ML19262A353
| ML19262A353 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 04/08/1976 |
| From: | Arnold R METROPOLITAN EDISON CO. |
| To: | Reid R Office of Nuclear Reactor Regulation |
| References | |
| GQL-0506, GQL-506, NUDOCS 7910260674 | |
| Download: ML19262A353 (26) | |
Text
I NRC ac'u 135 U.S. NUCLE AR REGUL ATOP V M
W OOCKET NUMDER s
,'t'- 2qG i i2 76
. JJRCDISTRIBUTION FOn PA7T 50 DOCKET MATERI AL FHOM:
z.,.
r _... -,
DATE OF COCUMENT lTO:
l DATE RECE;VCD BLETTER O NOTOR iz E D PROP
- NPUT FORM NUMBER OF COPIES RECElvED D ORIGIN AL S U NC L AS$1FIE D WCOpv I DESCRIPTION E N CLOSU R E
.-7 -:-
L-...:.,-.
'. t- !.... ;;. r '. ;,....
. ~).. -..........
r...
)
i 1485 167 PLANT NAME:
7
- ,c SAFETY FOR ACTION /INFORMATION EN!IRO
- ?-
A$SIGD AD :
/
BIONCH CHIEF :
ra i d _
BRANCll CHIEF.
V PROJECT MANAGER:
br.d e cs PROJECT MANAGER :
!./
fit. A$3T. :
b g,,,
LIC. ASST. :
1
~ l I
INTERNAL DISTRIBUTION
/ w UCC FILE
)
SY_SIEMS_ SAFETY PT. ANT 9YSTrug rmrTun 'pr is
/
NRC PDR FFIN'N A!!
TFDFqrn ERNST
/
I&E ( :)
ScimOEDFq BENAROYA BALLAR_D
/
071D ~
LAINAS SPANGLER
/
COSSICK & STAFF ENGIEERING IPPOLITO SITE TECll l
flPC u;CCAny CASE KNIGHT OPERATING REACTORS GAMMILL liANAUER SIIF4EIL STELLO STEPP
~
R\\RLESS PAWLICRI HUU1AN OPERATING TECH PROJECT MANAGEMENT REACTOR SAFETY
/
EISENilUT SITE ANALYSIS BOYD ROSS
/
SHA0 VOLU!ER P. COLLINS NOVAK
/
BAER BUNCH 110USTON ROSZTOCZY
/
SCHWENCER J. COLLINS PETERSON CllECK
/
GRIMES KREGER M5LTZ llELTEMES AT & I SITE SAFETY & ENVIRC SKOVHOLT SALTZMAN
! ANALY_ SIS RUTBERG I DENTON & MULLER l
EXTERNAL DISTRIBUTION CONTROL NL
/
LPDR : h': -c o., -
t_
NATL LAB BR00KilAVEN NATL LAB
[N TIC REG. V-IE ULRIKSON(ORNL)
NSlc T.A PDR n
ASLB CONSULTANTS
~).O ACRSIb IlOLDIN0/SEMP
" C' LMI: M M
t NRCFORM 195(2 70)
,7h192DNgy
N. 7M2 l w:..
~... - -,,,
METHOPOLil'AN EDISON COMPANY POST OFFICE BOX 542 READING, PENNSYLVANIA 19603 TELEPHONE 215 - 929-3601 April 6, 1976 m
f ' Q,.
GQL 05C6 ps APR 1 *, $7D A ?
Director of :!uclear Reactor Regulation a
y.s. w * ' Mercer ced,I A".5:I:
Mr. R. W. Reid, Chief
- ' d g')g ~gi\\ j- - '
Operating Reactors 3 ranch A U.S. Iiuclear Regulatory Ccnsission Washington, D.C.
20555 N/j e
Dear Sir:
TREEE MILE ISLA'iD !iUCLEAR STATIO: U:!IT 1 (T'II-1)
CPERATI:IG LICE:ISE :10. DFR-50 ECCYET :;0. 50-259 In response to your letter of March 29, 1976 requesting additional information pertaining to our Cycle 2 Reload Report, pleace firi enclosed three signed criginals (37 conformed copie sent separately) of our reepCnseS.
Sincerely, I
i R. C.'Arncli Vice President RCA:CWS:ilm Enclosure 1485 168 assi
RESPONSES TO liRC QUESTIONS CYCLE 2 RELCAD REPCRT URC QUESTION (1)
. Provide the following:
A description of the flov measurement technique used along with the a.
data measured and an error analysis for the measurements.
h.
A discussion of the bases for the pcVer to f1cv into reactor trip setting and the everpcVer trip setting (if it is based on measured flow).
A proposed surveillance and/or test program to confirm that the value c.
of the core flow rate has not decreased below the value used as the basis for reactor pover/ flow trip (and/or the overpower trip), including appropriate uncertainties.
RESPONSE TO QUESTION (1)
Attachment (1) provides the response to parts a and c above and attachment 2 provides the response to part b.
NRC QUESTION (2)
Rod boving is not considered in the report. A discussion should be included on the effect of rod bowing and either justification given for not including the effects or revised technical specifications to account for rod bcVing should be proposed.
RESPONSE TO QUESTION (2)
Proposed Technical Specifications to account for the effects of possible fuel rod bowing have been forwarded by our letter of April 2,1976.
1485 169
. :iRC AUESTION (3)
Provide justification for the =cderator and doppler coefficients used in the accident and transient analysis (Table 7.1-1 of Reload Report) and explain how thece values relate to those listed in Table 5.1-1.
RESPCNSE TO QUESTICN (3)
The ecderator and doppler coefficients given in Table 7.1-1 are taken frcs the FSAR. When the FSAR accident analyses were performed these values were used to provide conservative worst case values for these coefficients. On line measurements (% ECC) of these coefficients are reported in the TMI-l Initial Startup Report ( Sl'R ). The reported values are: =cderator
-0.222 x 10-h Ak/k/
F and doppler -1.t7 x 10-5 ok/k/ F.
(See Initial Startup Report Table 5.6-2)
AJ so note that doppler coefficient is extracted frcs power doppler coefficient measurements by calculation (See Startup Report Section 5.6).
ECC on line measurements for these coefficients have not been evaluated however, initial indications are that they are very close to the predicted Cycle 1 values given in Table 5 1-1 of the Cycle 2 Reload Repcrt.
The conclusion to be drawn from a comparisen of the =cderator and doppler coefficients given in Table 7.1-1 and Table 5.1-1 is that the values used for the purposes of the Accident Analyses are indeed ccnservative.
NRC QUESTION (h)
Indicate which of the values listed in Table 5.1-1 were observed in actual Cycle 1 operation and cc= pare these observed values with calculations.
Parameters of specific interest include critical boren concentrations, control rod worths, and core temperature coefficients.
RECPCNSE TO QUESTION (h) -
All values given in Table 5 1-1 are taken from the TMI-l Cycle 1 Physics Test Manual (PTM) except as noted. Attached is a copy of Table 5.1-1 with item numbers assigned to facilitate discussion. The following table lists the table or figure number from which each ites was taken.
1485 170
Tablo S h TMT-1, Cyclo 2 phyoicn Pr a torn Cycle 2 Cycle 1 296 466
(
Cycle icngth, EFPD 9144 14,396 Cycle burnup, tGd/mtU Average core burnup - EOC, igd /mtU 18,612 14,396 82.1 82.1 Initini core loading, stU
@ Critical borca - BOC, ppm (no xe,)
1350 1634 HZP - all rods out 1187 1494 HZP - Eroups 7 and 8 inserted 1004 1382 HFP - groups 7 and B inserted
@ Critical boren - EOC, ppa (Eq. Xe) 390 480
- HZP - all rods out 46 180 HFP - group 8 (37.5% withdrawn, equil. Xe)
@ Control rod worths - HFP, BOC, %4k/k 1.17 1.58 Group 6
.97 0.99 Croup 7
.54 0.44 Croup 8 (37.5% ud)
/ M Control rod worths - HTP, EOC, %Ak/k 1.32 1.37
"~
Croup 7
.50 0.26 Croup 8 (37.5% ud)
(
Ejected rod worth - HZP, %ak/h
.57+
0.48M BOC
.54+
0.72M ECC
' b Stuck rod uorth - HZP, %Sk/k 2.15 4.27 BOC 2.21 2.69 EOC
@ Power deficit, HZP to HFP, %Ak/h n
-1.64
-1.32 BOC
-2.48..
-2.10 EOC Q) Doppler coeff - BOC, 10~5 (6'4/k/ F)
-1.49
-1.51 100% power (0 Xe) h Doppler caeff - EOC, 10-5 (6k/k/ F)
-1.53
-1.67 100% power (equil Xc)
,jh Moderator coeff - HFP, 10' (ak/k/ F)
BOC (0 Xe, 1000 ppm. groups 7 and 8
-1.06
-0.23 inscreed)
EOC (equil Xc, 17 ppo, group 8 inserted)
-2.63
-2.70
(
1485 171 5-3
.o L j-5.1-1 (Continued)
/
Cycle 2 Cycle 1
~
(h'.Doron vorth - HFP, pps/%Sk/k 98 b
EOC (1000 ppm) 108 95 100 EOC (17 ppc) hXenonworth-HFP,%6k/k 2.61 2.71 BOC (4 days) 2.67
,2.65 EOC (equilibrium)
A (J3. ffective delayed neutron fraction (HFP)
)E
.00577
.00690 BOC v
.00516
.00514 EOC
+ Ejected rod value for group 5, 6, 7 and 8 inserted 4+ Ejected rod value for Group 6, 7, and 8 inserted
(_
.e
(
1485 172 5-4 e
e
TABLE I Iten "o.
Reference For Value (1)
PTM Table 2-h (2)
PTM Figure 2-61 & 2 h9 (3)
PTM Table 2-3 (h)
PTM Table 2-3 (5)
Recent 3&W Calculations (6)
PTM Figure 2-39 (full length APER)*
(7)
PTM Figure 2-55*
(8)
Recent 3&W Calculations (TM-159)
(9)
Recent B&W Calculations (10)
PTM Figure 2-52 & 2-57**
(11)
PTM Figure 2 h7 & 2-h8 (12)
PTM (12/10/73)
(13)
PTM Page 2.1-3
- ECC Value From Recent E&W Calculations
- With Operating Control Rod Alignment The values given in Table 5.1-1 are calculated and represent best estimates of on line values under the stated conditions. The techniques used by E&W to calculate these values are considered to be state-of-the-art techniques, and are refined, where possible and appropriate, based en expericental and operational data.
The same or similar techniques were used to predict Cycle 2 values.
Table 2 ccmpares the calculated and measured (where available) physics parameters. It can be seen that where observed values are available they ccupare very favarably with calculated values and suppcrt the validity of the calculaticnal techniques used for Cycle 1 and Cycle 2 physics parameters.
References for the measured parameters are given in the notes fcr Table 2.
1485 173
IEC &UES. TION (5 )
It is our understanding that the check value proposed in your correspondence c1 April 19, 1975 to address the problem of long-term eccling fellowing a loss of coolant accident (LOCA) vill not be available for installation prior to start-up of Cycle 2.
If this is the case it will be necessary for you to provide specific information on how power would be restored to the necessary valves following a failure of the 10 Engineered Safeguard Valve h80 V centrol center prior to utilisation of that aspect of the cooling system.
RESPONSE TO QUESTION (5)
The 2-inch stop check valve which originally was to be installed in the Decay Heat Pressuriser Auxilliary Spray line, is not available due to delivery delays. Ecwever, a 1 inch stop check valve is available and will be installed prior to Cycle 2 operations. The subject valve vill be installed between the existing inside containment isolation valve, DH-V63, and penetration 320s.
This new valve vill function as the inside containment isolation valve and will permit manual valves DH-V62 and DH-V63 to be locked open curing power operations.
In this manner, a ficw rate of about 70 gp: could be established through the auxilliary spray line for boren control following a LOCA. This vould be accc=plished by opening motor operated valve RC-Vh and the manual outside containment isolation valve DH-V6h.
Please note that the primary flow path for boron control following a LOCA is through the Decay Heat Drop Line (i.e. through motor operated valves DH-V1, DH-V2 and DH-V3) to either the reactor building sump or to the Decay Heat pu=ps.
The auxilliary spray line flow path is, therefore, only a back-up flow path to be utilised should a single failure, such as failure of DH-V1 or DH-V2, preclude establishment of flow through the Decay Heat Drop LIne.
As noted in our April 19, 1975 letter, the vorst single failure which could occur is the failure of the 1C Engineered Safeguards Valve h80 V control center. In such case power to open valves DH-71, DH-V2 and DH-V3 in the primary flow path or RC-Vh in the alternate flow path would not be i==ediately available.
In this case the following procedure would be used to open RC-Vh to establish long-term flow for post LCCA boron control'.
1.
Open breaker for RC-Vh at the 1C ES Valve Centrol Center.
2.
Verify MU-V2A is in its closed position and open the breaker for MU-V2A at the 1B ES Valve Control Center.
3 At penetration 315E, lift the power and control cables frc= the MU-V2A motor controller.
h.
At penetration 317E, which is located atcut 10 feet frc= penetration 315E, lift the power and control cables frc= RC-Vh =ctor control center.
1485 174
5 Using jumpers, connect the MU-72A motor controller power and control cables renoved frcm penetration 315E to penetration 317E to tenetration 317E connections for EC-7h.
~
6.
Utilizing MU-72A motor controller open RC-Vh.
As noted in our April 19, 1975 letter, 30 days would be available to acccmplisi. the above emergency action and, therefore, no plant changes are considered necessary.
1485 175
TABLE c OLE 1 FHYSICS PARAIETE?S Calculated Measured Hote
.Cricical boron - 1:,JC, ppa (No xe,)
HZP - all rods out 1634 161.7 (1)
BZP - groups 7 and 8 inserted 1494 15 h*5 (2)
HFP - groups 7 and 8 inserted 1382 Critical boron - EOC, pps Gq. Xc)
HZP - hil rods out 480 HFP - group 8 (37.5% withdrawn, equil. Xc) 180 199 (3)
Control rod worths - EFF, EOC, %ak/k Croup 6 1.58 1 52 (L)
Croup 7 0.99 0 98 (h)
Croup 8 (37.5% ud) 0.44 0.h6 (h)
Control rod worths - HFP, EOC, %Ak/k Croup 7 1.37 Croup 8 (37.5% ud) 0.26 Ejected rod worth - HZP, Zak/h BOC 0.48 0.688 (5) (6)
EOC 0.72 Stuck rod worth - HZ?, %Ak/k BOC 4.27 EOC 2 6
Power deficit, EZP to HFP, %ak/h EOC
-1.32
- 0 90 (7)
.EOC
-2.10 Doppler coeff - BOC,10~5 (ak/k/ F) 100% power (O Xe)
-1. 51'
- 1.07 (8)
Dop ler coeff - EOC, 10-5 (ak/k/ F) 100% power (equil Xc)
-1.67
~
lbderator coef f - HFP,10 (ak/k/ F)
BOC (0 Xc,1000 pp::2.
groups 7 and 8 inserted)
-0.23
- 0.222 (9)
EOC (equil Xc, 17 ppo, group 8 inserted)
-2.70
. Baron vorth - HFP, ppa /Zak/k 98 Boc (1000 ppo) 95 EOC (17 PPC)
Xenon vorth - RFP, %ak/k 2.71 BOC (4 days)
EOC (equilibrium)
- 2.65 Effective delayed neutron fraction (HFP)
.00514 y
1485 176
TABLE 2 (continued)
NOTES:
(1) SLT Page h.h-1 (2) SUR h.1-2 with group 7 at 26.5", withdrs.m (3) Hot Full Power and Operating Control Rod Alignment Last day of operation approximately h66 EFFD.
(h) SLP Table h.7-3 Hot Zero Fover (5) SUR Table h.8-1 (6) 0.706 at apprcximately 250 EFFD following rod swap usingboron swap method (7) SUR Page 5.6-2 (8) SUR Table 5.6-2 with equil. xenon (9) SLP Table 5.6-2 1485 177
R. C. FLOW MEASURING TECHUIQUE & ASSOCIATED ERRCRS The RC flevrate at TMI-l is determined by two metheds: a direct method using Gentille ficv meters in the Loop A and B reactor outlet piping and an indirect methed using plant heat balance data and measured feedvater flevrates for Loop A and B.
Historically, the primary side flov=eters have been indicating approximately 2% lover flows than the actual flov. After TMI-1 achieved rated power, an attempt was made to eliminate the -2% offset to re-calibrate the primary side flovneter instrumentation but gradually over a years time, the primary side flov=eter indication degraded again 2%. Additional effort spent trying to arrest this degradation was successful, however, the policy has evolved that the heat balance ecmbined with the feedvater flev=eter measure-ment is the best method for determining actual reactor coolant flovrate. This precedure is not unique to the TMI-1 plant but is utilized at all operating B&W plants. For pcVer levels near or equal to rated power, the net core pcver calculated free secondary side measurements is the value used to periodically re-calibrate the nuclear instrumentation for pcVer level. Thus, to indirectly determine the actual total reactor coolant flevrate with plant heat balance data and measured feedvater flowrates at full power is both accurate and consistent with plant calibration procedures.
The specific measurements required to determine tctal reactor coolant flovrate are presented in the following table:
Table 1 B&W NSS Heat Balance Measurements 1.
Lecp A reactor Outlet Temperature 2.
Average of 2 Loop A Reactor Inlet Temperatures 3
Lcop A RC pressure k.
Lcop A feedvater temperature 5
Loop A feedvater pressure (at OTSG inlet) 6.
Locp A steam temperature (at OTSG discharge) 7 Lecp A stean pressure (at OTSG discharge) 8.
Loop A feedvater flovrate 9
Leop B Reactor Outlet Tempe>.ature 10.
Average of 2 Lcop B reactor Inlet Temperatures 11.
Loop B RC pressure 12.
Lec; 3 feedvater temperature 13 Locp B feedvater pressure 33g 7g 1h.
Lcep 3 steam temperature 14U
_S.
Loop B steam pressure
- 16. Lcep B feedvater flowrate
17 Input power to all fc ur RC pumps 18.
Letdown flovrate 19 Makeup flevrate 20.
Letdown temperature 21.
Makeup temperature Measurements 1, 2, and 3 yield a net Loop A enthalphy change and 9, lo, and 11 similarly yield a net Loop B enthalpy change. Measurements h, 5, 6, and 7 are used to determine a Loop A secondary side enthalpy increase and items 12,13, lh and 15 provide the Loop B secondary side enthalpy increase.
For a normal secondary side heat balance cale.ulation of net reactor power, the Loop A and 3 feedvater flovrates and the Loop A and B enthalpy increase values are combined with pump input power values and energy losses and gains via the letdown and makeup flows to calculate the power level of the reactor. To dett r:dne the total reactor coolant flow from the plant heat balance ceasurements, a heat balance equation r s derived for the steam generators as shown in Appendix A.
Table II below displays typical heat balance data recorded and printed out by the TMI-l plant computer on February 16, 1976..
TABLE II TMI-l PLAHT HEAT BALA' ICE DATA Parameter Run #1 Run #2 T hot-A-F 602.2 602.1 T cold, A-1-F 557.6 557.k T cold, A-2-F 556.2 555.9 Avg T cold-A-F 556.9 556.65 BC Pressure-A-psia 2170 2169
- AH primary-Loop A 60.87 61.0L T fdu-A-F h58.1 h58.3 P fdu-A-psia (assuned) 970 970 T stean-A-F _
593.1 593.0 P stean-A-Psia 917 916
- AH secondary-Loop A 813.31 813.11 T hot-B-F 601.3 601.2 T cold, B-1-F 556.8 556.7 T cold, B-2-F 556.9 556.8 Av;; T cold-B-F 556.85 556.75 RC Pressure-B-psia 2170 2169
- AH primary - Lcop B 59.62 59.60 1485 179
T fdv-E-F h58.1 h58.3 P fdv-B-pria(assumed) 970 970 T steam-3-F 592.8 592.7 P steam -B-psia 91T 416
- AH secondary - Loop B 812.88 1*
6 6
5 396 x 10 Loop A Feedvater Flev lbs/ Hour 5.389 x 106 6
5.259 x 10 6 Loop B Feedvater Flev lts/ Hour 5.233 x 10 6 Total RC Flovrate(calculated) lbs/ hour lh3 35 x 10 lh3.61 x 10 Ratio to Design RC Flovrate 109 5%
109.7%
To determine the influence of the teasurement errors on the calculated value of the total RC flow, Met-Ed performed an error analysis identical to that carried out by C. L. Howard et al for the USAEC (Contract # AT(Oh-3)-189). The theory, assumptions and a list of the carefully evaluated measurement errors are in-cluded in Appendix A.
After considering all the errors and their functional dependence, it was possible co determine the deviations in Loop A and B RC flow-rates and in the total RC flovrate. It was found that the maximum protable error on the total RC flovrate vill not exceed 1.h9% for 95% of the data As to reactor coolant flovrate surveillance program, Metropolitan Edison Company vill verify the total RC flovrate within three conths after refueling and after that periodically every six months (plus/minus 30 dr.ys) using the heat balance technique described above.
1485 180
A P P E II D I X A
Error In Core Flou JJeterminatien At TMI-1 An error analysis has been conducted to Utermine the effect of instrumentation and calculational uncertainties in the deter =ination of reacter core flow. The analysis has been =ade on a statistical basis using the =ethod of C. L. Ecvard's at all (USAEC Contract AT (04-3)-189).
G As =entioned earlier in this report the practice has evolved at T:H-1 that the actual reactor core flow rate is determined frc= heat balance based en feedvater flev =easure=ent. The pri=ary loop flows (A/B) are determined from heat talance over the steam generators (SG) frc= the following relation (ks-4)
L W
- we (0
(L-L)
(L-L)
Where W = primary loop flow (A/B) (1b/hr)
W = feedvater flev (lb/hr)
F hs= steam enthalpy (Btu /lb) f h = feedvater enthalpy (Btu /lb) g h = hot leg enthalpy (Btu /lb) h h,= cold leg enthalpy (Btu /lb)
L = heat loss from the syste= surface bounded by the temperature sensors (Btu /hr)
The core flov for this analysis is defined to be the sun of the loop flows.
The ficvs obtained frc= relatien (1) are terned actual flevs (expected values) and are based en actual instru=ent readings. Se uncertainity in 1 cop flows because of instrument errors was deter =ined based on the following assumptiens:
- 1) instru=ent errors follow approxi=ately a normal probability distributien.
- 2) the tails of the distributien are truncated because those instru=ents that are not within specifications are rejected, and our carefully established surveillance and test program is the assurance that specification limits vill not be reached.
- 3) the manufacturer guaranteed maximum error is censidered to be 2 6 confidence limit (Supplement to ASME Pcver Test Ccdes, Part 5).
1485 181 h) the instrument string errors were determined frcs the randen maximum cceponent errors, manufacturer guaranteed taximum drift included.
- 5) since our main interest is decreased reactor safety the confidence level is based en use of only one " tail" of the distributien curve.
The cora flow error vill be the statistical ecmbination of the loop flow errors.
The variance of W, being a function of the uncorrelated variables W ; h ; hr r
s
. etc. is calculated frcm the relationship 1
9W
/ 'c)k/
j S h/
( 3
- Ts
-3bF 6,
'5w, 6p l
-=
I I
SWF GW i
Sh/
i 9 h/
~ $'
.J.
- 6L (9,/1 63 3%
l S A, l
3L Following partial differentiations 6
gF
-I-
@w Fw
=
3
$ kgc
)
Ab l
A kac Rc
/
l AA 2
Ak 2
~ ~ (' ' 5 = c &(*]
(3)
T (M. # "
~
~ * ( A L )'
The standard deviations of the enthalpies are determired frc=
9k I
9 ft T
'T g) j*
Fcr the calculations the numerical values of RU:i-2 vere used (See TABLE II of the report) in conjunction with the instrument string errors shown in TABLE I:
6 - Ar (y
n,.
=
1485 182 i
sa Au _ n e u
TABLE I Instrument String Errors Used for Loop Flov Error Analycis Feedvater flov i 0.635 Feedvater temp.
1 1 9h F Steam Pressure 1 3.0 psi Steam te=p.
1 2.120F RC hot leg temp.
1 0 93h0F RC cold leg temp.
1 0.52 F RC pressure 1 30.0 psi Heat loss to ambient
+ 50%
The feedvater flow error shown in TABLE I was calculated as follows:
The feedvater flow in both lcops is measured by flow elements supplied by BAILEY ETER CO.
The flow is ?etermined f*om t: c relation k
=KC kw (3.li60 a lo fW W
C
=
p Where c; f and b are; the mean discharge coefficient, feedvater y
density in flov element (1b/ft3) and differential pressure (in.
water at 68 F) ' < spectively. The mean discharge coefficients for both flow elements were determined by ALDEU RESEARCE LABOPATORIES, Holden, Mass..
The error in feedvater ficv measurement was determinM by the method discussed above.
The variance of Wf for either loen is obtained from the expressien kkh@ ' (f i
M b0
~
5
=K n-
} kh Isb were the 6 -s are ctandard deviations defined by their subscripts.
The errors used for the calculations are sho n in TABLE II
4-TABLE II Errors used 1or Feedvater Flov Error Analysis Nozzle discharge coefficient 1 0 5%
Feedvater te=p.
1 1 9hoy Nozzle pressure differential 1 5.45 in water The feedvater flow errors (2 F ) obtained for loop A and B vere
.612% and.628% respectively. A conservative value of.63% vas entered in TABLE I.
Based on the method, assumptions and numerical values discussed above the maximum probable error obtained for the primary loop A RC flow (1.6h5 6 ) was 1.512
- 106 lbs/hr (2.1%). Since the system conditions in each loop are nearly the same, the errors are nearly identical. Therefore the maximum probable error in core flow due to the independent rando: loop errors vill not exceed 1.49% for 95% of the data.
6 The 2 6 loop flev error is 1.8384
- 10 lbs/hr (2.56%) leading to a 2 6 core flov error of 2.60
- 106 lbs/hr (1.82%).
The results of this analysis indicate that the 106.5% design flav for Cycle 2 is acceptable relative to the 108% generic nominal" flow (See Cycle 2 Reload Report p. 6.1) based on the maximum probable error, and is highly conservative relative to the actual RC flow at T:C-1 of 109 3%. This value is based on numerous heat balance calculations carried out in 1975 and 1976, and is considered a time averaged stable flovrate without drift.
EFH:11=
1485 184
Attachment (2)
Bases For Power / Flow Trin I.
EACKGRCUND This enclosure provides a discussion of the bases for the power to flow into the reactor trip setting. The bases for the overpower trip setting is not included since it is not based on measured flow.
It is the purpose of this enclosure to de=cnstrate that the value of the flux / flow trip ratio is still conservative and adequate in the light of certain parameters different frc= those values assumed in the FSAR.
These parameters are:
A.
Core Flow Rate; measured flev rate (with 1 5% =argin) is used rather than design flow rate.
3.
Response Time Flux / Flow Tr,, /ield Change FC-137 increased the time constant of the reactor coolant flow sensing string to 1.h seconds frc= the 0.65 seconds assumed in the FSAE.
The effect of rod bowing has also been addressed. Consequences of rod bowing other than those affecting DUB are covered in our Technical Speci-fications Change Request 30 Amendment 2, suh=itted April 2, 1976.
II.
THERMAL-HYDRAULIC METdODS To determine the flux / flow trip setpoint that is necessary to meet the hot-channel DNB ratio criteria, several calculational steps are required.
These steps involve such things as the determination of steady-state operating ecnditions, fuel densification effects and transient calculations.
A.
Thermal-Hydraulie Conditions Durine Normal Oreration The hot channel thermal hydraulic conditicns are calculated for design conditions at 108% of rated power. The power level of 108%
includes operation at 102% of rated power plus a maxi =u= power level measurement error of 6% (h5 neutron flux error and 25 heat talance error).
This serves as the bench = ark calculation frc= which the densification penalty and the transient effects can be determined.
The steady state analysis is performed using the TEMF cc=puter code (3AW-10021) (1) with the appropriate bot channel factors, coolant inlet temperature and system pressure errors, and a 5% hot assembly flow =aldistribution factor applied.
These conservatis=s are cen-sistent with the calculational techniques e= ployed in the FSAR D
analyses. Tne design ficw rate of 131.32 x 10 #/hr. (88,000 GFM/
Pump) was used for first cycle analysis. For second cycle analysis,
the reanalysis used 106.5% of the design flew rate, based on system flow measurements made during the first cycle. For both cycles, the hot assembly pcwer distribution censisted of a 1 78 radial-local nuclear peaking factor (F 6 H) with a 1. 5 cosine axial flux shape.
Incorpor'.icn of the increased flew rate into the analysis was accc=panied by a correspending increase in the reactor coolant inlet temperature, frc= 55h? to 555.6F for the nc=inal, rated power e
condition (assumed to be 2,56cMWt for this analysis. ) Neither the increase in system flow nor the increase in inlet temperature repre-sents a change in the operation of the plant. The integrated control syste= =aintains a constant average reactor ecolant temperature, an increase in the steady-state system flow rate results in an increase in core inlet temperature and a corresponding decrease in reactor vessel outlet temperature.
As a result of pump and reactor coolant system tests, a majority of the orifice plugs were re=cved from peripheral fuel assemblies prior to startup of TMI-l and other similar 3&W 177 FA plants.
This was done to preclude operation with excessive coolant flow through the reactor core. The result was an increase in the maximum core bypass (or leakage) flow conservatively estimated to be 2.35 (frc= 6.Ch% to 6.3h% of total RCS flow). This increased leakage was not accounted for in those analyses based on design flow (Cycle 1) because it was a direct result of the higher system flew.
For those analyses based upon the increased system flow rate, the increased leakage was taken into account, thus for an increase of 6.5% in syste= flow, the corresponding core flow increase was h.2%.
Batch h fuel assemblies, which will be loaded primarily in peri-pheral locations for cycle 2 operations, have a sligntly lower resistance to flow than do the batch 2 and 3 asse=blies. Since the batch 3 fuel is located in the hottest core locations (hot assembly is in batch 3), the result is that the coolant flow through the hot assembly is slightly less than if all assemblies were identical.
This difference is conservatively accounted for in the thermal-hydraulic analysis by assuming that the cycle 2 core consists of two batches (116 asse=blies) of the less restrictive Mark 3h asse=blies and one batch (60 asse=blies) of the more restrictive Mark B3 assemblies, as discussed in section 6.1 of the TMI-l cycle 2 Reload Report. The resulting predicted minimum DNER is 2.05 (BAW-2) for the 108% overpower, maximum design condition (undensified, including temperature and pressure errors. )
3.
Densification Effects The fuel densification penalty applied to the hot channel for cycle 1 operation was determined by the methods discussed in the Oconee II Fuel Densification Report, BAW-1395, June 197', page A-5 a
conservative slumped and spiked 1.83 outlet peaked axial power shape was used in conjunction with a 1.h9 radial-local factor to determine the maximum fuel densif3 cation effect on DN3 ratio.
This reduced het channel DNS ratio 1, the basis for establishing the initial conditions for the tran sient calculations.
The power spike =cdel and densification penalty analyses for cycle 2 are discussed in subsections h.3 and 6.2, respectively, of the cycle 2 Reload Report. A conservative DN3R penalty for dersification was assured by the use of a penalty analysis based upon the most limiting (batch 3) fuel, without consideration of the effects cf burnup. Application of the densification penalty results in a reduction in the predicted minimum DNER for the 108% overpower, maximum design, case from 2.05 to 2.00.
)h0
C.
Effect of Oten Vent Valve Assu=ntion Fnr the flux-flow trip setpoint analysis, it has been conservatively assumed that cne core barrel vent valve is stuck open.
This assumption reduces the effective core flow rate by h.6% and results in a corresponding reduction in mini =wn DN3.
For second cycle analysis, the effect of this assumption is a reduction in predicted minimum DNER for the 108% overpower, maximum design case frc= 2.00 to 1.85 This valve represents the initial MD:GR for the transient analysis described belev.
The RADAR (3) cc=puter code is utilized to analyze two isolated channels, representing an average subchannel and the hot subchannel.
Input to the first channel includes nceinal suichannel and fuel red gec=etry.
Initial ecndition inputs to this channel include the nominal input power for a fuel rod, the reactor overpower (1.08%)
inlet entht py and outlet pressure representative of the 108%
oprating condition, including errors, and average channel f1cw rate.
Transient inputs include normalized reactor power and flow versus time. Heat input to the coolant is calculated by a transient fuel pin model which includes 15 radial nodes in the fuel, a calculated fuel-clad gap coefficient, and 2 radial nodes in the cladding. The axial heat input distribution within the fuel is represented by a sy==etrical 1 5 peak-to-average cosine axial flur shape distributed over 60 axial nodes. Primary output frc= this calculation is the pressure drop versus time for the average subchannel.
The second channel analyzed by RADAR represents the hottest sub-channel in the core. This channel, and its associated fuel rod, is modeled in the same manner as the first channel with appropriate hot channel factors added.
Input power to this channel is higher than that of channel 1 by the maximum design radial x local power factor of 1.783 plus an added factor to account for the densification penalty. The flow rate in this subchannel is calculated for both initial and transient conditions so that the hot channel pressure drop always matches that of the average channel. By analyzing the reactor core in this fashicn, the first channel represents the average core response during the transient while the second channel repre-sents the response of the hottest subchannel.
The result is a more severe hot channel transient than would be indicated if the transient core flow function were applied directly to the hot channel.
D.
Transient Hot Channel Conoitions During a Loss of Flow The flux / flow trip setpoint is derived to protect the core during a one pump coastdown. A ene pump coastdown is analyzed because re-dundant pump monitors are provided which will provide DUB protection for all other pump coastdowns including coastdowns while the plant is in partial pump operation. The pump monitor logic will not cause a reactor trip for the loss of one pu=p frc= four pump operation.
The thermal-hydraulic response of the hot channel is calculated by RADAR cceputer code (3AW-10C69) (3). The initial hot channel DN3 ratio is set equal to the steady state value with densification and open vent valve effects included.
The RADAR cutput in the form of Hot Channel DNB ratio versus time is the basis for establishing the flux /flev ratio trip setpcint.
1485 187
E'.
Rod Boving Effects Analysis was performed with the COBRA III-C code to determine the effect of a fuel rod boving into the hot channel and reducing the flow area of that channel. The results de=onstrate that rod bow of the magnitude predicted is adequately compensated for by tha flov area reduction factor. Rod bow away from the hot channel was also analyzed.
In this analysis, the effect of a power spike was added to the hot rod in the area of the minimum DNER.
This analysis also demonstrated that the current TMI-1, Cycle 2 DNER results conservatively account for the effects of fuel rod bowing.
III. FRCCEDURE FOR DETERMINING FLUX / FLOW SETFOINT The determination of the flux /flov setpoint is accomplished in four basic steps. The result of these steps is designed to yield a value of the flux / flow ratio that will prevent the minimum hot channel DN3R frc=
going below the limiting desigr DNER for the coastdown for which pro-tection is required. These steps are as follows:
A.
Total Time Determination From a plot of =inimum DNER versus time find the time that yields a DNBR of 1.3 for the maximum power level (1.08%) for the maxi =um nu=ber of pumps lost for which the flux / flow trip must provide protection (one pu=p for TMI-1, although the original Technical Specifications were based on a two pump coastdown since redundant pump monitors were installed subsequent to the original calculations).
B.
Coasting Time Determination The total time to reach a DNER of 1.3 minus a conservative value of the total trip delay time gives the maximum allowable coasting time prior to trip initiation.
C.
Minimum Flow Determination Frc= a plot of flow versus time for the coastdown of interest, the percent flow for the maxi =um allovable coasting time is found.
This yields the flow at which trip must be initiated.
D.
Flux / Flow Ratio Calculation The maximum allowable flux / flow ratio is the maximum real power level of interest (108%) minus the power level measurement error (6%) divided by the minimum flow.
IV.
CALCULATIONAL RESULTS Figure 1 shows the flow versus time that is the design basis for the determination of the flux / flow ratio.
1485 188 Figure 2 shows the calculated DNER versus time with the effects of densification included.
Frc= this figure it is seen that a DNER of 1.3 (W-3) is reached at about 3.35 seconds. Using figure 1 and the technique explained previously, this yields a flux / flow ratio of 1.08.
This is the value presented in the FSAR Technical Specifications for densified fuel. Figure 3 shows DNER versus time for a TMI-l one pump coastdown using the 3AW-2 (2) correlation.
Curves are shown for both the cycle 2 analysis and the most recent cycle 1 analysis. Differences in the initial minimum DUER occur because of the increased flow rate accounted for in cycle 2 (106.5% of design ficv) and because of core configuration and fuel assembly =cdeling differences discussed in the above paragraphs and in sectica 6.1 of the Reload Report.
Frc= figure 3, the limiting design DHER (1.32 for cycle 1, 1.30 for cycle 2) is reached at 5.h5 seconds for both cases. Using the Method defined in III. above, with a trip delay of 1.3 seconds, the maxi =u allowable flux /
flow ratio is then 1.12.
In recent flux / flow setpoint analysis, such as the Oconee Unit 1 third cycle analysis (h), the methed defined in III. above, has been refined slightly to include the effect of "DUER Turnaround".
This effect results frc= the fact that sc=e finite time is required after centrol rod motion starts before the minimum DNER is reached. Fcr the TMI-l setpoint analysis, this effect can be conservatively accounted for by adding 0.5 seconds to the " trip delay" time. Using a value of 1-9 seconds for trip delay, the maximum flux /flev setpoint would then be reduced frc= 1.12 to 1.11.
It should be emphasized that the above described analyses are based on the assumption that one vent valve is stuck open. This assumption reduces the effective core flow by h.6%.
Elimination of this ccnservative assump-tion would have the effect of increasing the calculated allowable flux /
flow setpoint by approximately 0.0h.
This conservatism of the Tech Spec value (1.08) is assured.
V.
CONCLUSIONS Cycle 2 analyses for TMI-l have been based upon reactor coolant flow measurements taken during the first cycle operation and have incorporated revisions to the standard S&W analysis techniques. The analysis described in this report has de=cnstrated that the technical specification value of the flux / flow trip setpoint (1.08) is conservative.
REFERENCES (1) 3&W Topical Report 3AW-10021, " TEMP-Thermal Enthalpy Mixing Program" April 1970.
(2) B&W Topical Report EAW-10000, " Correlation of Critical Heat Flux in a Bundle Cooled by Pressurized Water," March 1970.
(3) 3&W Topical Report BAW-10069, "PJCAR-Reactor Thermal and Hydraulic Analysis During Reactor Flow Coastdown," July 1973 (h)
TMI-1, Cycle 2 Relcad Repcrt i485 189
n _ :__ ). :4. __:::__
__. _._n4.. 2... l..
p__.. __:. _q._.1:..cj.; 'l.
q..._'L_.. l -
"Y I
~
'-[
_..[ -..
r' ?
~
..; l... __.
r
~
.f i_N6T.. C f%.Mtat L :.77 rg [ r+ tom. D WcR.;
...* S...
i
.. I E, b 63 4 _
E.~_C. h PcM M._]*_..g r-1g_.f,.f g., r.
1 l
i
- I-3.
.i.
i
.o
~
y..)_. _
e,._. _ a. w,....... u
_~
I 1
I
. :I d: ' CE. _O."b. L.Ed.Q. ~.L_NJI~wl. lid.PE.D -
4l:
. ", w J8
....i. 2 o.
. I..
.s........
..I.
.. l...
.i.....
... -.Jn.. _.
..l
.t..
.x
.v
, t.. e,e.
.t,.
._..t,_.2..
._ l
.... i. :....
.....t._
.t._...._..
.r......
_. 3..
....t.
..I...,......
....t............
- t......,..,,...
... t..
.a.._
.s..
1
- o..
l g
n< >..
. _.,....t..,
..g.
- 7..
-..... _... _....i._..... t...
.i
.r.
.,....6...
.t
.. 1....................... _.
.. i......
_v.. _.....__._...._t..._.._._._..
t. _..
._..r__.
__ _ t_
...I..
.t.
....... i
.n.,,
__..i._........_..p.._.._.._.l.
......s.s...........l.
.,.._.._...t.
........ l,.
-... \\....
..t..
,..i.
....... _ _ v._......,.
._...__.._...4__..._..,..._...._._._3...
........ 1...._... _._...._........_... _.........
..L.._._.___
_._...f. _ _... _. _ _.
.r.__.__._._.._.............._._.._......_,__l._...8....._._.___......_._.___.;.__t_.._.i._.... _.. ! __..
... i.....
_v_._ _.
.._.. l.. _..... _.. _... _.... _
r i
.g,___
...__.__..._.._._.__..t....r.._.6.____._._
..N.._..___.t._.....,......... _.,.... _. _..,.
.. _ _. __t....._._..
N. _. _. w.._._..... _......... _..
..,_)_.._. _ _. _.,...._......_....._.._...._t.._.....__.._t_......_.__._._.4_N...
...r..._.... _.
. r.._
...._._.'.-=_.,.M_
l L........'....,.
g _
.)........ 1
.__...__....__..__l.__..__T_....__...._....._.,___..._._._.._..
...:_.. _.. _... _ _.. _.. _.... _....__.t._.____.L.......... _. _... <. _ _ _ _.....
r..
.....t..
. g _.4..
_. t
_ _. __,.__.__._...__ t.__..._.._.._..._.
..... _....._..t........
... r.........
.._.r......
..i.__.._............._.._._.... _
.._.t--
.._._._._r._....,...._.....,.. _
- ~-
z,...
. _.q__
..g_...
.7...
._.....__4._.._._._
........,..... u......:.
. _r...
.... _.1.
. l... _.1_ _. _
.v
. _. _..._...._..__........t. _..
........t..
i.
I._.
V.....
. _ +. _ _ _ _
...i.._ _._..,... _,_ _.
.l...,....1......_
._.__._......l..._............_..s......
.t.......
_._t.._
.... _... r.. _
... t... r f.....
. _. ~.. _ _........
_.._.!..t.
.....t,..
.g....__.__._.2.__,...r._.._
.,...._.._.......,t
..._.1_..__J._.._
..i..
....... _._J....... _ _ _....... _..I
._.i_._.._.
.r u... __.,....
...t........
- t...
..i._.__...._.2._
..... r_.
._. 1. _
_.t...._
....... l
.... o... : _,
. _..._..r_.__.............,..........
....i.
../
m,.
. _. 4. N..v.
. e.
.....1....
.........._.l..
....t
....~....
....l.... t... l..l....;..I
. _._ r.......
.t.
.. _....)......
,....i
.l..l.
..t............
.t...._........
...... g
..l.
u......
.c
..l
.l.
.I..
.t..
i...
..i.. l.,3 l.
.l.
. i,.
. o. _ :_
p.
-..-4 g.
.t.
r.
.[
. T'.__p. $ (
.i..i, t.
._.. I
.. /..$d.Ep....
. _d. i. i.
,i[. ; 1r.'
- n:i: _
....f.. -
- i. -
1
}99
..:... u..
/ -Q l. j....
......i....i..
..---.....i...i..
.. :. w. :..
n
-....r.....
.,n ::
M yp M....y.u m,....-.. ".. a...
4
... s, - pa-.
- in
.i..
.... N.._.;.1...
.I..
p.
.,..5.,s....
j
- c.. s.t -..._.. _ _....
..........s....
i...
.... s_.
- i. n..
..i......
u.
t..
.. i.. w,.
,.....i..
.,...i...........
..=
.e..
_....3 _.
.. o..
,..._._.,i._
..i.
i...
..e.
..a..
r.__.
1,............
...i-.. _
.t
.i._..r_......
..... _.~._....
...i...
I..
..r
.._.i...
........t.......
......... t..
..... _......i.._.i...
....t..__._.__
. i..
sm
...i..
. t. __ p. _..... __.
.._....,...i_....
.._.i._.._..... _. _
.t._........._.......
s
. i...
..t.._..
.i..
.....:..........t
. i....
.a
_. s.
_....._..._............_..,......_.....z.,.
_..t_..............t._._.._._._......._....... t...
..........i_.................. :......
c
.i_......_.....___....:....
......... i....
....,e..._.i.._.,.._.
...._.i..
_.1._....._....._..
o...i......._._._.._...._.
...t.
......=
...l e
s
....i..i_..,>.........
..t
... r
.:.4..
,....g _..s___. _.. i ___a.
_ _.=o._.
. m.
... s, _.
.....t...
.._._.._.___.._._,.._L.__t.._... -.....
.......__1____.,____._.-_...._.a,....._..
...+..
... ~, _ _.
.................._.._r._...
... _,..i... _.,..._...._ _.. /. t _... _ i...
...i _..
._2...__. _. _. _...
.__...._.._i._._.
.e....._..
__._4...._.._.,_...;.,.__.-.r...,__..__-.._-.:..t.._.l............._.........t_.........-..-.....-.--.,..
p.n
...___..._..._.,...,_L._........_
_g..,._.__.
,,g.
..u.......,...n
. t.
,_..._._....._....__.i_._...._.1_..._...
.--..-...t....,_--. _ -....
t
._...O..
. _..__.. _ 1....... _. _. _ _ _... _.... _. _.._
...._.....s..
.... f..._-.......-... o -._........,.__......=..._.
.r--
n
... r....
. i..... _ r._.
...i..
...._..m__.....m.. _...
L........ l. _....r _......,. _.. _... _..... _... -...,. _ _.......
......_._i._._..
... _.. __ -t.. t __.
...._.i_..
i....... _.
..,_....._ -.. __. __.___.n._...i._.
...r......_..._..._r.._..__1._.__....
...._.._._..t..._..
_ t.._._._.L....
..i.._.__...._.._..._._...__.._..._._..._....s._.,._.__......._._._.._t._...
...i...
r._.._........
_..._. s...-. _......, _.._.......
._. s....
...__..,.,__.._._.t._._.__.._._.,_._.
_..t __...., _._.. _ _. _
_._.,f...__,_.t _....__.. _.. _ -
_. A..
_...__,,...__i___..,,___......,__.___._._.~...,..._...
._...._._.3_..
................... i...:.................
....._.i._.._..i_............
_.t. _....
._.._.. _.i....._........;.........
_.... _...__._....: _._..e. _..._.,. __.._
_......t__..
.........i._ _...
..t._.__._.._.__...... _...........
_.a......,..._..._..__.._._.__._._._.,.._.._._._._.__......,_....._...._..i,._....__..._....,....._:_..
._...,_. __.,.- _.4...
_e_.....3.__....._.c_...._._
... t,, _ rs... D.
3....
~
._...._.._.._:.._i........._....
.._.A......._...._.r._....__l__...:...._..__._.... _....... _..__._..._.. _.....
r..
t..._..._..._
.m.....-..._.,....
....._._...r.
t.._......._.. _ 1._..........
..(
.__.._.._2.._._._s.._.._.
...__i..... i..
_ i......
_. 4..
<s
.i t...
. _..... ;s..... i
........_.....1_.__._..._..._._.._.i...._..._._.,_.....
... r. _......
.r.
... a.
........_.c...r_..__..t.......,.......
- s..... d,.n..
...._._..m._._t.....................
,. _ _..... 1......
._.1._......
...-...._ _....i.. a.
. _... i_..-_..._...
r.
.,. t
._.t.,
....t
....1....
.t
....., A... J _
.~
.. _. t _..;..,
.._.,.._.,._.....i........
s
- s.
__. r.... _... I ____..
....i.._..
.... J.....
..s....,....._....:-........,_......._ r....... _
.._..._..i...
_...... -....s.._.._..t
...i..... _.
_.i.._.
.._. u..
- r..
Og....
..i.......
~.....
...i....
._._..__._..i...,._.,._
...r......._.i.
...i...
._-f,...,..
- a.....
..........1....
_.. r..._
..-_-:.....r......__...............-...._....-.. _..
. tw..
....t_.__.
. -r.._- -
... :...... t
...... m_, _....:. _.
i.
. t,.. _ _..
.N...._._..
...,.._.;._...._..r._.
.........._i...
. _.. l _.
- 4. _
-. =.
_ n,..
._.i..._.. _
.__.i_..... _
..t.
- f. -
...~._.r...-..............
........r.....
- -. s. _._. 1....._.
. s, _...
-...t......
.. _..i _.__.
5 -.
q,...s.......
.-..s
..n..-.
_.. t _...
._. g..
1
. -....,. p.4
...i...... _
...._1...
.. 1.
...i..
.. r..........i___.
__..~i..
..,s............_ _
_ m. n_._.
.3,...
. r..
- t..
...i.. _.._.....:._ r..... _
...s.
t.__
_. s,_. _......
,...t.
...,..---._.L.........
f..
...).
. - -... J.
. _. _u...
_....l
...,t__.
.....I..
....l-
.e
..I.
_. _....l....L.
,,.,:[......:,, :,,, ::,:g. n :..~... J.-l;.
.P 4. "T
- r.
- .E r
..\\
.~.g._.
I. _
g - P ::l:..
..=-
..t.....
..g~.g
.. ~E _
f).. :
..t.
f, t.
1 I
s.
8 4.. I
. l.
i l[....
...g....l..l
...J....
.-. s..
I t..
- 7......
_L-.
- ].
- l l
l g y yggggg,,2.
.I I**!
?
..,. L. N.
'*3 e
_ [
l' 8
8 1 N-i
".i
.4 i
. 3 l..
..I - %.-l 9.;. [.
.3 o.._
.l.
._ _w _ L _. % :<. - g.
.._. 5__ :
.f Ni j% j P.'
K j b I Ni
-l-l d l "
i i
I
.f
.l
.l. j
..l l.
.s I
8
..l l.l t
..{ __j s
..g
,a
_ a.
.l..
- -.i. _
g
.g l
.g
{
. j t
...j.
[ --*}-
a 1
}
-l.
..l..-
.Q i
...l...
I
'~u-
. o'.. =
... ;I,'...
.,1;
'l.
.l l'
Mi 4.;..
l
.. l.
.i.
i
.4 I '.
i
. g...
...g...
.1.,...*..
s g
..I
.=
l
..\\
.l..
.l..
7 7
. ' ". ~
....g..... _..j..
3
- 1..
s
....).....,...
i
..,g.3 3..
..I.
..I..
.=
, =. =....
.=y i
..r
.. u...=
i
..i.
e.
.s.
l....
..L
. l' i
...I.....
\\
i
....l i.
l..
..l 1..
.I..
.....e.
l.............
..., %.p.
. I
.. ~.......
l.
g l.
.4.
.l..
...l.
.1
.. 1..
).
.I 1...
, s3 _..
i.g.
_ g.
.I
.1.
- 3...
3.
.., I
...... l
- 4..
.l.-
o..
3
.3
.6 l.
.a
...i..
..% T 1-
. t I_.-, p...
.'h
..l..
l.
.,l. 8
.l..
f l....
3.
. 1.. l.
(..
.. l.4
[
. I _-
...g..
l
.,.g a
..I..
e.
e.
....i.
- . j.
1..
..l......
i
~.
.l-r-
g
...y 8 8
....g
.4....
,. a.
....,4.
s
.l a.-.I.N e. t
.....a I.
I
~~T l.
. *\\.
e
...f.
a..l
- 3.......l..
.. :l.
..... r
.. -. \\\\.
s p.
- t. _
..l.
l l.
(,.
4
..l.
.e:
)..
.s l t _t
... 3... i.. l._. L.
...,.e %,,.
1,.
.~
. _.... E......ll...I....s
.I.
....,s...
.l.
..l..
N 1
s...
.i
.i
.t\\.),.. N..
s
.'T.
.g..
3
.l..
..)
l....I.
.. y..
. 3 s.
..l.
.g..
1 _..
. a
..g,
...6.
J.. l..
I........ - -. l..l.
..... f e
a
.g.
g
_. m.
.l.
. l _...
.g..
.._.1...
.1.
4
.... s.
s..
g l.
e,.
.l...!.
. g.
- 4..
...f.
~1 e..
j e
.i g
e
.-. N.,
.i....
g.
.l e.-.,...
.....6
.l.
l-i I... =..
_ - -. g..
e
.s
.l l.....
w t
.t.
..... I. :
l
.. I... s.
t
.l 1...
I l-l i..,l
..l
..4...
I.
.l
..3..
.g
..s..
.g
.. I..
.l
..... e.
f.
G, l
..... l
..g e
. n.
- g. -
.._g..
=4 g
l
.3 6. -.
.l
.I
....l..e o
i 1..
I
-,l
.q -
l.-
._.4
-s 3
...l
.I..
a
,\\.
..-.4 l..i g
... )..
.g.
,..i e
...l
. _.. l..... t...
..l.
- e. _
8-.,..-
.._ L. _, - _._.
.. _ s..
..l l-
..e I G.
1 s
1 l..
.a
. l......
... I
._t
..J.. l..
- 5.. l...._
,..