ML19247A269
| ML19247A269 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 06/29/1979 |
| From: | Phillips L Office of Nuclear Reactor Regulation |
| To: | Denise R Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 7907300469 | |
| Download: ML19247A269 (15) | |
Text
r
/
- 8-v UNITED STATES c
.4 NUCLEAR REGULATORY COMMisstON
.s WASHINGTON. D. C. 20555 w r'
[W
'%..".y..f aus 2 3 1319
/
MEMORANDUM FOR:
Richard P. Denise, Acting Assistant D' M actor for Reactor Safety Division of Systems Safety FROM:
Laurence Phillips, Acting Sr cn Chier Analysis Branch Division of Systems Saf y
/
SUBJECT:
DOCUMENTATION TMi-BENCFMARK CALCULATIONS Since the first few days after the TMI-2 accident, several calculations were performed in the Reactor Analysis Section of AB. The purpose of this memo is to document these calculations and to point out conclusions and quastions which may be considered for further study by the task forces now assigned to the TMI-2 review.
(1) Core Uncovery Calculations Based on studies of the plant process data which were plotted for the period i: mediately after the accident, it was concludad that core uncovery first occurred at 105 minutes after turbine trip. This was indicated by primary coolant temperature data and steam generator pressure data which indicated a loss of primary to secondary heat transfer, nerefore inferring cessation of primary coolant circulation and condensation in the steam generator. Subcooling of primary coolant had ceased at approximately sixty minutes after the event. and all reactor coolant circulation pumps were turned off by 100 minutes.
Calculations were performed to estimate the necessary primary coolant mass discharge rate to achieve top of the core uncovery at 105 minutes.
The calculations were based on the RCS pressure history and the assumo-tions which follow:
(1) No HPI coolant after approximately eight minutes.
(2) Mass discharge rate varied as the square root of the pressure times density product, (3) All liquid mass above the core exit elevation (including pressurizer) was discnarged at 105 minutes.
57G0ett 4
7 90 7300 y67 5/Vrott-7785
Richerd P. Denise JUN 4 W3 Results of these calculations are indicated in Figure 1 and show that a mass discharge rate of 395,000 lbm/ hour (rated @2250 psi) would be required.
This compares to a rated steam discharge rate of 118,000 lbm/ hour 92250 psi through the power operated relief valve (PORV).
Since the required mass discharge could not be achieved by steam flow through the PORV alone, the effect of varying the discharge quality using various critical flow models was studied.
These results are indicated in Figure 2.
It was concluded that very nearly 100 percent liquid discharge and little or no HPI flow for the 100 minutes preceding core uncoverv would be required to explain the system mass loss by discharge through the PORY alone. Since the new HPI flow (in excess of letdown flow) is believed to have bean 30 to 100 gpm or more during this period and since continuous discharge of very low quality coolant through the PORV for this entire interval is highly unlikely, a leakage source other than normal PORV discharge is inferred by these calculations. Additionally, later comoilations of sequence of events data show that the Reactor Building Sump high level alarm (4.650 feet above the bottom of the sump) was received at about eleven minutes after turbine trip. Since the rupture diaphragm on the reactor coolant drain tank did not burst until fifteen minutes, PORV discharge does not explain the high building watt level.
The core uncovery calculation was continued based on steam generation as a function of the decay heat rate and the portion of the core covered.
The predicted core steaming rate during the core uncovery interval is given in Figure 3 based on (a) mass and energy conservation caiculations and (b) mass and volume conservation calculations. The corresponding plots of core water level are given in Figure 4; an early B&W estimate of the core water level is also indicated on the plot. The plots indicate that the minimum core water level during this interval was no more than three feet and may have been below the bottom of the core.
Corresconding core heat-up calculations during the uncovery interval were performed using T000EE. These calculations indicate that the clad melting point occurred in advance of total zirconium oxidation. More details of the core heat-up results will be documented separately.
(2) Once Throuch Steam Generator Heat Sink Cacacity Calculations were performed to estimate the effects of PORV setpoint and reactor trip response on the response of once through steam generators to the Loss of Feedwater Transient, assuming that no auxiliary feedwater is available.
Design data for the Midland plant were used as the basis for the calculations. Three cases are tabulated in Table I.
Case 1 assumes pre-TMI setpoints for PORY pressure relief and for reactor tria.
Case 2 assumes a reduction in the overpressure reactor trip setpoint to 23C0 psig, versus a 2450 psig higner setooint for the PORY which 576093
d'"
&J3 Richard P. Denise may preclude its opening during the transient. Case 3 further reduces the time to reactor trip by providing a reactor trip on turbine trip.
An energy balance was performed for the first 110 seconds of the transient, which was the time required to boil the steam generator dry for Case 1.
Steam generator heat transfer response and the minimum primary coolant temperature of 552F corresponding to secondary steam pressure control conditions were taken from B&W calculations of the transient. Decay heat rates are based on 1971 ANS data with a best estimate multiplier of 0.9.
A feedwater coastdown of 10 seconds is assumed.
For Case 1, the primary heat sources including stored energy released when the primary system drops from 582F average temperature to 552F provide sufficient energy to boil dry the 109,000 pounds of mass in the steam generator and to produce 592F superheated steam at an assumed saturation pressure of 1020 psia. However, it is important to note that the steam generator dry out process has reduced the energy level of the system so that 538 seconds of decay heat would be required to return the primary system to initial average temperature of 582F.
Therefore, initiation of auxiliary feedwater could be delayed up to nine minutes without rise in the primary system energy level above the initial level.
Case 2 results in about one second earlier trip time and corresponding reduction in the full power seconds generated after turbine trip.
Three percent of the steam generator coolant inventory remains available at 110 seconds, and approximately 585 seconds of decay heat are required to restore the system to its initial energy level at 582F.
Case 3 results in instantaneous trip with less than one second at full power. The primary coolant system is reduced to the assumed minimum temperature of 552F (secondary saturation temperature is 547F) with 29 percent of the secondary coolant inventory remaining. Approximately 17 minutes of decay heat are required to boil the balance of steam generator coolant and restore the system to its initial energy level of 582F.
It can be concluded that the early trip time is equivalent to increasing the steam generator coolant inventory by:
.29 x 109,000 = 31,610 pounds.
Typical steam generator coolant inventory and boil-off data were computed for loss of ac/dc Power Task Action Plan A-20 as follows.
Plant Tyce Power (uw)
S.G. Mass (15.)
Time to Boil Dry
- Midland S&W 2552 92000 17 minutes St. Lucie CE 2570 258000 61 minutes Zion W
3238 357146 70 minutes
- Assumed 1971 AtiS decay neat with nc multiplier.
5%DM
Richard P. Denise JUN t 9 5 9 It is clear that an early reactor trip setting is of little significance to the pcst-accident response of CE and W steam generators with large coolant inventory, even if auxiliary feehater is not available.
For B&W steam generators with icw water inventory (probably less than the design value), an early trip would significantly delay the occurrence of steam generator dry out and provide more time for initiation of auxiliary feedwater.
It is also noteworthy that the steam generator coolant inventory depletion can be delayed by higher secondary pressure relief setpoints (CE vs. B&W),
However, higher secondary pressure also limits the primary cooling at a higner temperature level, so that less time is requireo to reheat the primary system if the heat sink is lost. Therefore, there appears to be no ultimate advantage to higher secondary pressure relief setpoints.
(3) Evaluation of TMI-2 Benchmark Calculations It was noted that both B&W and INEL benchmark calculations produced over-cooling c/ the primary system during the first two minutes of the transient.
An energy balance based on plant process data was performed by hand calcu-lations for the intervals (0-110) sec., (110-300) sec., (300-360) sec.,
and (360-540) sec., in order to better understand the indicated transient and reasons for the error in computer models.
Taole II is a tabulation of the bases and results for the Reactor Coolant System (RCS) energy balance calculations.
The steam generator mass, including feedwater added during a linear ten second coastdown, was depleted during the (0-110) second interval.
Mass inventory of 55,970 pounds was computed from the measured level of 160 inches based on a shell side flow area of 44.4 ft.2 in each steam generator.
This compares to a B&W reported mass of 97,000 pounds which is believed to have been used in their benchmark calculations.
The difference of 41,030 pounds is believed due to the difference between actual heat transfer performance and design heat transfer performance.
Better performance results in lower liquid level in a once througn steam generator.
The additional mass would correspond to an added heat sink of approximately 30,000 F%-sec and would lead to an over prediction of the Reactor Coolant System cocidown during steam generator dry out.
In fact, the LOFW analysis would look like Case 1 of Table I with the RCS cooled to 552F compared to the measured temperature of 577F.
The sensitivity of safety analyses to assumed mass inventory of the steam generator should be considered in future review of plants having once through steam generators.
The energy balance during the first 110 seconds resulted in excess energy of 1495 >%-sec.
Decay heat energy is believed to be at least as great as the estimate and possibly 10 percent more.
Uncertainty in the nutoer of full power seconds or in the steam generator heat sink could account for 57GF 0
Richard P. Denise JUN 2 3 673 the deficit.
For the purposes of the tabulation, the amount of steam that could be generated by the excess energy of 13.6 Nt each second is assumed to flash and discharge through an unidentified leak (e.g.,
a steam generator tube).
One peculiar aspect of the TMI-2 data is the mismatch in reactor coolant cold leg temperatures at the start of the transient. Loop A was at 568F wnile Loop B was i.c 557F. The control system would normally be expected to maintain a much closer match between these temperatures, e.g., + 2F, and the reason for this condition should be investigated.
The (110-300) second time interval should be ideal for an energy balance. During this interval, the HPI was known to be operating at full capacity and was the only heat sink other than normal heat loss to the containment. The RCS average temperature was nearly constant during this interval. The only heat sources were decay heat (estimated uncertainty of -0 + 10%) and the reactor coolant pumps. This heat balance shows a large deficit of 5219 N-sec or 27.5 Nt per second. The only plausible explanation of the deficit is flashing of the primary coolant. Sin:e at least 10F subcooling is indicated during this interval, flashing could only occur at a leak location. Leakage through the stuc'.' open PORV would be suoplied by flashing in the pressurizer, which was analy:ed separately from the RCS heat balance. Leakage flow rate for this energy deficit would be 47.8 pounds per second of steam, compared to 26.9 pounds per second for the deficit indicated during the first 110 seconds.
The balance was continued for the interval of (300-360) seconds when the RCS temperature rises to 582F and reaches saturation tempe ra ture. An energy deficit of 12 Nt per second (equivalent steam leakage of 19.8 pounds per second) is indicated for this interval. However, there was greater uncertainty in the HPI coolant injection rate and in the energy sucolied to the reactor coolant system during the reheating. A :alculation performed for the six to nine minute interval with RCS at saturation and rising in temperature showed only a 3.6 Nt/sec deficit, indicative of very little flasning during that period. The latter result is surprising and possibly indicative of icwer quality leakage from the satura ted system. However, the calculations may be in error after six minutes since the cressurizer cannot be properly separated from tne RCS heat balance after saturation is reached.
Table III is an energy balance of the pressurizer to evaluate the calculated leakage and the calculated level based on the system pressure history during the first six minutes of the transient.
Time intervals were chosen to match Table II except that no balance was made after six minutes when saturation temperatures was reached.
An equilibrium pressurizer model was assumed with flashing energy su:: plied by all of the not fluid in the pressurizer at the beginning of a calculation interval.
RCS water was not incluced in tne balance.
5N1CT
Richard P. Denise JUN 0 3 W9 Steam relieving rates through the PORV were normalized to a discharge pressure of 2250 psi for comparison of discharge capacity during the three time intervals and for comparison to the rated relief capacity of 118,125 lbm/hr.
Excellent agreement with rated capacity was obtained, particularly during the (110-300) sec. period when the calculational uncertainty is at a minimum. This tends to confirm that pressurizer leakage was via the stuck open PORV and an additional RCS leak is needed to explain the mass balance.
Also of interest is the computed liquid level versus the measured liquid level. The calculations indicate that the indicated level was too high by 35 inches at 110 seconds, 151 inches at 300 seconds, and 137 inches at 360 seconds. These calculations are believed to be reasonably accurate and indicative of substantial error in the liquid level reading during the first several minutes of the tran-sient. The high indicated levels suggest flashing in the reference leg during the depressurization.
In sumary, the following conclusions were reached from the benchmark calculations.
(a) The B&W computer model for CADDS benchmarking of TMI appears to have several deficiencies; e.g., too much water inventory in the steam generator; 3% heat demand to simulate HPI cooling effect from two to five minutes (this is too much); etc., which could be pursued to cbtain a more acceptable benchmark.
(b) The INEL model had several problems with improper handling of auxiliary feecwater being the largest error contributor.
They are now aware of these problems and are making appropriate corrections.
(c) All calculations seem to point to leakage in addition to that through the stuck open PORV.
(d) A reactor trip on turbine trip has the same effect as additional inventory in the steam generator and appears to be of no value for plants having steam generators which reduce the prirary temperature to near the secondary saturation temperature without boiling dry.
57G1.CS
GI3 jf,. 3 Richard P. Denise (e) Safety analyses sensitive to steam generator coolant inventory should be reviewed carefully to assure that conservative water levels are used.
( f) There remain sufficient questions about the TMI-2 response to warrant additional bencnmarking analyses.
i
~ ~ h., u
~ ~,m c,
Laurence E. Phillips, Acting Branch Chief Analysis Branch Division of Systems Safety
Enclosures:
Tables cc:
F. Schroeder R. Tedesco V. Stello S. Fabic Z. Ros: toc:y P. Norian F. Odar E. Throm N. Lauben B. Sheron G. Holahan W. Lyon J. Holonicn Gordon Edison
/.11 TMI Task Force Chiefs R. Bernero, Task Leader, Task Group No. 5 4th Floor Arlington Road Building R. Vollmer R. Mattson D. Ross D. Eisenhut 57G13
TABLE I PRIMARY COOLANT ENERGY RELEASE TO BOIL DRY THE STEAM GENERATORS
Reference:
Midland Data CASE 1 CASE 2*
CASE 3**
(Pre-TMI)
(Post-TMI)
(Post-TMI)
PORV Set Pressure (psig) 2340 2450 2450 Time to Reactor Trip (sec) 8.93 8.0 0
Time to Rod Movement (sec) 9.63 8.7 0.9 Feedwater Flcw Rate (1bm/sec) 3500 3500 35C0 Feedwater Added During Coastdown (lbm) 17000 17000 17000 Steam Generator Secondary Water (lbm) 92000 92000 92000 Total Coolant Mass Boiled (lbm @ 1035psig)l09000 109000 109000 Avg. Reactor Coolant System Temp. 9 0 sec. ( F) 582 582 582 Avg. Reactor Coolant System Temp. @
110 sec. ( F) 552 552***
552***
Avg. Secondary Steam Pressure (psia) 1020 1020 1020 Estimated PORV Flow (avg. lbm/sec)
Power Level (Mwt) 2552 2552 2552 PRIMARY HEAT SINKS (MW-sec)
(a) Steam Generator 74157 74157 74157 (b) PORV NEGLECTED PRIMARY HEAT SOURCES (Mw-sec) to 110 sec. after Turbine Tr.p (a)
Full Power (2552 Kat) Energy Input 24575 22202 2297 (b) ' Fuel Stored Energy (1350F--y550 F )
12913 12913 12913 (c) System Stored En gy (582F--7552F) 25392 25392 25392
-(d) Decay Heat to 110 sec. 9297 9297 10057 (e) Reactor Coolant Pumos (
18 Mwt) tu 110 sec. 1980 1980 1980 TOTAL ENERGY SUPPLY TO S.G. @ 110 sec. 74157 71784 52639 FRACTION OF STEAM GENERATCR CCOLANT NOT BOILED 0
.03
.29 TOTAL DECAY HEAT FOR ADIABATIC HEATING OF THE SYSTEM TO 582F ENERGY LEVEL 34689 37062 56967 TIME (sec.) recuired to generate decav heat 538 585 1006
- 0verpressure trio 3 2300 osig; S.G., would not be dry 3110 sec.
- Reactor trio on turbine trip; S.G. would have substantial inventory 9110 sec.
- Primary coolant tercerature cannot be icwered below 552F due to limiting secondary saturation temcerature of 547F wnicn limits further heat transfer.
57Gi$
TABLE II TMI ENERGY BALANCE BENCHMARK Basis: Plant Data Describino the Accident (0-110) sec (110-300) sec (300-360) sec (360-500) sec P'ORV Set Pressure (psig) 2,255 OPEN OPEN OPEN Time to Reactor Trip (sec) 9 Time to Rod Insertion (sec) 10 Feedwater flow Rate (lbm/sec) 3,180 NONE NONE NONE Feedwater Temperature ( F) 463 Feedwater Added During Coast-down (ibm) 15,897 Steam Generator Coolant Inventory 55,970 NONE NOME NONE Total Coolant Available (lbm) 71,870 NONE NONE NONE Steam Superneat Temperature ( F) 592 Avg. Secondary Steam Pressure (psia) 1,035 950 865 825 Avg. Reactor Coolant Temp.
9 0 sec ("F) 582 582 582 582 Final Reactor Coolant Temp.
Avg. ( F) 577 578 582 595 Avg. Primary Stean Press. (psia) 1,900 1,560 1,375 1,425
- Estimated Leak Flow ( Avg. Itm/sec) 25.9
- 47.8 19.8 6.1 HPI Coolant Added (Gallons) 243 2,136 200 600 Power Level (N t) 2,688 PRIMARY HEAT SOURCE (Mw-sec)
(a)
Full Power (2688 Nt) Energy Inout 26,880 0
0 0
(b)
Fuel Stored Energy (1350 F-T Coolant) 12,590 0
-81
-212 (c)
System Stored Energy 4,250
-850
-2,800
-9,100 (d)
Decay Heat 9,792 13,290 3,604 9,970 (e)
Reactor Coolant Pumps (18 Nt) 1.980 3,420 1,080 3,2a0 TOTAL ENERGY SUPPLY 55,492 15,360 1,804 3,398 Fraction of S.G. Coolant not boiled 0
~-
TOTAL DECAY HEAT FOR ADIABATIC HEATING OF THE SYSTEM TO 582 F ENERGY LEVEL 14,042 Time (sec) Required to Generate Cecay Heat 175 sec PRIMARY HEAT SINKS (PW-sec)
(a J Stean Generator 52,610 0
0 0
(b) HPI Coolant Heating 1,167 10,261 961 2,282 (c) System Heat Losses 220 380 120 360 (d) Daficit 1,a35 5,210~
723 656
" ENERGY Deficit ::er sec (Nt) 13.59 27.2/
i2.05 3.6a "For 1005 cuality leakage from the RCS (excluding the pressurizer) to cc-oensate the ener:y de#icit
" Note nat a icwer average energy discnarge indicates tnat lower quality coolant is ceing discnarge.
5761C6
TABLE III TMI-II PRESSURE VOLUME / ENERGY BALANCE (0-110) sec (110-300) sec (300-360) sec PZR Pressure (psig) 2150 - 1745 1745 - 1400 1400 - 1355 3
V (ft /lbm)
.1685
.2275
.2275
.2974
.2974
.3094 Cociant Properties:
g V (ft3/lbm)
.0265
.0245
.0245
.0232
.0232
.0230 f
h (BTU /lbm) 1123.3 - 1153.3 1153.3 - 1172.6 1172.6 - 1175.0 g
h (BTU /lba) 689.1 - 641.1 641.1 - 600.7 600.7 - 594.7 f
~ ITectric Heater Input (Mw-Sec) 180.18 311.22 98.28 Pressurizer Volume per ft. Level Change (ft3/ft) 38.48 38.48 38.48 Reactor Coolant Average Temperature O F) 582 - 577 577 - 573 578 - 582 Fraction of Hot PZR Fluid Flashed
.0937
.0706
.0103 LICUID VOLUME (V ), Ft3 795.35 - 604.4 604.4 - 811.2 311.2 - 932.a f
DELTA LIQUID VOLUME (.V )
f RCS Shrinkage, Ft3
-99.4 0
-99.4 PZR Shrinkage, c 3
-60.0
-32.1
-7.0 Liquid Flashed, Ft",
-68.9 40.4
-5.3 Liquid Boiled by Heaters Ft3
-8.6
-12.5
-3.7 Liquid Added to RCS by HP! Ft3
+45.9
+291.9
+37.3 STEAM VOLUME (V ) Ft3 g
704.65 - 895.6 895.6 - 688.8 688.3 - 567.6 STEAM VASS (W )
g Steam at Beginning of Interval ibn a181.9 3936.7 2316.2 Ocded by lashing, Ibm 2312.6 1741.7 231.4 Added by Heating, Ibn 350.
Saa.a 161.3 Steam at end of Interval, iba 3936.7 2316.2 183a.5 Salance los thru 30RV, lbm 3407.8 3906.6 57a.9 Ste?m Relieving Rate (lbm/Sec) 31.0 2.
g la.6 TABLE III (Cont'd)
Equivalent Relief Rate (lbmAicur) @ 2250 psi 134150 114502 92279 PZR LIOUID LEVEL: Measured (in) 220-145 193-3'6 376-400 Calculated (in) 220-160.4 160.4-224.9 224.9-262.7 57G1C F7
9
.-e._
.----e
-M-+.-.'-=---
_-e--.-
e---m-
--e+-
ee---e e=
- ---W Rb R.Q(~I$ C Q Q We } T1-. -
e-t.;a.
6._. - s a -.-
4
._ _ y
w-_-_.-..
___ _ Q x_-
o y
Q N
M
=
- _ _.. _ _.. _.- _ +. _ _ - -. _..
g_-.--.__
>=C ~~
1
_.y-
.f W-m
-W-M<
x w.
__.g.
a* E
-N
-a
- c. rw.
__d t
1_
__,. A y
,p
- py4-f----
-T O y; M".
\\
s_
_v_
g.--
4 h4 O
C'--'.-. _ __~,__
_.5 0
__.-'g_
++ F 1
- p_...
_.._g r=
m&
m t
+- -
- q-
- T --
w-
.a___ b
_.._g i_
~~ -.
s._.
3_
.. g_
, _g
- g-g.. _
__w
.. w.a
+
w ^
t_.
O --t --
TJI--' !_~.-
~
g 1.
WS-g
.C N I,, -
v_-
-- n 2 -, -
r F
o C 771 - _
n_E w-
_.__ m
.e m-A
$ -4 e
A y
~
v C
s%e, lISEY YNh,_d. i_. h.-
dE MJ{ [
NN b_
..L_.-
....y5-__
._2
~
W m.
-W
+--\\
M.
k W d*y
.k.._.
O N
m,___
"A'_
m,.
c_--
_y-v.
w s
%m.
m-_.
r-h__-
+
q--
"Q
,,4_P D.
~
Z p+,._
~
- cJ D
'^-
- c. _
~ ~~ T ^ ~ ^^: ^ X ~
~ ~. - - - -
~
W s7cics
i mms ;&:mpr6sD _' /Ane - Y/D ~
l r
- d --- - -- S
---E--
h j
0 1
4 u
o
-t
(
i t
i Qt>
1 i
=g-.
i
.w.
' i,k (: -
. _ ;- - j. _ [. a-h]
i
=
_-1..,+=.
m - g g _.-.__ _. _.
gs g.
g.
AT
- s..
k
! 9 9
'/l
?
i
)1 \\
i/k!
6b
\\
' NN i:
by b
-!/ 5 9,
f.
W is p\\
\\
W LL :(
/;
J 5S b
N Il i 'T if
-8 Nd M
W i \\\\
/
/1 4%
v i
'l
\\
/i S
-/i $
[
Q Il c.
d :i
\\l/~.
~
- N t/' #
A d
!j
'1 "1
i
\\-
A-
-1
/ v.q
_2
',1 b
e!
m 3
~
- (-'i / k
'/:
Ah 2
ru.
i T
tA T
- /
Ag
/
p e
i--
- v t
1: A
\\
/ -
R 1:
ss u
3 1-_-. y' : 3 -- r - g r --
y
. gh yy 3
p t,
- \\-
g y- :
l e
=
. _.._ g.,.
. - _ p. p._
. g j - _., _ _3_ eg_ _. c g~
u w
L i_
V
!\\ -
/
~
3.
- 1 1
---fi
-i :N
- /
s.
Q
. 4
-p2A-3\\ =.. : _.
g g3
.n u
x3--.
-f
-e r-
/
t y
3
\\
-f -
x:
k-l_
--I
- t
- \\--! /
A d
S y
%! - 3mi t
~---sg=
v
, i i
,~
rd O
1.
0 4, '.
. g..
g.
.)' 3,.
g,
-- g
~
- j....j.
djM
(
.-1-i t xw s
t 1-a my%
.e
+
e
.c s ~e u
r h
' ~ -
.._g 4_4&__a_
3-3_ _ _
y _.
4 G '
34 a
r
== t - _ _
3~
3 f
1 g-1 f
.~, Q#
C, *.",,
G 3E [^
^$
k QQ \\
O2_. ; ',
. _ _ _u
- g S
a~
\\
576P,0 *l
GT].v v, a
u
-- _.=_ - - -
_. c b,
n G.,
- m --- - - - ---- 5 =----- m p :
2:1:.-___
= -- ---
_C.,Q
-n2-mn_--
- -us
_. = _. _ _ = = =.
._==
9
- =; 2m_ -
s
_.-g__.,-
en.
w.
e _.. g= q ---
g__..__.
w'=g%
p
~
.. _. r-._-.
h,
.w __
9._..
p
- -. 2-~ ^ * ^
m'-
%,ms
--_-._._'.'_'__~.__=__~~._.~._'_.__.__..__1_!..=_._.__.__._.__.._.._._..
._!.__._~~.
~..
-_,.a__#__.__., __ _
M
..._l~___
.!~-'l_(7~.~~~^*'~_--' (^'_~~~-'-_-.__._ _. ~~ _.. ~ - -~~ *~' 11.: l ' _ ^~. d C.~.'~ !. _.:.'-'I_
~b--'__..~
e r3 ____
_. _ _ _ x._
. = __x _=_.=.==;===__-____.__
_=..._
? ~-
~ ~ ~ '. ' ' -
'kT-A
- 5: ~ ~ ' _ _.
- ' W.
~ ^ h ^:5_
~
m==.:__._2_=_==
_- - - --- -=:-
^-
~
-^-~~_:
._-^2'^~
---^..:. :: ~ ^" ':-_5
_ L -~ ^ M. ' ' '
_2,
' ^ ~ ^ ^. ^^
..u :. _ _^
'h _
__...2
. = ~ _ =
- -- M _ -' n; b - !_. _.:r-'
2-
=2
~~_
q
-1
~
- ~~::!
.i__ _..
-_ni: : =__!. :-.__..-. -- _ _. _ _ _
_. _ _. _ _ _ ~
._.e
i..~,..~_C'.i'-~-^_.
- -*!.l_--~~-~.'_.__,_--'_.._--_*._---*"'___,__*:--___.._ = ]
~_'~C-*-.^_-
~ ~ ~ -.. _ ".. _..M'~
4
_. _ - - _~ -
e.
~
_ & -- ~---- - =i' ~ L_-~L. =~:il: 5 g;^Q_ -i.~ :
- _-_^~ _-
-- L_:=::-_ -- -. = -
- = :
.= : _ r _= - = =2 E===25:_ ^ .&i--" '" y-- _ _ _.
Q ;..=-
Q._ z-- - = = - -~_l. _. ~' '. ~ -- .. _ _. _, _. _ _ - _ ~ -~ I ~ _ _ ' '*d =_ _-.-_.u__--_=_z__-._
- =
- _.-- : __.. :....: _ -- =
___,.-___3 ' ' ~ ' ' _.. 4.
- ____.____.--_'!.~'1._--"'^71.!_-^.
. _.. _ _ _.* - ' - ~ -_ __.. [_.).---* _ _. ~ _ ~ ~ ~ _~' _~1._~~~ - -j; } ~ J ~-_ Q =..-' ]_=====q_jz_=f = c1.- - ---- *~+=_=_ i _i .= = 9 := =_- _- _= _ _ _.,_._.__ _l.. p..e.a3. e 699 88 88M ,,s - _ _ ~ _ '. - ~ _--.. _. ~
- " ^ - " " " ' " - * ' " - ~ - * - ' - - ' ' -.
b ~ ~ ~ ~ '
- '*'~
3 '.y__h hff b
- ^
~-} l - ^ 'f^^*~-'-_-*-^^^. .~:;-h '_^ _^ ' N'- l.,_',._ ._. ---. --. __-.- _- -._ - _ - - m._ .., &a g_"1_ _ _ _ _._ __.y_, g g n - _. L71 --*~':*.~~~"-'-~~~--*._*_.._-:^~.--_'-_.'__'__.*----I~*I~_____.--'.' - ' '. ' _ * - ~ _ - -. " ^z-'=.-~'-^~'.:- Cam ~:. a.. ';: : : :~.= : ^. :=._ -~ ' ' ' ~ ~;: ^ ~. ^ ~ - * ' - -*_- ' Q
- u g,
_... : :--. ~ =--. -. .g---*----- _ : - _- n- =_:v_ __ -._.==--._.~:__-=:;,n--.. ..cg ._-a 5.--!__- .2d-- '. _ $ 2 _ =_ _. - _ _.._ _ -. - -. _ _ _. -. =. -- M __...... _.-
- ^
- 3 k m w ~- f ' - ^~ : _. ] _j = ~- ^jl' }~-~-~ij._'. ^i;- "'.l.~= } ' * -- A- - -'-_-W=3 - ~ -
- ^----- -
_:2 -1 =&. _.. _.. _ _' _ _ 3. _ _+..:__^=: i_- =_ _ i _~ _ T_ : ~ ' _ _.^ _=_ =- ~ i, s. :-E. __^ _ _= _ __-. & 2. _ _.. =_ _+_: _ ~ ~ 2. _ c. -.i._.._. T = c. _ _- =_ _=E. _E_ _ _: ~E _--- z_ _ :_:=_. =_i.=_ =__: -_E_-_. 5 a_=_. _=5 =_ =_--- = = ; -_ :==. _. - z -_--_--_.w.- +- {' - m } ]*- - - - - + . = = :- 7.;- 7.- _. - n.- ._.4-~~_.. -~_ j __ --f __ _ _ ; __ ~ -. g - _1 p < g _ _ - -i = 7_;E - jg ;=-g "ggyu. f_ _ ; 7 ; _=_=. -- - __ 4 _=;g -- _m _-=_=-.__=.=__=.==yq . _ j., __. _ _. _ =; _; .m .= _ _ f.: _. m ~ _ = _=_ _~ =_ _ ~ " / ' L --__ ; f._y: _ _ _ L_; ? _ j z_ ~ _.. _i. = g _ _ _. 7-- 1 t --. =q + + c ;=== 1:- ~----- ( _ y; _* - u==:
- x== : 25E=- g4 =a
- . :~: : :~~..Y~.:.~T^*-._._J-~g
= = - ---- =-
=-.
_== _- .__. _ _ - - - -_ _...~~ ^^'T^ '-.__=__~'r~ - =.. _ _ - -... _ c _ .7._._==__._. __,__:. - j m. __ _-- - -_, --________*=__r.___-__._______: __=:=__.__._-:_- .__[^'.2 -._'.._ ',_7_. _ - - - -. - - - - - - _ - _" a M 2 2 * ^^,- - - -D ~ ..__;=
- =
^ . L __--- :.; - ~ ' - ' ' * ' _ _ --~": _ _. -_... = - ' * ..M_^-'_ .. ' _ = -l._. _.- - - _ ~. _l_ - ' ' ~ ~ p'~,_..____._,.__'I_- v.- p -_.._.-g-._-_._.-. A -~~-:..~~-'Q ,-'..,.,i_. a _ l ': 2 ~ -.-- ^ _.-.e. -. g _.-._g ..__ wa ___-g.==:--~--
- -~ W
~ - - - . _. _ L _ _% = z ~ =- t- _..: _ =._ = 4 :: = = =_ g .. _::- z = a - =- 1 = ---__.-..-A_ ~.~'1 4 [ { 7. S '~
- i. 1 1 = '_r:_.:- j== :Z=~Ei
~=f - --~ _ _ j _._^ _. _ j __ ' _ _ _- j Q =, - - 't_ _: ===: :_i.T=- -R-T - : : _- =__u~* _..._a - - -------__:===:=-- T'=*:' ' ^ ' ~ ~ ~ _ ' * ~ ^ ~ ' ~ ~ '. _ ---:_= _.. _ ~==_ _u =.:=.: _ _,_..___....-~~L-^*~~^_-..^~*~ .--_.-.-_.___3--.-*~ ' ' ' ' *._:.^^-*-^~=:.*._._..__ _..._^*:._ .. _.. _. - ~. = = _ - -'".*~-'--*__~_.1-^_;: .____L
- ._-*' ^
\\. --
- _ - - - - - - _ ~
- - - + - _ _ _. - - __ e
1 kW k.C 10 w y n c r,* &&s8 9 19 e [ y g _._3 ..m Cf ~~ g . c, ___.. _.. _. _. _. _. _ _ _.. _ -. _.-. _ v
- 1
- ' ' ~ ~ ' * ^ _.. ~ A.-.a^ ^ - - - _. _ _ . _.. -.. ~ ~ - _. ^ ~ ~__^-_---^~.':--~_-' ~ _r _ _ __ c. 4 _ _ _ __ _ __. _ __.u_= = h. .. ". _-h=_=______=.=-._._-=.-__=..=---.=.=- _ =__.- a ; = = = - ^ = -- = = : ~i ^ = :
- - = - + = : -. -
Q --T--^----:- . ~ l'i ' ^ ' - _-~- =- _=. ': -i Ti.=~ z ~ _ : -.. __ - - i== L 2 = _. ^- - =_: = =_ =L==. _._ _= = =_: : :.. _ }..==. - -. _. k ^:--. _j_=-R -.g'1'-^-.._._: _ _.,*^- _ ~ _. ~ ' _. _. -. - _ - _. -. _ - - - - ^ _ m __ y - ^ ^^ '^- _. 'i K. __~'^f_.~
- ~ _ ___ q g ' ^-
- _
- - A== __
& -^ =! ~ ^^ ~~ -. -8$ n i :^^' '.-W = ~~;5 --- _5_- - _= - - : -i -- -. E_.1 --& - =
- - = -.. _.
I
- n. __..
__ d_. _,l. L- - -.-~'}~~ _ * - - ' - ~ -. _ - -lJ ~ _ _ ^ ^ ' - ' ' ^ - - ~ - -. - ~ ~ - ' i ~~} ~ ~ --. e.. =.- _. _ e_ _ = - :.__ _...-_ s - ;=_ 2w. =_== - - - r---- = _ + - --% =_ t d- =~ _--d ' _ibi E-[. _ _i-- ' _ - h -b__ -- - '_ _~ ~; 5-- ~ ~ _ b-d _- :J '. - ' - - - 1 l_~__;_
- +=_
l_~
- ; '=_ t 7_. =--j,=-----j=_~;=--~~__. _. -- = f 3 ;' ---j __
- ' ^._
.. -. -...__;___._~:. _ _.'=T_*~~-^' ' ' - -.. _ _ _ _. _ _.. _ _*_~'7=-'~~~~~.~__J.-'__I_=.Y_._ _ _1 :-~~~ :-_ _. -
- .. ~ ~ ~ ~ - -
= - - _. - :-f== =.=i==: = =. -. _ _. _ ~_ ~ = 3_ i----- - -j G ; } = - = p- .--,__._a . - _. b ** t ensamm % p __ L_. _. _-.x y===_g-wr = 2 y - - y -- 3 7 =-- =; g n =
== =i e_. ; =a==:=i=3_ -;g = E,=+ M412=2_.=. =i-4 =g _+ =; = ;
- -- =_ = _ - - = -
-.._._ A, w '_. _ _ _ -. _ - _ -.. _ = __:.-_._. -m _ _=_.. - -._.,_:...__z_..__%...--.__.________-..e. _ -. - j _ _ -. _ _ --- = a_ g 4.-y=._-.._==__.--___.,--_._-_-_=__----_____:__:__ ~'~-- -1 ~ --* N { { _ -.. q ---* -. th-~. 31. 7-- 2 2 21--i -~ _Y--~.. _ ~. -- ih - - ^ -. _. --'__ .=_=.'__=_'___*_."1:-_=-l--" - =;- ; =. m - - - - - * -_.__-=_!--*,_--'7/, p_. z._ "m -
- w u.
.
. - ~ .-Z__ _ _ ~ _ _ C s= 4 .__'g .__u__ g _g_ =4 4 _ _. _ _ a _ g __ _f _g__. _____g__";_._'. 'w g 7- =n. _.. - w g = = -w:== ={=. 2=-@@ = = = g _ g -- -- --_ =s= y= = - _a ~ _. _ ' ' X ~:- :-- ~5_--;G _ } 1 -- --'"; m = l _:'--i - f: T~ -- h -5 'i.~ .(:Y ~1 c-- - 7- -=4-.m w.c gj g g-4 f., =-- = <,_ = = r e c,- --._=.-_----,= =.x._._ p__ -_=._.q._ _. _ =. _ -. =- .i. -,v q __._.._.__ _ _ g._
- __n, 1
y m . ~. r.. g_... __ *.- =d - _ h { - i---_.;_.~_~.:_:._~,-----'------j~ ' - ' . b=14 J= - _. e( =; -~;7 -- ~ -i ' -+.. = U~ i=- ~7[ ~ i i - _it ] --- -_ 1=.-7 E_____.-. _ _. ; _ 3 f
- -~_'-. _ _ _.. ' '. _. _ _. _ _ _.. 3 _J ; - M""_~~
' = _ =.. ". - ~~ : n' r-- - y . _ _. - -.. = = ~C 3z _ :-:_=.__:.=_.--.------.-_._=--- _ _ __..=_=._:=-=z_=__---=-_-~. =_- -z._,-_.=_~.__.__., =- -z x'; G: - -.: z =.=__T _ '. ::--- - = p#_ _ -. _ _ _ _ _. _^ Q =- - - = - - -. _ _. _T_ -'-. _ _ ~ _ ~.. -. _-.-_...L--~.=1' _ _ _ _.. _ _.' _ - '. _ - ~ - - _ _. _ ' ' - - - '_-~_-n----~._:._.--T, g --- _ 2_ _ _ _.. - _.. _ _ ~... _ - -. _;C 7__.3-. -bi-- ~ -~~;---'*:..:_ - - * ' - ' + - - - = 1. ~ . =. -- - _ _ _. .=_.-.__-=:_====----- _ _. _. - _ - _ _ -_.. =. _ _ = =.----l===...__ . _.=;- Q _'-,-. :, =_ _:_m .._ __._-_; _. ;_.<.-.__.-_;....___ ' _,..g y 7-__. : a. _. .- %. - 5 : a _ L_ Q--- .__ _:. _: _ r-- - - - - i _ 2 z._2__:;2..--a---^.-w___..._--_.. ^ ~ - _-___=.;n_
- l. -. _^.1. -~~ -z _ : -- = __
..c _ - - - - - =~ = = x; - - :' ' - - ~ ~ ' ~ _ ~...; ('5Y ~_' -- -:_-'_.._._i __.. 5 :$. _._ '_Y N_ ~ ). .f1 ik~
- Q [ ~. [_.1 p.. ',,}",'"2
-. ^*~_.-~_~-'~;-. p__'_-.--~._._.':'ZI'.!.___,'_;4_'_:-__1..'-~.~_~'.--*.-__~- ~~..-'---*^___"_1__'_".-.-1__7_'__._'_"..~7____._.___;._"!_!_'*'__~!C_.=.__E_*'._._-.... 4 ? *T- .___3 _ _ - ~. _ '
- ___'__'_:m'_.__.-._ _.. _ _.
'L'.
- {__^
.__-.__.:-=_______;_--~.:_T*~-'^ y__ {' .__;_-----;_-.=-----~~.--~~1-*_-:'_--*--~'*-'.=-__;~~.J!.__~--^ .}}