'h11090118810827
,19DR 1A i WDELL 1-313 PDR i
-.3
a 2
+
6
?
1 II. J. Kouts
- connection with specific plants, for exampic R. E. Cinna,11. B. Robinson and Palisades. No experimental test data exists regarding the margins of safety of B & W steam generator tube size.
For the purpose of cstablishing design allowable stresses, it is necessary to obtain
})*j statistically significant results over a wider range of tube degradation parameters.
Jnformation Deficiencies and Proposed Program f
~
Based on an evaluation of the test data obtained by Westinghouse and l'
~~
Combustion Engineering on the collapse and burst strengths of their I
steam generator tubes, it is apparent that additional testing is
{ **f{},
necessary within the next year.
NRR strongly recommends tha t an NRC g'.
funded test program should 'be developed in the following areas:
5----
'l. Burst and Collapse Pressure Data I
I Provide an independent verification of the experimental predictions of burst and collapse pres'sures of Incodel 600 (ASME SB-163 Alloy
. 600) degraded steam generator tubing under biaxial tensile and i
compressive loads on the four sizes of tubing (.875 in. OD x 0.05",
0.75" x 0.043", 0.75" x 0.048" and 0.625 x 0.034") used in the Westinghouse, Combustion Engineering and B & W steam generators.
I'*"
The test specimens should be artificially defected to simulate general thinning due to wastage, stress corrosion cracking and superimposed cracks on thinned regions. The depths of penetration of both the cracked and wastage type defects should range from 80 percent remaining wall thickness to thru-wall defects.
The number of j
tests should be sufficiently large so as to be statistically signif-icant. Westinghouse an ! Combustion Engineering have used diff erent types of artificial defects to simulate thinning due to wastage.
A standardized artificial defect should be established for simulating h.
vastage, stress corrosion cracking and superimposed cracking on g====
l l
thinned tubes. A uniform rate of loading should be established for l
both burst pressure tests and collapse tests to simulate transient P.
loadings during postulated accident conditions. The burst pressure l.'
tests for thru wall cracks should be carried beyond the point of a
(
plastic bulge appearing around the simulated defect. The rupture pressure when the cracked section folistes and has an area opening
(.
at 1 cast equnt tu the originhl tube cross section, should be
(
obtained in these tests. This information has not been obtained l
by either Westinghouse or Combustion Engineering in their test
(*
programs, and it is considered highly significant in determining a,
actual margins to failure.
7:..
t-.
t,e I
- i P
\\
3 i
t
..d m
l
~
.. ~,
' ~
APR r 076 l
11.
J,. Kouts 3,-
I The test data should be obtained at temperatures and other conditions prevailing in a steam generator environment. A majority of the test results provided by the vendors were obtained at room temperatures.
f.f.',..
C.*
2.
Leak Rate Tests Leakage flow rates from primary side to secondary and vice versa i
should be obtained for a range of cracked geometries under operating.>
conditions as well as postulated accident conditions. These tests should be conducted at temperatures and wf.th chemistry conditions prevailing in a steam generator environment.
3.
Determination of Material Properties of Tubes Prior to Start of Test M ~,,'
Program Inconel 600 tubes purchased prior to August 1971, as per ASME Code l
Section Ill - SB 163, were required to have a minimum yield strength of 35,000 psi. This specification was modified by Code Case 1484 in August 1971, and the minimum yield strength requirement was changed to 40,000 psi with a maximum of 65,000 psi.
The procurement specifications should identify the materials for the test specimens. Consideration should also be given to dif ferences between actual material properties g'%
q.
versus specified minimum values, variances in properties from dif ferent sources and variances from the same source such as lot-to-lot variations.
y I
Additional considerations that should be accounted for in this initial test program are the design flexibility and instrumentation to allow l
subsequent investigation of the following:
1
~.. N.
[
a.
Determinat on o Ld.e_g aded tub.e_s_trengths under, combined pressure T
{._..
and bending loads.
b.
Determination of the collapse strength of degraded tubes with
,C""
(
specified ovality.-
l{,,
c.
Determination of critical fracture toughness and other parameters
>j.
for Inconel 600 required in analytical calculations of margins to failure.
.m.
t.-
These items will be included in the second phase of the test program.
p Completian Ikite Initial program results consisting of collapse and burst test data should be available to NRR in 6 to 9 months. The leek rate data should be y#.,
,~4 completed.within one year.
Results of ongoing tests should be provided to NRR as they become available.
~'
y,..
t g-I 1
.t
.......I.*_...
M
t
.; r.....
II. 4. Kouts 4-I Licensing impact t
^
At a recent ASLB hearing the board raised questions regarding the failure
{;
pressures and design, margins for steam generator tubes under various I (;_,
f<
operating and postulated accident conditions. These and similar questions which have a significant impact on the licensing process can only be
'satisf actorily answered on the basis of experimental data obtained using uniform and consistent procedures for producing artificial defects and.
I testing under simulated accident conditions.
I i
l:,
The information to be generatcJ by the proposed program is fundamentally applicable to all steam generator.i of PWR plants. Design allowances for
~ '
current steam generator tube designs are evaluated on the basis of the j!
limited experimental and theoretJeal dat availabia at this time.
i It is anticipated that once the margins to tailure of degraded steam generator tubes have been determined more precisely ac a result of this proposed
~1
. test program, the design allosances can be evaluated on a more realistic
,;. basis.
l Magnitude of Research Effort g
t Staf f members of the Mechanical Engineering Branch, Division of Systems h?' "
bafety, have had preliminary discussions with David Tayle..r Naval Ship I
R & D Center, Annapolis, MD. and have visited their facilities.
We believe that DTNSRDC, Annapolis has the test facilities, experience and personnel necessary to ccnduet an ind.cpendent confirmatory research program to meet our needs. They P.ava expressed their willingness to l
conduct tests within a time frame which will commence in FY 1976 and which is expected to be completed by March 1977.
It is estimated that
. F_ _
this initial phase of research effort will est approxis ely $150 K.
3
/
s l
g
^
v.
i, B. C. Rucche, Director i
Office of Nuclear Reactor Regulation
/,,
t cc:
R. E. Ileineman, SS
~;
R. R. Hacenry, SS" J. P.. Knight, SS J. k. Hiller, NRR l
S. S. Pawlicki, SS i-W. F. Anderson, SD L. Frank, SD W. S. !!azelton, OR R. f. Bosnak, SS
- g-
~.
L.
S.- Rubenstein, NRR I
H. J. Faulkner, SS l
J. R. Rajan, SS t
F e
-t
- + -
,.,y-m
- - -m-,
---,-w---r--g i----
-up n
v