ML20095J441
| ML20095J441 | |
| Person / Time | |
|---|---|
| Site: | Fort Saint Vrain  |
| Issue date: | 08/13/1984 |
| From: | Lee O PUBLIC SERVICE CO. OF COLORADO |
| To: | Johnson E NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| References | |
| P-84275, NUDOCS 8408290256 | |
| Download: ML20095J441 (7) | |
Text
_
l l
(
PUBLIC SERVICE COMPANY OF COLORADO
[
P.
O. BOX 84O DENVER.
COLORADO 80201 August 13, 1984 OSCAR R. m Fort St. Vrain Unit No. 1 P-84275 32@20MMl%
Mr. E. H. Johnson, Chief
)
'i Reactor Project Branch 1 E 2 21984 Nuclear Regulatory Commission k
Region IV j
h 611 Ryan Plaza Drive, Suite 1000 Arlington, Texas 76011 DOCKET N0. 50-267
SUBJECT:
Response to NRC/LANL Concerns on Cracked Fuel Elements
REFERENCE:
NRC Letter from E.H. Johnsen to
- 0. R. Lee Dated May 11, 1984 (G-84158)
Dear Mr. Johnson:
Please find enclosed forty (40) copies of the report, " Response to NRC/LANL Concerns Regarding Cracked Fuel Element Integrity."
GA Technologies Inc. and Public Service Company of Colorado have reviewed the technical issues raised by LANL and have concluded that these issues pose no problems with regard to the performance of fuel elements with cracked webs at Fort St. Vrain.
This report fulfills part (1) of the requirement in your letter of May 11, 1984 (Reference 1).
If you have any further questions on this matter, please contact Mr. M. H. Holmes (303) 571-8409.
Very truly yours, Oh o!aohjj7 PDR
- 0. R. Lee, Vice President Electric Production ORL/SH:pa d )g Attachment y guFN,p.ar I
4
[
y
.c RESPONSE TO NRC/LANL CONCERNS REGARDING CRACKED FUEL ELEMENT INTEGRITY
=
~
Introduction
~
The - Los Almans National Laboratory (LANL) report of April 30, 1984,
" Fort St. Vrain Fuel Elements," presents a review of material sucait-ted to the NRC by PSC regarding two Segment 2 fuel elements with cracked graphite webs. In the report summary, LANL concurs with PSC's information~ regarding the likely cause of the cracks and the low proe-ability of extensive further cracking under normal operating condi-tions.
LANL also concurs with PSC's position regarding the icw proe-ability of cracks affecting the integrity of the reactor fuel itself.
However, in its discussion of cracked fuel element integrity LANL raises two concerns associated with off-normal conditions:
1.
The effects of the thermal stress field in combination with static or dynamic loading of the cracked element have not been addressed.
2.
The effects of dyn.unic loading through the dowel. socket system on the stress field in the interior of the cracked element and on pctential crack progression have not oeen addressed.
These LANL concerns are addressiid in the following discussion.
Thermal Stress Field Effects LANL expresses the concern that the static loading tests perforned by GA on unirradiated H-327 graphite slabs with simulated cracks, "do not 4
l account for the presence of a strong, thermal stress field in the specimen, nor do they account for the possibility that the crack could reduce the strength of the element under dynamic loading conditions."
,V.
Calculations have been performed oy GA in' which the stress analyses of the cracked fuel element presented to the NRC on April 4, 1984, in Bethesda (Ref. 1) were perturned oy analytically imposing a static compressive load on the l' bel element at the time of peak calculated stress.
Crack configurations from zero to five cracked wees were assumed to exist. The pattern of these assumed cracks was the same as that observed for the three cracked wees in element 1-2415, or an extrapolation thereof. (Ref.1).
The maximum seismic load on a FSV fuel element during a Design Basis Earthquake (DBE) (0.1 g c:aximum horizontal ground acceleration) has oeen determined to be 1500 10.
This load is determined from the weights of individual fuel and reflector elements that can act upon a single element during a seismic event and from the 0.26 g maximum lateral acceleration experienced in the core during the DBE (Ref. 2).
Accordingly, a 1500 10 static compressive load was used in these analyses.
The results of these calculations are summarized in Taole 1.
The
+
taole shows that the largest calculated increase in the peak in-plane stress / strength ratio resulting from imposition of a 1500 lo seismic load on the cracked element is 0.02.
This increment is a minor per-turbation on the stresses resulting from thermal and irradiation-in-duced strains. It is concluded from these analyses that imposition of the maximum seismic loads resulting from the DBE on the thermal and t
irradiation-induced stress fields in the cracked fuel element has a negligiole effect on fuel element performance.
While the loading tests conducted on unirradiated H-327 graphite slaos and the calculations described acove noth involved imposition of static loads on the cracked l' bel element, experimental evaluations of HTGR fuel element seismic strength have shown that fuel element per-formance under Doth static and dynamic loading conditions is essenti-1 l
ally the same for relative impact velocities up to 120 in/sec (Ref.
3).
Relative impact velocities in the HTGR core during a seismic event are less than 120 in/sec. These tests have shown that:
l
.'.n 19 l
1.
crack patterns resulting from static and dynamic loading are nearly identical, 2.
strain behavior under static and dynamic loading, cotained from strain gage traces, is fundamentally the same, and 3
the magnitudes of the static and dynamic loads required to produce failure are the same (acout 70,000 lb) and are much larger than the 1500 lo FSV DBE seismic load.
It is concluded, therefore, that analytical representation of fuel element seismic behavior by imposition of static compressive loads provides a valid approximation of fuel element performance under seismic (dynamic) conditions.
This approximation is particularly valid in view of the large difference between the maximum DBE seismic load in FSV (1500 lb) and the loads required to produce fuel element l
cracking (70,000 lb).
TABLE 1 GFECT CF FSV DBE SEISMIC LOAD ON 4
CRACKED FUEL ELEMENT STRESS FIELD l
Peak In-Plane Stress / Strength Ratio l
Without With 1500 lo l
Numoer of Cracxed Webs Seismic Load Seismic Load 0
0 70 0.70 3
0 90 0 91 5
0.69 0.71
- /
Dowel-Socket System Loading
~ LANL contends that, "ducir.g a seismic event the Fort St. Vrain core (as currently constrained by the core restraint devices) will transmit dynamic loads primarily through the dowel pins and socket arrangement located on the ends of the fuel elements.
This dynamic load trar,sfer will produce a complex stress field in the interior of the element, and could subsequently cause cracks to further propagate, depending on the magnitude of loads being transmitted."
In fact, the behavior of the fuel elements under dowel-socket system loading is well characterized and has been a subject of previous PSC-NRC correspondence (Ref. 4) related to the core fluctuations investigaton.
As explained in Reference 4, when the FSV dowel-socket system is loaded to its ultimate capacity, failure occurs in the sidewall of the element between either the socket or the dowel and the nearept face of the fuel element.
(See Figure 1.)
No failure occurs in the dowel pin itself.
This sidewll failure typically initiates as two cracks at the base of the socket or dowel which, under continued i
application of the load, progress outward to the element edge to encompass a fragment that is about seven inches along the horizontal edge and three inches in the axial direction.
Tests on FSV fuel element geometries (Ref. 5) have shown that this failure mode is the same for both static and impact loading conditions, with the dowel-socket system strength being somewhat higher under i= pact loading.
This failure mode indicates that, rather than being a complex stress field in the interior of the element, the stress field resulting from
. dowel-socket system loading is (1) a relatively simple stress cone, and (2) relatively localized at the end of the element with regard to i
stresses of sufficient magnitude to cause graphite failure.
I i
i
\\
f
.~
e L
1 1
l l
I r
i i
l
)
l I
i FIGURE 1: Typical Dowel / Socket Failure
l
.O,
Q y ss y,
I It is possible that the presence of fuel element wee cracks such as those seen in the Segment 2 fuel elements in the vicinity of the "S" face dowel might reduce the ultimate capacity of that one (out of three) ' dowel-socket system.
It is imittely, however, that the
~
resulting localized stress field would produce any notable effect on the orack in the interior of the element.
Therefore, extensive crack progression throughout the element is not expected to result from dowel-socket system loading.
References 1.
Don W. Wargebourg (PSC) letter to John T. Collins (NRC), " Fort St. Vrain Unit No.1 Fuel Element Meeting," P-84104, April 6, 1984.
2.
Response to AEC-DRL Question 5.6, Amendment No.16 to Application of Public Service Company of Colorado for Construction Permit and Class 104 License for the Fort St. Vrain Generating Station, Docket No. 50-267, November, 1970.
3 L. Sevier, "LETGR Graphite Fuel Element Seismic Strength," GA-A13920, April 30,1976.
4 J. K. Fuller (PSC) letter to William Gammill (NRC), " Fort St.
Vrain Operations and Oscillations Testing," P-78174, Octooer 20, 1978.
5.
Puolic Service Company of Colorado 330-MW(e) High-Temperature, Gas-Cooled Reactor Research and Development Program Quarterly Progress Report for the Period Ending June 30, 1966, GA-7314, Septeamer 1966, pp 29-31.