ML20035E721
| ML20035E721 | |
| Person / Time | |
|---|---|
| Site: | Mcguire |
| Issue date: | 04/07/1993 |
| From: | Mcmeekin T DUKE POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9304190143 | |
| Download: ML20035E721 (9) | |
Text
_ _ _ _ _ _ _ _ _
11 1
Duke Iburr Company (704)$i5-4000 M:Guirr Nucircr Station I??00 Ikpers Terry Rnd Huntersvitie. NC280;S.FM5 DUKEPOWER l
April 7, 1993 U.S. Nuclear Regulatory Commission Document Control Desk Washington, D.C.
20555
Subject:
McGuire Nuclear Station Unit 2 Docket No. 50-370 Licensee Event Report 370/85/24 Gentlemen:
The purpose of this letter is to provide the basis for relief / amendment of the McGuire Nuclear Station LER commitment for adjusting AT 1 percent conservative prior to unit start-up.
This document will detail the new AT evaluation process, which will begin during IEOC8, and demonstrat(. that this process is more comprehensive than the existing process as required by LER 370/85-24.
Very truly yours, YJ T.C. McMeekin TLP/bcb xc:
Mr.
S.D. Ebneter INPO Records Center Administrator, Region II Suite 1500 U.S.
Nuclear Regulatory Commission 1100 Circle 75 Parkway 101 Marietta St., NW, Suite 2900 Atlanta, GA 30339 Atlanta, GA 30323 Mr. Tim Reed Mr.
P.K. Van Doorn U.S. Nuclear Regulatory Commission NRC Resident Inspector Office of Nuclear Reactor Regulation McGuire Nuclear Station Washington, D.C.
20555 1
9304190143 930407 PDR ADOCK 05000370 S
pg l
d l
ATTACHMENT 1
SUBJECT:
1 Percent Conservative AT LER Commitment Relief / Amendment Existing AT Evaluation Process:
Outage /Startup AT Adiustments:
AT is adjusted 1 percent conservative prior to unit start-up (based upon AT value at unit shutdown).
AT is adjusted to linearized 100% value, as determined by NC System flow balance, typically within 14 days after unit start-up.
On-line AT Adjustments:
No on-line AT evaluations are performed.
New AT Evaluation Process:
Outage /Startup AT Adiustments:
PT/0/A/4150/21 (Reactor Group's controlling procedure for unit start-up) has a hold point at 90% power level, where a preliminary evaluation of AT values is performed.
Based upon this evaluation, new AT data will be placed in the Plant Data Book and AT will be adjusted as needed prior to proceeding Unit start-up.
PT/0/A/4150/40 (Reactor groups AT, etc. analysis) will be performed shortly after unit start-up (after Venturi fouling constants are removed) to analyze 100% AT values.
Based upon this evaluation, new AT will be placed in the Plant Data Book and AT will be adjusted as necessary.
IP/1,2/A/3000/05C, D, E, and F (AT/Tavg Loop Calibration Procedures) will be revised to use Plant Data Book data for loop calibrations.
i On-line AT Adiustments:
PT/0/A/4150/40 will be implemented quarterly and will determine actual AT values.
If the new AT data differs from the 'old' AT data by greater than 0.5 degrees F, the new data will be placed in the Plant Data Book.
The 7300 Monthly calibration procedures (PT/1,2/A/4601/01,02,03, and 04) will be revised to calibrate AT loops per values per dcta in Plant Data Book.
This will result in quarterly AT adjustments where necessary.
i 8
l O
t There are three major advantages of the new AT evaluation process over l
the existing process:
1.
AT values will be evaluated prior to start-up (090% hold point) and adjusted as necessary prior to continued start-up.
i 2.
Actual 100% AT values will be known and the need for loop calibration determined more closely following full power operation.
t 3.
AT values will be evaluated, and possibly adjusted quarterly during fuel cycles.
l Based upon the arguments presented above, the requirement for the 1 degree conservative AT adjustments prior to start-up will be revised to take credit for the new process adopted at McGuire Nuclear Station.
t I
d I
l l
I I
i 1
i 1
l
ATTACHMENT 2 r3.
. co _
no e o,...o LICENSEE EVENT F.EPO~.T (LEN 4
.&CsLITv AnahAE fu DOCESTERJuSRA @)
F&st g McGuire Nuclear Statten - Unit 2 015 I o !o ! o I 117 h 1 l0Fl Ol 5 Yetk8 #es Gradual Decrease in indicated Full Power Delta-T sys=v pave si j
6am munsasm isi espost para m crase eaciuriss,vo6ves ei wo=v-ca. l
.s.
j vaan "tg f
- g g wo v.
ca.
. sam
- *csiv
=*=as pocasv =vueen.si 01 510lolol ! I 1l 0 1 l 4l B LS 8l 5 0l 2 l4 0 l0 1l1 1l3 8l5
~
~
0 1 5 1 0 1 0 1 0 t i t v is me one e sue =tero eva wa=t to r-e asoviasme=rs e, is e,
,,c
, eni
=a** *n 1
se m.,
m m n.au.
rttimi m m.nuin u ww.n m rewai..,
vanw 1 10 t0 m a== nim
= mwa' asnaura=>
= a;s,;,;,g,,,,,
ni se ass 4.mnw
{
se tsweue en73 wen.enu Jesas
.n u,
. n i=>=
. n..n N Genifittoi es 73een@Heisi M 736ae@Hal ucswas co= Tact som tus ten esas maus tatsraons =vuosa
- ta cope Tem 9.
The v - T f r paef w 7 I Al L 1 1 71 1 l_ l T in f '3l 7
~ conspLars oma tems som taca conspores=v estuma otecaean i= two mesce? nai
"' Q C
%%'N' wh%ac ga ne causa svr'su conspo=a =T cassa systew coemacsamt l I !
! ! I l
! ! I l I f f
I I i l i I i
I l 1 I ! !
su ts
=v t.., a meten n.
wo.v.
c.,
w.. is.o.
'""1 vis,,,- - en.se-se s,-ss,o. e re, T) o non~ e se, +~.,...x
~. ~i n
. ~- -
n,
Following the startup of Unit 2 Cycle 2 in May 1985, a gradual decrease in the indicated value of the full power Delta-T vas identified. The indicated Delta-T decreased linearly in each loop by approximately 1 degree-F. Delta-T is used as a measure of reactor power for both the overpower and overte=perature Delta-T reactor trip setpoints. The decrease in Delta-T has caused these dynamic reactor trip functions to be improperly scaled in a non-conservative direction.
Two potential causes for the decrease have been identified: a change in hot leg te=perature streaming patterns which supply coolant samples to the temperature sensors; a reduction in thermal power, possibly caused by fouling of the feedvater flow venturi meters.
The channel errors have been analyzed to determine the impact on FSAR accident analyses. The results of this evaluation indicate that only a narrow range of steam line break accidents may have exceeded the design basis during this incident.
Calibration procedures will be developed to set Delta-T values for each new fuel cycle; and the possible feedvater venturi fouling vill be investigated.
..c.- m.
- $ $31
Get See m y3 aIWCLE2A 848ULATCAT % '
UCENSEE EVENT REPORT (LERI TEXT CONTINUATION
.- onc aue =n veo s===es ure
'aceute =easetu pocaat emmesa op
" M *,;.* 6 ! Om; e.=
McGuire Nuclear Station - Unit 2 o l5 lc jo le l3 l 7l0 8l5 0l5 0 l 2l 4
- 0 l0 0l2 or w.-.
-==c a.- an. wm Following the startup of Unit 2 Cycle 2. a gradual decrease in the indicated value of the full-power reactor coolant temperature difference was identified. During the eight weeks of full-power operation following the cycle 2 refueling, the indicated Delta-T had decreased linearly in each of the four loops by approximately 1 degree-F. Delta-T is used as a measure of reactor power for both the over-temperature and overpower Delta-T reactor trip setpoints. The decrease in Delta-T had caused these dynamic reactor trip functions to be improperly scaled in a non-conservative direction. The Delta-T channels were underpredicting full. core potter by as much as 5%.
In evaluation which followed the discovery, two potential causes for the Delta-T decrease were identified:
- a change in the hot leg temperature streaming patterns which uupply coolant samples to the temperature sensors.
a reduction in thermal power, possibly caused by fouling of the feedvater flow venturi meters.
The Delta-T channel errors have been analyzed to determine th'e impact on the Final Safety Analysis Report accidents which take credit for the overtemperature and overpower trip functions. The results of this safety evaluation indicate that only a narrow range of steam line break accidents (0.4 to 0.9 square feet breaks) may have exceeded the design baris during this decreasing Delta-T incident.
Background
Delta-T is used as a measure of reactor power and is calibrated (scaled) to read 100% as determined by the precision secondary system calorimetric measurements.
The scaled value is used in the overtemperature and overpower Delta T core protection circuits.
The overtemperature Delta-T trip actuation circuit loops automatically vary with:
- 1) coolant average temperature, 2) pressurizer pressure, and 3) axial power distribution to protect the reactor core from departure from nucleate boiling (DNB) during certain design transiests.
The overpower Delta-T trip circuit loops automatically vary with:
- 1) coolant temperature, 2) rate of change of average temperature, and 3) axial power dis-i tribution to protect the fuel from excessive heat generation rates (KW/ft).
The overtemperature and overpower Delta-T trip setpoints are detailed in Technical Specification Table 2.2-1.
..e,o.
=
$433
i a
= ace. me.
va muetaan.. v6aro, cow LK:!NSEE EVENT REPORT (LER) TEXT CONTINUATION
.-aoveo o.e =n va-e=
i esawe j
e.c.ur.
o c. 1 u. m j
m:=n.:.=
j 012l4 0] O Ol3 or 0 ;$
McGuire Nuclear Station - Unit 2 0 !s lc j o l o 13 l 710 85 rurr r a ama nn t
Description of Jvent I
The results of the evaluatior. have identified the two possible causes of decreased i
4 l
Delta-T as:
- 1) changes in the hot leg streaming patterns, and 2) fouling of the feedwater flow venturi nozzles.
The hot leg streaming patterns are results of fuel assembly exit temperature diffsrence. The high flow rates and differences in individual fuel assembly outlet temperatures cause temperature stratification within the Reactor Coolant system i
piping. A special device is installed in the hot leg pipir.g which samples the reactor coolant at different areas within the piping. These samples six and pass by the hot leg temperature sensor to provide an average hot leg temperature.
The fuel assembly exit temperatures did change during the period when the Delta-T
[
was decreasing. This is an indication that the temperature streaming patterns in the piping also changed. The evaluation states that it is possible-for a 0.5 degree-F decrease in Delta-T on Unit 2 based on the data provided. This would account for about half of the total decrease.
l The venturi nozzle fouling is a result of a crud buildup on the feedwater flow sensing device. The buildup affsets the flow readings obtained for the heat balance measurements. The venturi is a device which develops a differential pressure when flow is passed through it.
This differential pressure can be accurately scaled to indicate flow as long as the venturi dimensions are not altered. The crud buildup inside the venturi affects the accuracy of the developed
- essure across the vent tri.
Venturi fouling has poss1My caused a Delta-T
)
cc. crease of approx 1L.ately 0.5 degree-F at McGuire.
The total 1.0 degree-F Delta-T drif t caused by hot leg streamng and venturi nozzle fouling was within the uncertainty allowances assumed in the safety analysis.
The additional errors associated with e at calibrating the circuits during the cycle 2 startup caused the total Delta-T uncertain *,f to exceed the uncertainty allowances assumed in the safety analysis. The actual error for each loop in units of percent fuli pt_ver was:
Loop A Loop B Loop C Loop D 2.5% F.P.
5.3% F.P 5.5% F.P 3.4% F.P These errors have been evaluated for their impact on certain Final Safety Analysis Report accidents. The evaluation results indicate that only a narrow range of steam line break accidents (0.4 to 0.9 square feet) may have been in violation of the design basis.
..e o.., mu S@
4 d
.. ~,.
LICENSEE EVENT REPORT (LERI TEXT CONTINUATION
.a aoveo ow.a vu+o.
l memes w e I
i
..c.u r.
c....,
t...,.
a, "M:-
"':.l*::
\\
"=
l i
l McGuire Nuclear Station - Unir. 2 o is l0 l 0 l s 1317l 0 8 I 5 0 l 214 0l 0 0 l 4 0FQ f
ver,-
c
.m
~
The hot leg streaming patterns have been identified to the industry as early as 1
1968. The RTD Bypass System, which is installed at McGuire, was developed to i
reduce the effects of the temperature gradients in the piping. Westinghouse i
reports that no significant problems have arisen in the industry pertaining to the streaming patterns although some variations of 0.5 degree-F or less have been 1
r._orted.
j i
The feedvater f1owventuri nozzles have been reported by several nuclear plants as indicating a fouling condition. The stations affected by this condition have taken measures to compensate for the errors and are monitoring the plant parameters.
i The omission of the calibration of the Delta-T channels during power escalation has affected both Unit 1 and Unit 2.
Although the decreasing Delta-T in Unit 2 brought the problem into view, the Unit I channels had not been calibrated during i
the startup of each fuel cycle. This requirement was not clearly defined to the
{
station personnel. The channel calibration was being performed on an 18 month l
l basis without regard to the cycle startup requirements.
l l
2 i
CORRECTIVE ACTIONS:
j 1mmediate:
None i
i Subsequent:
1.
The results of the reactor coolant flow :est and how it affected l
the Delta-T circuits heve been reviewed.
l 2.
A work request to recalibrate the Delta-T circuits was initiated.
3.
Westinghouse reviewed the key plant parameters to evaluate the Delta-T drift and submitted a report to Duke Power.
l Planned:
1.
A calibration procedure to set the Delta-T to a conservative value prior to reactor startup at the beginning of each cycle i
will be established. Necessary controls will be implemented to ensure the work is completed prior to reactor startup.
2.
A calibration procedure to set the Delta-T values for the new fuel cycle based on final measured Delta-T values will be i
i established. Testing will. determine the actual 100% power level.
3.
A program to monitor the full power Delta-T during the entire fuel cycle and to report significato: temperature drifts to the Reacter Safety groep for analysis will be developed. Data has already begun regarding this program.
4.
Measures will be taken to identify the ettent of the venturi fouling and initiate actions to compensate for the apparent loss of power. Alternate tests have been tuitiated to determine the extent of error, A
meC 80** 3888 19416
. - ~. - - - - -. - - -,,
._n I e m.
v2 aucum..vt.,es, ca 7
i UCENSEE EVENT CEPG!1* <LERI TEXT CONTINUATION
.eso o a vio.co.
DFIRE 8/31 S
..c L.
sec.sv won m un uinea.
.. a 43
"'q i
aM,;p i
0l 0 0l5 or 0l5 j
McGuire Nuclear Station - Unit 2 o 15 lo lc j o l @ 7l0
$5 0 l 2l 4 runt = -
m.mc r
. nn I
SAFETY ANALYSIS i
I The evaluation stated that the approximately 1 degree-F drif t associated with the i
Delta-T channels is within the uncertainty allowances assumed in the safety analysis.
i Therefore, the precision heat balance performed in June may be considered valid.
t in addition, since the error associated with the calculated flow is within the j
uncertainty allowance assumed in the safety analyses, there are no flow related l
Tech Spec violations associated with the Dsita-T drif t incident. However, when the j
i degree-F drif t is added to the errors associated with not calibrating the Delta-T i
channels during startup, the total error associated with the Delta-T channels is greater than the uncertainty allowances assumed in the safety cualyses.
j 4
1 4
The Delta-T channel errors can potentially impact the TSAR accidents which take
{
credit for a reactor trip on the overtemperature or overpower Delta-T trip I
j functions. There is sufficient margin included in the overtemperature Delta-T l
setpoint calculation to account for the channel errors. Therefore, che accident j
analyses which take credit for a reactor trip on overtemperature Delta-T (RCCA j
Withdrawal at Power, RCCA Misalignment, and Boron Dilution) remain valid. However.
I for the overpower trip function, there is not sufficient margin in the setpoint i
calculation to account for the channel errors.
s I
j lt should be noted that credit is not explicitly taken for the overpower Delta-T 1
trip function in any of the accidents analyzed in Chapter 15 of tho McGuire FSAR.
t However, WCAP-9226, " Reactor Core Response to Excessive Secondary Steam Releases".
]
shows that the overpower Delta-T trip function may be relied upon to provide DNB l
protection for some steam line breaks at power.
Plant-specific analyses have been performed for a spectrum of intermediate steam line breaks at power. It sus determined that, for brea.k sizes equal to or greater l
than 0.9 square feet, reactor trip occurred on S1 actuation on low steam line l
pressure. For break sizes less than 0.4 square feet, either no trip was necessary.
or a trip occurred on low-low steam generator level. For breaks inside containment in the size range of interest. reactor trip occurs on SI actuation on high 4
j containment pressure. Thus, the overpower Delta-T trip function provides the J
reactor trip only for breaks between 0.4 square f eet. and 0.9 squ,re feet outside j
containment.
i There were no pipe break events during the period which would have affected the
',i health and safety of the public.
It is also-possible tb:st other trip functions j
such as the high flux trip, or trading off available aaalytical margin may have
{
revealed acceptable results for the narrow range of affected steam line breaks.
i 4
2
..c...
m.
19 431 1
..,y
,