ML19344B556
| ML19344B556 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 10/17/1980 |
| From: | Hukill H METROPOLITAN EDISON CO. |
| To: | Brunner E NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| References | |
| TLL-509, NUDOCS 8010210453 | |
| Download: ML19344B556 (4) | |
Text
F 4
A;
. 'm, 7 2: jf Metropolitan Edison Company v':'"~~.r* WJ -,
('-
Middletown, Pennsylvania 17057 Post Office Box 480 i
- m..m.4.~.maan
/
717 9444041 l
Wnter's Direct Dial Number l
l
' October 17, 1980 TLL 509 Reactor Operations -& Nuclear Support Brmch
. Attn:
E. J. Brunner, Chief U. S. Nuclear Regulatory Commission Region'I-63.' Park Avenue Kitr.; of Prussia,- Pa.
19406 i
Dear Sir:
Three Mile Island Nuclear Station, Unit 1 (TMI-1)
Operating License No. DPR-50 Docket No. 50-289 l
OTSG. Auxiliary Feedwater Nozzle Thermal Sleeve / Collar Weld Attachment l
The following information is in response to your inquiry of June 10, 1980 concerning the above mentioned subject.
1.
Design / environmental conditions leading to possible degradation of subject veld integrity.
Identification of probable failure mode.
Injection of. condensate or' riser water into a nozzle thermal sleeve (Figure 1) at steady state temperture will cause thermal shock.
Translating flow'conditious into stress, transient thermal conditions create peak principal stresses high enough'to produce crack initiation in-as little as 10 cycles for a temperture differance, between the sleeve and the collar, equal to.5000F. About 100 cycles can be anticipa*ed for a-temperature difference. equal.to'2500F.- Each of these estimates is based on the ASME, BPVC,Section III, Design Fatigue Curves. These temperature
~
differences are realizable as a result of short bursts of cold emergency feedwater'or more sustained flow'using somewhat warmer emergency feedwater.
t
- One emergency feedwater usage represents one thermal fatigue cycle f
regardless:of how many times feedwater was called for during the emergency event or what-the feedwater temperature was.
The basis for this assumption is that once t'.ac thermal sleeve was thermally shocked, to 3 peak stress field is constant.
2.
- Interict action taken, if any.'
We will partially drain the OTSG and inspect Z-axis no::zles.
If cracking i 4,
is noted on these nozzles, -the remainder will also be inspected and weld a
repaired as required. Z-axis nozzles ~were inspected in 1977_and found L
free of: cracks.: Only two. thermal cycles have been recorded since these inspections.
soto s1gemmmm_p pn gg smm._
s
-e E. J. BRUNNER TLL 509 3.
Possible long term modifications.
No long term design modification is anticipated at this time.
Z-axis AFW nozzles will be removed and blanked due to conce:n for potential OTSG tube inspection lane flow impingement.
4.
Safety consequences B&W evaluation (TMI-80-053 dated March 17, 1980) of the auxiliary feedwater thermal sleeve / collar weld attachment failure finds that decay heat removal is not impaired if the nozzle is short of the shroud due to either assembly tolerance or postulated backing into the external piping elbows after thermal sleeve separation.
This B&W evaluation does not include the safety consequences of water filling the steam lines.
The safety consequences of static and dynamic conditions resulting from water accumulating in the main steam line can be addressed in terms of the following terms:
a.
Unbalanced Forces at Elbows Auxiliary feedwater leakage may result in moisture carryover into the main steam lines. While the mass of the water content may be low (8 1/2% leakage, maximum), its velocity is high resulting in high unbalanced forces. Steam hammer pipe restraints will react the unbalanced forces.
b.
Filled Main Steam - Deadweight and Increased Pressure If the feedwater fills the main steam lines, the additional deadweight can be accommodated by the existing supports.
Internal pressure in the main steam line will increase because the feedwater pressure exceeds the steam pressure.
The main steam lines are shown to be structurally sound by hydro test to 1.5 times the design pressure.
c.
Dischar2ing of Safety Valves Due to increased pressure identified in Item 2, above, the spring loaded safety valves can be expected to open resulting in higher than normal discharge forces due to pressure and density effects.
Margin is provided via the thermally compensated elbow supports.
Even if the nominal safety valve reactions are exceeded the thermally compensated supports will protect the structural intearity of the main steam piping.
Each of the thermally compensated supports was designed for 55 kips of unbalanced dynamic force as per USAS 331.1-1967.
d.
Seismic Stresses For a full main steam.line, it is likely that the resulting seismic stresses will exceed the allowable values of USAS B31.1-1967. However, the pressure boundary may still be L.__
O E. J. BRUNNER TLL 509 preserved due to the existence of steam hammer restraints and pipe deflection limit devices such as rupture restraints and concrete penetrations. Furthermore, simultaneous occurrence of the cubject overfill and seismic event are beyond the B&W safety concern and has a very low probability of actual realization.
Sincerely, get'] M.D. '.'.NI!!
H. D. Hukill Director,DfI-l HDH:DCM:hh Attachment cc:
J. T. Collins R. W. Reid B. H. Grier D. Dilanni H. Silver
.o D"D
- D'3~Y
.2.k o
66-e OTSC VEcSEL t'ALL C
l e
COLLAR 1/8" SLEIVE I
.f z
i NN l
-_)
q
/'[j 1 /,
Y I
I EISER ELEOW R!
CEA'TERING RING -
i
. !LLIARY FEED *w'ATER ptAggg uu i
ELEXITALLIC GASKET c nwm AUXILLIARY FEEDWATER N0Z::LE/ COLLAR /THgpKM SLEEVE CONFIGURATION FIGURE 1
.