ML20054E210
| ML20054E210 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 04/21/1982 |
| From: | Sieber J DUQUESNE LIGHT CO. |
| To: | Varga S Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20054E211 | List: |
| References | |
| ND1MSL:1685, TAC-48018, NUDOCS 8204260261 | |
| Download: ML20054E210 (14) | |
Text
.
o
- 1, 4
f ll h$
75 i
A1 M
Teiepnon (412)45e4000 E"ol'g,'((* "
April 21, 1982 C
Shipp6ngport, PA 150f7024 ND1MSL:1685
{
c A
Director of Nuclear Reactor Regulation M
United Stutes Nuclear Regulatory Commission ggN@ED y
I' Attn:
Mr. Steven A. Varga, Chief k.
9 l'*
. Operating Reactors Branch No. 1 2
APR23 84tWr g%r% g,2 A
.}
Division of Licensing Washington, D.
C.
20555 E
Rehehence:
Beaver Valley Power Station, Unit No. 1 U
Docket No. 50-334, License No. DPR-66 4
l Cycle 3 Reload Safety Evaluation l
Centlemen:
This letter responds to your staff request that we provide additional technical information related to the Rod Position Indication Technical Specifications forwarded to the Staff by our letter dated February 23, 1982 concerning the Cycle 3 Reload Safety Evaluation for Beaver Valley Power Station.
As I indicated in that conversation, we are developing an alternate Technical Specification which responds to the Staff's desire that maximum rod deviation be limited to i 24 steps including instrument error.
A draft copy of that alternate Technical Specification is included with this letter for your review.
I must point out that this draft Specification has not yet undergone the reviews required to be conducted by Duquesne Light Company prior to formal submittal to the NRC.
We are very interested in arriving at an appropriate solution to this issue as soon as possible and I assure you that we will cooperate in any way we can to achieve this end.
Our February 23, 1982 submittal contained a request for Technical Specification amendment which basically extended the present Technical Specification to include Cycle 3 and following Cycles of operation.
This Technical Specification basically allows a total deviation of rods (including instrument inaccuracy) of i 32 steps and includes provision for use of primary voltage measurements to determine rod position as a backup method.
We contracted with Westinghouse to perform a safety analysis which considered the 1 32 step deviation and the analysis which Westinghouse performed assumed a single rod misalignment from the group indication at this maximum deviation.
The Westinghouse analysis was similar to that used in Cycle 2 to determine the peaking factor penalties resulting from i 32 step misalignments of single C and D Bank rods.
Together with 82 04 2 60RGI d
Mr. S.
A.
Varga Acril 21, 1982 N'lMSL:1685 D
Page 2 fine mesh 2D-TURTLE computer model, a TURTLE benchmarked 3D-PALADON model was employed, using 2x2 meshes per assembly radially and 20 meshes axially.
Calculations were made at various power levels with control rods positioned at the appropriate insertion limits (Figure 1).
Single D and C bank rods were misaligned by 32 steps and the resulting effects on peaking factors determined.
The following factors were considered and adequate margin was found to exist when this analysis was reviewed:
Moderator Coefficient Doppler Coefficient Seff and Prompt Neutron Lifetime Normalized Trip Reactivity vs. Position Trip Reactivity Worth Shutdown Margin Overpower kW/ft Rod Insertion Limits Boron Dilution During Refueling Rod Withdrawal From Subcritical Single Rod Withdrawal Rod Withdrawal at Power Rod Out of Position Dropped RCCA Rod Ejection Steamline Break in addition to the reviews conducted above, examination of rod misalignment FaH, rod misalignment Fxy and rod misalignment FQ(Z) were performed.
With respect to rod misalignment FAH, calculations were done at BOL with both equilibrium xenon and non-xenon conditions considered.
FAH limit values were compared to a limit described by the following equation FAH (P) = 1.55 [1 + 0.3(1-P)] where "P"
is the fraction of rated thermal power.
As in Cycle 2, at significant power levels where DNB is a concern, the rod misalignment FiH is smaller thca the design basis FaH limit.
For pcwer levels below 20%, the rod misalignment FaH values exceed i
the proposed Technical Specification limit by 2.5% or less.
For these low power levels, however, the core limits are vessel exit boiling limited and are not a function of peaking factors.
Consequently, there is no safety concern.
Figure 3, attached, provides a plot of F;H as a function of Rated Thermal Pcwer.
i
Mr.
S. A. Varga April 21, 1982
.NDlMSL:1685 Page 3 Worst case Fxy (Z) values were collected for each axial mesh from the three-dimensional misalignment calculations.
Figures 4 and 5 show the maximum calculated Fxy(Z) values for power levels of 100% and 45%, respectively.
These Fxy's lie below the proposed Fxy limits and include an.8% uncertainty. factor.
For very low power levels and deep rod insertion, the Fxy limits could be exceeded.
At low power levels, however, the Fg limit is very large, 4.64 below 50% power, and power peaking is not a concern.
The F limits xy proposed are as follows:
1.
For unrodded core planes:
RTP Fxy < l.68 up to 2.4 ft. elevation RTP Fxy < l.68 from 2.4 ft. to 7.8 ft. elevation RTP Fxy < l.65 above 7.8 ft. elevation 2.
RTP Fxy < l.71 for all core planes containing D Bank The power dependence of the F limits will become:
xy Limit RTP F
(P)
=F
[1 + 0.3 (1-P)]
xy xy where "P"
is the fraction of Rated Thermal Power.
With respect to rod misalignment Fg(Z), these values were conservatively generated for steady state conditions by synthesizing the limiting Fxy(Z) with the largest axial relative power for each axial plane.
Tne Fg ( Z) values for the reference case, i.e.
the HFP case with Bank D inserted to the full power insertion limit, are shown on Figure 6.
Also shown on Figure 6 are the limiting F9(Z) values associated with the limiting rod misalignment Fxy ( Z ) values discussed above.
HFP equilibrium xenon conditions were assumed for both the reference case and the raisalignment case.
The values plotted on Figure 6 include the 1.05 and 1 03 Technical Specification uncertainty factor.
Note that the misalignment penalty cccurs at the top of the core (in the redded region) and that considerable margin to the LCCA Fg envelope exists at all core elevations.
o Mr. S. A. Varga April 21, 1982 NDlMSL:1685 Page 4 For load follow operation, the limiting points from the Cycle 3 FAC~ analysis are also shown in Figure 6.
These points were generated using the Technical Specification Fxy values which are conservative by 5 to 10% with respect to as calculated values (including uncertainties) and by 16 to 21% with respect to best estimate Fxy's.
All of the limiting Fg x Power points occurred during full power time steps.
Furthermore, in the upper part of the core where the misalignment penalty is the largest, the limiting points occurred for D Bank insertions of < 24 steps.
Typically, D Bank is inserted only 12 steps for these points.
Consequently, upward rod misalignments would fall within range of the standard misalignment (24 step) Technical Specification which Beaver Valley used prior to the Technical Specification change in Cycle 2.
Therefore, for upward misalignments, no additional FQ x Power increase will result from the proposed Technical Specification.
Downward rod misalignments result in only small (approximately 3%) FQ increases in the vicinity of the misalignment.
There is sufficient conservatism in the analysis to accomodate these small increases.
Duquesne Light Company pursued expanded rod deviation limits because of the extensive testing which we have performed on the analog Rod Position Indication system at Beaver Valley before and during Cycle 2 operation.
We were convinced that a Technical Specification which limited the allowable inaccuracy of the Rod Position Indication System to + 12 steps would not be suitable for Beaver Valley Power Station.
We are not convinced that this problem is generic to Westinghouse plants since there are many plant specific aspects to the problem including variations in equipment characteristics within manufacturer's tolerances and spatial variations in the installed RCCA cooling system.
We believe that the temperature related phenomena which influence j
the accuracy of the Rod Position Indication system do not indicate a state of failure or malfunction but are manifesting characteristics of the instrumentation which we have measured and evaluated.
The characteristics of the instruments are such that they display a degree of nonlinearity in the steady state response and a transient (time dependent) drift which is due to thermal effects in the detector assembly.
With respect to the steady state nonlinearity, it appears that all control rods can.be calibrated to remain within the + 12 steo band of the group demand position in the range of from 20 steps 'to 223 steps.
Belov. 20 steps, it appears that additional margin is required to allow the caltbration to be performed in a practical manner.
There is some variation in the fegree of nonlinearity from red to rod but all rods at our facility appear to be enveloped as described above.
=
Mr.
S. A. Varga April 21, 1982 NDlMSL:1685 Page 5 With respect to the time-dependent transient response of the Rod Position Indication System, we believe that this response is due to changes in the magnetic circuit within the detector itself over time which causes the detector output to rise with increasing temperature.
The component having the most impact on the magnitude and duration of this transient response is postulated to be the lead screw and extension as it is influenced by conduction from the primary coolant at the upper head temperature.
This transient is most pronounced when a control rod is withdrawn and rod withdrawal appears to cause the channel to read high until the mechanism cools down.
A one hour waiting time appears to be sufficient to allow stabilization to approximately the calibration curve value of output.
Another thermally induced source of inaccuracy not addressed by the proposed generic Westinghouse Technical Specification for Rod Position Indication Systems is the fact that there is an upward shift in the calibration curve as a function of increasing reactor power.
This shift is believed to be caused by conduction of heat from the reactor vessel as upper head temperature increases with higher power levels.
This thermal response is a steady state effect not observable at zero power where rod calibrations are performed.
The impact of this phenomena is to cause the detector response curves to drift high and outside the i 12 step limits imposed by the Specifications.
This effect is not significant with the rods near the fully withdrawn position since ample margin exists in the calibration curve at that point.
However, when rods are near the insertion limit, this thermal effect will place us into the ACTION statement unless the Specification recognizes this effect.
To aid in the understanding of this complex phenomena, I have attached two figures which represent the calibration of a rod which is characteristic of a typical rod exhibiting these characteristics.
Figure 1 shows the entire range of the calibration curve along with a representation of the reference position and the i 12 step limits.
Also shown is the hypothetical curve for the rod at 100 percent power, illustrating the drift in calibration which occurs.
This curve is hypothetical in that we only have data within the insertion limics ahich represent this phencmena.
The portion of the curve shown belcw the insertion limit is an extrapolation of existing data and is probably irrelevent to the problem being discussed here.
Figure 2 is the same informatitn as is depicted in Figure 1 but is expanded to show the characteristics of the Rod Position Indication System in the area of interest (above 150 steps).
Our concern is related to the shaded area above the - 12 step limit which we expect to encounter at higher pcwer levels when rods are partially inserted.
Mr.
S. A. Varga April 21, 1982 ND1MSL:1685 Page 6 We have determined by testing that thermal effects are less pronounced on tne Rod Position Indication System primary voltage than they are when measurements are taken from the Rod Position analog indicators.
This could be due to the lessor impact which thermal ef fects in the magnetic circuit have on reluctance associated with the primary coil as opposed to the impact which these thermal effects have on the magnetic circuit when the Rod Position Indicator coils are used as transformers.
It is for this reason that we have elected to include the use of Rod Position Indication primary voltage measurements.
We believe that we will need to rely on backup indication based upon calibration curves of Rod Position Indication primary voltage.
versus group demand.
These calibration curves should be referenced to a constant power supply voltage and should include the 100 percent power voltage measurements.
Obviously, data cannot be developed for power operation until after startup and, obviously, we cannot violate the Technical Specifications to record the data.
It is because of the above stated technical complexities which we believe that a restriction to i 12 steps is not warranted.
Furthermore, from a human factors standpoint, we would be making the reactor operator's job much more complex by adopting a Technical Specification which must rely upon test instruments and test methods to carry out a normal operational function.
We believe that the Westinghouse proposed Technical Specification is not adequate, unless modified, and that the modifications necessary to make the specification acceptable will cause the specification to be more complex that need be to accomplish a relatively simple task.
- Lastly, it is extremely difficult to work within Technical Specification limits to acquire the data to develop the 100 percent power calibration curve for the Rod Position Indication System.
We believe that a i 12 step curve will result in restricted operation and could be the cause of Licensee Event Reports which we endeavor to prevent, as a matter of Company policy.
It was for these reasons that the Company contracted with Westinghouse to perform the analysis to allow relaxation of the Technical Specification to 1 16 steps.
1 In accordance with your request, however, we have modified the proposed Westinghouse generic Technical Specification for Rod Position Indication Systems to make it applicable to our facility.
The areas of modifica icn are as follows:
Mr.
S.
A. Varga April 21, 1982 NDlMSL:1685 Page 7 1)
We have expanded the limits of the Rod Position Indication System to i 16 steps from 0 to 30 steps to permit shifting the calibration curves on several group C and D rods to permit remaining within the i 12 step limit from 150 to 228 steps.
We have no data to show whether this fix is enough to permit calibration of the remaining banks of rods but we are led to believe that it will be sufficient.
2)
We have deleted reference to part-length rods since they are not installed at Beaver Valley Power Station.
3)
We have retained language in 3.1.3.2 which recognizes the use of Rod Position Indication detector voltages.
4)
We have modified the BASES to reflect the use of Rod Position Indicator detector voltages.
5)
We have deleted the APPLICABILITY of Technical Specification 3.1.3.3 for Modes 4 and 5 to avoid a situation where the Reactor Coolant System could be depressurized and non-condensible gases accumulate in the control rod housings, preventing lubrication of the RCCAs when the SURVEILLANCE REQUIRED in Section 4.1.3.3 is performed.
We fully appreciate the Staff's position that it does not desire to approve a change to i 16_ steps, both from the standpoint of the strain which this larger allowed rod deviation places on traditional analytical techniques for rod misalignment and upon the increased burden placed upon the Staff by reviewing plant specific, cycle dependant analytical results.
Perhaps this brief explanation of our operational experience will provide insight as to why we submitted proposed Technical Specifications which called for the e i
deviation.
- 16 step r
l
, Wg are certain that the proposed Westinghouse generic speci:1 cation coes not fully address the thermal phenomena whicb we nave cbserved at our plant.
We believe that the changes we have made j
to t,at proposed specification go a long way toward addressing the n
l Inacecuacies which we perceive - at the expense of human factors consicerations.
These changes do not address the acquisition of the l
, Rcd Pcsitlon Indicator calibration curve at 100 percent ecwer.
We Delieve tnat this data acquisition effort can only be oe'forred by a
' r special exception to the proposed Technical Specifica: ion or by l
Permi: ping us to use Cycle 2 data (if Cycle 2 data correlates
~
su :1cgently to new Rod Position Indication data to be accuired durinc Cycle g startup) to satisfy the accuracy requirements of $he Technicai Speci:,1 cations wn.11e new data is being acquired.
1 I
Mr.
S. A.
Varga April 21, 1982 NDlMSL:1685 Page 8 I will appreciate your prompt review and comments on this information and I am prepared to discuss this matter with you or others of the NRC staff at your convenience.
We will promptly make another submittal on the docket which addresses the concerns of the Staff related to the remaining Technical Specification changes submitted by our February 23, 1982 letter.
Very truly yours, h
.b J.
D.
Sieber, Manager Nuclear Safety & Licensing Department Enclosures cc:
Mr.
D.
L.
Wigginton BVPS Unit il Licensing Project Manager Mr.
D.
A.
Beckman BVPS Unit #1 Resident Inspector Ms. M.
Chatterton USNRC Document Management Branch
FtGURE
.i
?N*
DUQ,0e. CMC LEHT CCMPAMT
+
EkiMtcR, VALLffY pcuaese tTWi ou
/
uu sy i WPtCAL CAU3RATioM CURVE j
R S P P o G t T i o u i>J O tc e g i cpa S y s T E: M j
' Loo-v)
O.it! 120 -
/
loo */o Power R.
c AL.i tartAricw cqnvc b}
v
/
l'
~
/
z.eno powen.
c.nuia e.ATico cunuc l.,
t oo-s,
/
1
/
/
k 12c,-
7
()
/
u t c.o -
7 Q
1
/
/
Ro -
4-
-l
/
/
f Lo --
/
sq D
//
/
+ n. cre e sli,4r
.t.o -
/
7 e F u g.e uc e vos enou
/
//
- /
- 12. r-Tc P Li s rt-to. f o - /n 3DS 4/2 [22.
i i
t i
i i
i i
i i
i i
6 2.o Ao Go ao tco 12o 14o 16o 19o 2.co 2.2o 2Ao C> ROU P DEMAuD (s re Ps)
2 6
(
t.
2
/4 sD J
/
/
"5
/
' /
2 i
2 p/
)
SP 4/
/
E
/y/
T 0s l
i1
(
2 4/
g
./
n 8
6
/
g o
/
1 r
M
- /
P, T
T E
/
Et c
E t
s i V
,9 -
k b
YU R AY i
1 M
U s
+
F
//
D A
- C Y go M,p['o P O S
o M C J
/
M t
I S7 O T0M o
CA c,
g 4
C A O
1 iS g
M T TI
&,W*
n l
TS e/
s
=
c T
2 HR R A t
0 7
4
,/
O 2
EG B c
(
t g_n7_
fr
/
P J
I RL
>v i
ll D 5u U.N A h v_e'<_, /+
I T
i lo
/
o E
G Y C l
l N
/
R E
S L
u I
i FE I
A o f
f G
g!y t
t U A C T
LC V I
o7</[
t R. P s o
O iS D
4 Y o
\\
-ETP v
l\\A A_
E D
B O
R s_
c o
s 0
u io n
3 e
5 z
1 1
n n$ eI 3 c $L 4qLY e9 a 2
- l I
4 4-a e
~2 4
g e
o t
3 ed
,H
-V1 xx 6
t;4 x
d L1-d
.I g
6 o
@0
'J x W
-9 1 e
32 O
wf9 a0 J
a 1
d
[
C {J D
2 v
,c I
i 3
U d
d
-i-s 4
0 d
k>$
4 z
0-a 1-U Y
hi y a
d a
ja a
y b
b o
P o
4 A
9 M
3 e
a w
m 0
--To 2
2 Po a
4 ti L1-4 0
i d
O G
r0 i
6 i
i i
i k
C e
J 4
~
a e
N9d l
FIGURE 4 eer vue vnutax pows. mAMca Fw(O vs. Cone IM ht 3
H oi-Iull Po w e.n.
l.9 Ruje su4o Lec4ico LaVi4s F@e_eo tweJ3Y.
e l
I.L e
>-x 11.
e o
6 O
O O
~ t- \\.C 2
u, I
2 d
9 [.4
.i g
- f e Rob MisAtianueuT ixy
- RaJd d Fxy Lb'it 1.a L[,.C un aom ea Fxy List l.L 1
I i
i l
i i
I o
i 2.
3
+
5 (o
7 8
9 to ii 12.
CORE H E lG HT (Ft.)
ms e,/n.
FtGU RE 5'
B t:Avt R.
v ea-tcg. icv > e R.
STATtou Rop Mt SALIG OMt:vT Ry VS.
2,1 C.O RE 14 Etc HT 45 /o Remleg. Tnemai_ b.a.,
ECSC itiko hJCt. rel-iou L ip g-f s
> 2.o-A Lt.
}_
1,c -'-
t 2
L2a o
/
(,9-A e
e
.T's 6
4
"' t,-l -
I 2
O o
Fxy
$] [,(, -
G rod M ts AuCN MEUT
- - - unnoppen per
_iu i r
- i l,g.
Roo pe.r2 F y Li,J. I 4 r- ---- - l 1
1 I
t i
i O
I 2.
3 4
5 4
'l 2
9 lo il CORE H E I G blT CFt.)
aos 4hi/ez.
FlGU RE G.
sue-m vAu.mx powe g mno,a MAuMUM(FT x het) a vetsus E'4 ~
CORE H EIG PtT D O R I N Gs N O G. tM R OPE RATIN cycts a 2.2. -
4 O
Z.o -
x 9
X O
X l.S -
x e
q a% l.b -
x s
LL U
O
< l.4 x
~
m O
l l
- 1. 2. _.*
x I
l l
- 1. 0 -
o f
1 i
cc 2.
4 6
3 o
i :_
CcRE H E GHT (=0 nq c.,!,; j'e
DEFINITIONS REFERENCE POSITION 1.36 Analog Rod Position Indication System REFERENCE POSITION is defined as:
1.
For all Shutdown Banks and Control Banks A and B, the group demand ji a.
counter indicated position between 0 and 30 steps withdrawn inclusive and between 200 and 228 steps withdrawn inclusive.
iil' b.
For Centrol Banks C and 0, the group demand counter indicated position between 0 and 30 steps withdrawn inclusive and between 150 and 228 steps withdrawn inclusive. For the withdrawal range of 31 to 149 steps inclusive, the REFERENCE POSITION shall be the individual rod calibration curve noting indicated analog rod position vs indicated group demand counter position.
!!'i I
l
'l I
i
'l t
7 4
SE.2iER '/ ALLEY - UNIT i 1-3