ML20209F013
| ML20209F013 | |
| Person / Time | |
|---|---|
| Site: | Maine Yankee |
| Issue date: | 09/04/1986 |
| From: | Whittier G Maine Yankee |
| To: | Thadani A Office of Nuclear Reactor Regulation |
| References | |
| GDW-86-207, MN-86-115, NUDOCS 8609120017 | |
| Download: ML20209F013 (22) | |
Text
!
o a
m l
MAIRE HARHEE ATulMCPOWERCOMPARUe EDISON DRIVE AUGUSTA, MA!NE 04336 j
(207) 623-3521 9
September 4, 1986 HN-86-115 GDH-86-207 Director of Nuclear Reactor Regulation United States Nuclear Regulatory Commission Washington, D. C.
20555 Attention:
Mr. Ashok C. Thadani, Director PHR Project Directorate #8 Division of Licensing
References:
(a)
License No. DPR-36 (Docket No. 50-309)
(b) USNRC Letter to MYAPCO dated June 24, 1986
Subject:
Response to Request for Information Concerning Spent Fuel Pool Masonry Hall Gentlemen:
l The response to your June 24, 1986 request for aaditional information, Reference (b), is enclosed in Attachment A.
For clarity, your requests have been reproduced with our responses following.
This information should provide the technical details necessary for you to evaluate your concerns identified in Reference (b).
l Please feel free to contact us if you have any questions.
Very truly yours, MAINE YANKEE ATOMIC POWER COMPANY bh0V
!bk kDOCK O 000 09 G. D. Whittier, Manager P
PDR Nuclear Engineering and Licensing GDH/bjp Attachment l
cc: Dr. Thomas E. Murley g
Mr. Pat Sears O\\
Mr. Cornelius F. Holden (0
\\
7997L-LM0 1
}s
-S o
MAINE YANKEE ATOMIC POWER COMPANY ATTACHMENT A RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION CONCERNING SPENT FUEL POOL MASONRY HALL NRC Reauest 1 In your submittal dated May 26, 1983, you stated that boiling of the spent fuel pool will not result in offsite dose releases in excess of technical specification limits per performed analysis.
Provide the results of your analysis of the acceptability of pool boiling in the event the fuel pool cooling system is damaged as a result of the falling of the masonry wall.
Also, provide the results of your analysis which confirm the adequacy of spent fuel cooling in the event of the failure of the masonry wall as indicated in your submittal dated December 11, 1985.
Maine Yankee Resoonse la A loss of spent fuel pool cooling could potentially result in the boiling of the pool water.
Subsequently, the increase in vaporization of activity bearing water would result in an increase in gaseous activity evolving from the spent fuel pool. A calculation was performed to place an upper bound on the amount of gaseous activity released as a consequence of boiling in the spent fuel pool.
Itwasassumedthattheonjysignificant quantities of activity relegsed are in the form of H3 and I dl with concentrations of 6.3 X 10-' uCi/ml and 2.0 X 10-6 uCi/ml respectively in the spent fugl pool.
(Recent measurements at Maine Yankee indicate H3 at 8.4 x 10-4 uCi/ml and 1131 (t 1.2 x 10-6 uCi/ml). Also, at a boil off rate of 100 gpm, the H3 and I13l activity contained in 33%
fuel pool water is released in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
The pool ofthevolumeofgpengofwaterwhenboilingbegins.
contains 5.8 X 10* ft Decontamination factors of 10 are assumed for 1131 as it vaporizes and as it passes through the charcoal filtered spent fuel pool ventilation system. The analysis has shown that boiling the spent fuel pool would result in release rates that do not exceed the technical specification limits at Maine Yankee.
Maine Yankee Response Ib An analysis was carried out to assess the coolability (and continuing clad integrity) of spent fuel stored in spent fuel storage rack cavities whose exit funnels may be blocked by debris or portions of the concrete block wall adjacent to the spent fuel pool, in the event the wall should fail and fall into the pool.
Analysis Assumations 1.
Although the fallen debris may completely block the exit funnel of the storage cavity, a coolant flow path exits through the four half-inch diameter holes drilled in the angled portions of the storage cavity below the top of the funnels in the new racks. Steam or water exiting this flow areas reaches the open pool area by traversing the intercavity annulus underneath the wall debris.
(Figure 1).
7997L-LM0
MAINE YANKEE ATOMIC POWER COMPANY
, 2.
Bulk pool conditions were chosen to bound the range of potential conditions which might be experienced following the wall failure.
It was postulated that debris from the wall could cause damage to the spent fuel coolant piping, resulting in loss of pool cooling and subsequent heatup and boil off of the pool inventory to a level just above the top of the spent fuel storage racks and accumulated wall debris. Thus, a range of pool bulk fluid temperatures from 100'F subcooled to and including saturated boiling conditions were considered in the cooling analysis. Saturated pressure at the top of the fuel racks was taken to be 16 psia.
3.
Since it is impossible to determine the exact flow path once the flow exits the half inch diameter holes, the flow resistance after the coolant exits the holes is neglected as insignificant.
In comparison to the loss coefficient associated with the holes themselves flow resistance is negligible.
4.
For purposes of calculating the flow through the inside of the storage cavity, the spent fuel axial heat flux distribution was assumed to be uniform.
For post-dryout clad temperature estimations and comparison to critical heat flux (CHF), an axial peak of 1.2 was assumed.
The latter is consistent with the flat axial flux shapes found in the fuel pins at end of cycle conditions.
Acceptance Criteria Two acceptance criteria were used as indications of adequate fuel cooling:
1.
No CHF is predicted to occur in the stored fuel in the blocked cavity.
2.
IF CHF is predicted to be a possibility (due to the uncertainty in the applicable CHF correlation) then predicted fuel cladding temperature must be well below 900*C (1600*F), the point where sustained zirconium-water oxidation would occur.
Methodoloav A model of the blocked spent fuel storage cavity was developed, which solves the momentum and energy equations for flow through the cavity. The model considers elevation, friction and form losses as well as spatial acceleration through the cavity for the single and two-phase regions.
Four regions were modeled, the subcooled inlet region (if the pool bulk temperature was considered to be subcooled), the two-phase boiling region, the single phase steam region to the top of the active fuel length (if dryout occurred), and the single or two-phase region from the top of the active fuel to the exit flow holes.
The model of the blocked cavity accounts for two-phase flow multipliers on friction and form losses and used the Zuber-Findlay void-quality model (with constant Co, Vgj).
7997L-LHO
MO!P$E VANKEE ATOMIC POWER COMPANV
, The model was used to determine mass flux through the cavity as a function of stored spent fuel heat flux and external driving head. The external driving head for flow through the cavity consists of the elevation head of the surrounding bulk pool fluid (based on assumed bulk pool fluid conditions) over the vertical distance from the cavity exit hole elevation to the bottom of the spent fuel active fuel region, less any flow losses along the path from the open pool region to the blocked cavity inlet, taking into consideration the flow drawn along the same flow path to cool fuel stored in unblocked cavities fed by the same flow path (Figure 1).
The exit void fraction was determined from the predicted exit quality (here exit refers to the top of the active fuel and not the exit flow holes) using a more conservative void-quality model which used Co - 1.0 and a " ramp" model for Vgj, shown in Figure 2.
The exit void fraction was then compared to a " critical void fraction" at which the Zuber-Griffith CHF correlation would predict CHF to occur. While the nominal " critical void fraction" at the low heat fluxes addressed here approaches 1.0 (Figure 3), due to the scatter in the data at higher heat fluxes, it was considered prudent to use " critical void fraction" of 0.8 for conservatism.
Finally, for cases where the critical void fraction was exceeded at its conservative value of 0.8, the fuel cladding temperature was estimated using the Dougali-Rohsenow Post-Dryout heat transfer correlation.
Results For freshly discharged fuel stored in blocked cavities with decay times greater than about 80 days, no CHF is predicted to occur for any bulk pool condition (i.e., saturated or sybcooled). Predicted cavity mass flux ranges from 1250-4150 LBM/HR-FT', depending on the number of intervening unblocked cavities assumed along the flow path feeding the blocked assembly.
Peak void fraction at the exit of the active fuel is approximately 0.50.
If CHF is arbitrarily assumed to occur anyway, the clad temperature predicted by the post dry-out heat transfer correlation is less than 500*F.
Since, it has been recommended that fuel be allowed to decay for an. additional 100 days (i.e., total decay time of 180 days) prior to it being stored in the poo' in the racks near the cinder block wall, these results are conservative.
For Consolidated Pin Bundles, CHF is predicted if the " critical void fraction" is taken at the conservative value of 0.8, but is not if the true critical void fraction is above 0.85.
In any case, the predicted clad temperature for bundles with 180 days decay time using the post dry-out heat transfer correlation is also less than 500*F due to the extremely low heat flux at this relatively long decay time.
Conclusion Coolability of spent fuel stored in cavities which have the potential of being blocked if the cinder block wall adjacent to the spent fuel pool fails, is adequate provided the minimum decay time of the fuel stored within these cavities is greater than or equal to 180 days.
7997L-LHO
M AINE YANKEE ATOMIC POWER COMPANY
, NRC Reauest 2 In your submittals dated May 26, 1983 and December 11, 1985, you stated that the spent fuel racks will adequately protect the fuel assemblies from potential damage caused by falling blocks based on your evaluation.
Provide the results of your analytical calculations which show that no unacceptable fuel damage will occur. Your response should include:
NRC Reauest 2a The fuel bundle drop analysis by GCA Corporation which qualified the spent fuel racks.
Maine Yankee Resoonse 2a The fuel bundle drop analysis by GCA Corporation is summarized in Attachment B.
NRC Reauest 2b The calculation sheets for the impact on the spent fuel racks due to the failed block wall.
Maine Yankee Response 2b The south wall of the Fuel Building is constructed of eight-inch hollow block and extends from Elevation 44'-6" to Elevation 71'-6".
The wall is supported laterally by the Fuel Building's steel superstructure at Elevations 58'-6" and 71'-6", effectively creating two vertical spans of 14 feet and 13 feet (Reference Maine Yankee Drawing Nos. 11550-FA-128-5 and 1150-FS-26A-5).
The worst case collapse analysis assumed a simple mid-span plate failure of the upper 13 foot high panel.
This collapse assumed complete severing of the masonry wall along a horizontal plane at Elevation 64'-6", with the subsequent collapse, edge first, of the top seven feet of the wall into the spent fuel pool.
(Note: The vertical M, N, and P-line columns, along with the existing wind bracing, and crane support steel will severely limit the length of the assumed seven foot high wall-plate that can reach the pool. However, the collapse was analyzed on a per foot of length basis, thus covering the possibility of the entire length of wall avoiding the intervening structural steel, and impacting the pool.) The center of gravity of the collapsing wall section was taken as Elevation 68'-3" and its weight was calculated as 385 pounds per foot of wall length.
The potential energy of the falling masonry was then equated to kinetic energy and an impact velocity of 41.89 fps was calculated at the spent fuel pool water level (Elevation 41'-0")
(Note:
This velocity increased to 44 fps when the wall section was completely submerged in the pool).
By the use of partial differential equations, the effects of buoyant and drag forces were incorporated into the kinetic energy analysis and a final impact velocity of 25.29 fps was calculated at the top of the spent fuel racks (Elevation 22'-2.75").
This velocity corresponds to a kinetic energy at impact of 45,888 inch-pounds per foot of wall section.
7997L-LM0
MAINE YANKEE ATOMIC POWER COMPAP$Y
. NRC Reauest 2c The failure mode of the block wall used in Item b above and the justification for that mode.
Maine Yankee Resoonse 2c As discussed in the previous response (Item 2b), a simple mid-span plate failure of the upper 13 foot high panel was chosen as the worst case collapse. By ignoring:
(1) the energy dissipation of the masonry wall / water impact, (2) the shielding provided by the existing building steel, which would limit the size of the masonry debris than can reach the pool, and (3) the breaking up of the relatively weak unreinforced masonry upon impact with the spent fuel pool water and/or racks, this analysis provides a conservative calculation of the impact energy.
NRC Reauest 2d A numerical comparison of results between Items a and b above.
Maine Yankee Response 2d The Spent Fuel Building South Hall Collapse Analysis yielded a value of 45,888 inch-pounds (per foot of wall section) for the impact energy. This value is two percent greater than the 45,000 inch-pound energy analyzed by GCA Corporation par Systems in their " Equivalent Fuel Impact Static Load Analysis Design Report". Since energy dissipation and breakup of the masonry wall were conservatively ignored as indicated in part 2c of this response, and since the impact energies calculated were comparable, the analysis concluded that the masonry wall collapse was acceptably bounded by the par fuel bundle drop. The par systems analysis found that an 18-inch drop of a 2500 pound fuel bundle would result in only localized deformation of the upper portion of the storage racks. The maximum rack deformation calculated was less than 1 inch.
7997L-LHO
MAINE YANKEE ATOMIC POWER COMPAP$Y ATTACHMENT B 4
1 Sumary of GCA Corporation par Systems Calculation DR 9016-4
" Equivalent Fuel Impact Static Load Analysis Design Report" 1
7997L-LM0 c.-----n. - -. -
DR-9016-4
-v R;v. O 1.0
SUMMARY
OF RESULTS The results of fuel impact loads and fuel penetration for the
(
spent fuel modules are-summarized in Table 1-1 as follows :
TABLE l-1 LOAD COMB.4 IMPACT CONDITION IMPACT' LOAD PENETRATION CONDITION (Kips)
(Inches) 1 18" Drop on Middle of Rack a) Middle of Can 50.26
.928 (Max) e-m
)c:]
/
u-s b) Intersection of Not Four Cans _
131.00 Calculated e ---
m (Max.)
I r,
(
[
L J ww 2
18" Drop on Corner of Rack r1 69.0
.757 E-.J.
)
sA 3
Full Length Drop thru Complete a can impacting middle 90.0 She ar-Ou t of bottom grid of Cruci-form (Ref.
Appendix A for calculations) l l
t i
1-1 r-
V I l.'
DR-9016-4 Rev.
O 2.0 MEHTOD OF ANALYSIS FOR DETERMINING EQUIVALENT FUEL IMPACT STATIC LOADS AND PENETRATIONS
(
The following methodology was used in determining the equivalent static loads and amount of penetration for the following drop conditions: (Ref. Para. 5.1 of DC-9016-1).
1)
An 18" fuel drop in the middle of the top grid.
2)
An 18" fuel drop on the corner of the top grid.
3)
A fuel drop 18" above the module, falling the full length through the cavity and impacting on the bottom grid.
For the first two drop conditions, the impacting energy was conservatively determined by excluding energy losses (i.e., fluid drag and impact), resulting in a final impact energy (E) as
(
determined from the following:
E= Wh Where-W= Buoyant weight of fuel bundle = 2500 lbs.
h= Drop height = 18" This impact energy (E) is conserved by the elastic and inelastic deformation of the module impact area (i.e.-upper portion of can).
For condition 3, shear-out of the grid cruciform was assumed and analyzed accordingly.
(
2-1
DR-9016-4 R;v. 0 2.1 CONDITIONS 1 AND 2 In condition 1 it is assumed that failure in the impact g
area is localized with only the upper portion of the cavity failing.
For this scenario the critical buckling load is calculated from the critical buckling stress of a plate, the size of the upper portion of the can, with all edges simply supported and a uniform compression load applied.
The computer results from applying the calculated buckling load to the static model verified that the stress was localized to the upper portion of the can (die-formed area).
t In condition 2 it is assumed that the two outer walls of the corner can will deform along the length of the can, with
(-
minimal failure to the upper portion of the can. This is considered local failure with only the outside walls of the can failing.
This assumption was verified by the computer results from applying the corner can buckling load to the detailed static model.
The equation used to calculate the critical stresses for both conditions 1 and 2 is given below.
v ocrit = k (
2)
I b)
Ref:
Roark and Young, l'- v Formulas for Stress and Strain, 5 th Ed. p. 5 50 (cond. 1 case la, cond. 2 case Ic).
0 2-2
DR-9016-4
.R;v. O where -
factor based on ratio f k
=
(
plate length a
=
b-=
plate width on which compressive force acts plate thickness t
=
Poisson's Ratio v
=
Modulus of Elasticity E
=
The critical stress and impact loads for conditions 1 and 2 are determined in Appendix A.
The elastic deflection (6 )
at point 1 (see Figure 1-1 and 7
2-1 of App. A) for conditions 1 and 2 were calculated from the rack spring rate and critical load as follows:
Pcrit 1
k
(
where -
Pcrit
= critical load for buckling k
= spring rate (see Appendix A)
The spring rates (K) at each point of impact for conditions 1 and 2 were determined by computer analysis.
This consisted of applying a 10,000 lb. load at the impact point.
With this 10 kip load an'd the resulting deflection j
the spring rate was determined.
To determine the maximum penetration, deflection total impact energy-(6 ) at point 2 2
(E) is equated to the area under the force - deflection curve (Figure 1-1 and 2-1 of App. A).
The penetration values are determined in Appendix A.
2-3
, s i g ij DR-9016-4 s
Rev. O The maximum impact load for condition 1 when the fuel assembly hits at the i_ntersection of four cans is greater than the impact load for the middle of the can impact.
This is due to an increase in impact area.
In condition 2 when the fuel ar.sembly impacts the i
corner of the grid there is penetration in the form of buckling of the outer two sides of the can from the impact energy being absorbed.
2.2 CONDITION 3 For condition 3,where the fuel assembly drops the full length of the cavity, there is no penetration in the rack; the fuel assembly shears out the cruciform at the bottom of the rack.
The load required to shear
(
out the cruciform is the impact load and is determined from the statically applied load to the middle of the bottom grid.
See Appendix A for calculations.
1 2-4 s
. -. ~ -. _ _ _, _.,. - -,. _ - - _ _ _ - _ _ -. _ _ - - - - _. -.. _.
DR-9016-4
- J d,v, g
Rev. O i
l 4
APPENDIX A IMPACT LOADS AND PENETRATION CALCULATIONS 9
e W-
. ev..#. _ raya.ll.11-81:ussacT..
cwaav no....l... or.a C HKD.
Y.
.. C AT L11:-12.- S l
......-..~..~ - - -......
J03 No...DR: 9.01.6-A Rev. O
- -.~.-..---
1 4
.- :, - ~
1.0 LOAD CONDITION 1 I
Ref:
DC-9016-1 Sec. 5".1
~
18" Fuel drop on middle of rack 1.1 CASE A - MAXIMUM PENETRATION Maximum penetration occurs when the fuel hits in the middle of the side of the can as shown below:
(this assumes 2 can sides effective)
..!U p-Impact Area
l L_ _J G
J
(
Local buckling stress on top of can - (Ref: Roark, 5th Edition, p. 550 case la.)
Ih) acrit =
2 (1-v )
cA hb
= 10.25
. 3 9 =+ k=7. 31. -
=
I.105)2 7.31 (27700) 2 10.25 23.35hsi acrit
=
=
1.3 Load required to buckle 2 plates, Pcrit = 2 (10.25).105 (23.35) = M kips
(
cwo, gk. rATr.Ah.11:-81sueJ:CT...
ey...#A SHIET NO..f.... 07.-- 9
,,,g4ya 11-12-81 Joe No..J.R.-1016. L B.*.V
- O
-.. ~..
- )
.. a,e a.,
Penetration force deflection diagram k
50.26 Force FIGURP. 1-1 l
Deflection I
d 6
i 2
The rack will react elastically up to 50.26 kips (buckling load) it will then become inelastic and buckle until the area under the curve (Fig. 1-1) 1 equals the impact energy.
Impact enorgy
(.
E
= 2500 (18) = 45000 in-lb. = 45 in-K Elastic spring rate of rack, C,
V' c.
10 K, =.013153 760.28 kip /in.
=
6 1 b8
.0661" (elastic deflection)
=
=
Equate the impact energy to the area under the curve -
0.26 b
+ (6 45 =
2~0) 50.26 1
50.26 (.0661)
=
+
(6
.0661) 50.'26 2
1.661
+ 50.266
- 3.3222
=
2 45 - 1.661 + 3.3222 62=
50.26
.928"
=
The maximum penetration will he.928" Y
--a
cwn:,eY. N. e4Ts.4.4.:.4.k-8 4ues:cr.........
eY.. Pf cwear u3.. f
.or 9
~
11-1 ~01
..CATE JOS NO,..DR.,1016 A na. a.,...........
., 2 v t 1.2 CASE B - MAXIMUM LOAD I
The usedrW#d maxi' mum' lo'$d %ccurs Nife'n'th's 'ftYei ~ asdeitibl9 hits at the intersection of 4 cans (see below) wi'th:4' cah' hides sharing
.the load,.
The buckling Stres's tremains'the seme'bu't' the area of 1.mpac(4. doubles. so the Pcrit'=. 50. 25'-(2) 100.52 kips.
=
p e
E$** N IN s
Can F.lement No.
l ls i
i
- s
/
ai I
@I v.
e t
NM~&
(
Re5ulting computer stresses on upper elements from 100.52 kips load applied at intersection of 4 cans:
21.3 o
=
y 18. D
=
2 I = 108.07 ksi 14.2 "3
=
a 108'.1 si ave =
=
y.
=
20.6 6
4 15.3 og
=
og
=
17.7 of
=
.3 Uncoupled o
~
8 100.52 5.2 Plates ef fective
=
.105(10.25)18.0 i
l l
nased on exact distribution obtained from the computer analysis it is determined that 5.2 can sides share the load, therefore the modified load is -
i P
modified = 23.35 (.105) 10.25 (5.7.)
cr k
= 130.6 k
(Use 131 impact load)
w a. ~ v.
.s....v
. cv...u..
cm.unu a wuracT..
JC] NO....O.R,,9 0,1,6,,,,4 cuo.
. cats..l.1.l.2.-81
.R,Ey.1 0,,,,,,,,,,,
l.,
.,...s 2.0 LOAD CONDITION 2 Ref:
DC-9016-1 Sec. 5.1 18" fuel dr.op on corner of rack Maximum penetration arid
- maximum
- impact load occhrs for the same corner drop case as shown below:
(assume two can sides share the load)
Impact Area i
~
l i
L__._J
}
Local buckling of can outer sidewalls -
Ref:
Roark, 5th Edition, p. 550 case 1C ccrit = KE I )2 2
1-v 165 18.86 M k = 5.73 a/b = 8.75
=
5.73(27700)
I.120)2 32.80 ksi ccrit =
2 G
1.3 Pcrit = (2) 32.80 (.120) 8.75 = 68.88 kips
CATt.A.4.-A A-81susarcT c HIET NO... 2....T F..Y..._
or..
......t ATs. ll 81 Joe so,,,,, DR-9 016 - 4 cuno.
Rev. g_,,,,,,,
Penetration force-deflection diagram
\\
k_
68.88 Force,
FIGtIRE 2-1 Deflection The rack will react elastically up to 68.88 kips (buckling load) and will then become inelastic and buckle until the area under the curve equals the impact energy.
Impact energy -
E = 2500 (18) = 45000 in-lb.
= 45 in-K
(
Elastic spring rate of rack -
i 10 330.96 kip /in.
k=
=
.0302155
. 081" y=
3h 6
d
=
Equate the impact energy to the area under the curve -
i
.886 45 =
1
+
(6
~0) 68.88 2
1 2
45 = 6 8.88.2081)
.2081) 68.88 gg 2
14.3339
= 7.1670 + 68.886 2
6 45 - 7.1670 + 14.3339
.757"
=
2
=
68.88
(
There is no local penetration but the two outer
(
sides of the can will buckle.757".
. ev..
.. avs.l.I:/bf/ c ur.ccr..
c w a av N a.. 6.....o r. 9.._
C Ar E/d878 --..----~~~......
J03 NO..9.E".I016,-4 ewwo.
-... ~
..Rev g
- ~....
- < w.1 v.
t
- i...,.
~
Resulting computer stresses on elements 9 and 12, below t
k upper transition, from 69 load applied at corner of rack:
12.-
~
Element r-9 i
Below Transition g
Area
/
~
35.14 ksi a
=
9 35.13 ksi a
=
"V" 012 =
35.12 kai c,y, = 3 5.13 ksi) " crit = 3 2. 8 ksi
(
69.0 load is sufficient in causing the elements 9 and 12 to buckle under the corner drop.
i l
l t
l l
I FI G-u rt. E 1.
i I
i i
13 ut. s<
Pbot. votJonti l
Coo (4evr F4oW
\\
\\
& j u g ;y @ l R. 7 : f,; ; g y -
Oz it,s f(,e L,
p j
/.! s
>T)r ourg,gy
.yt,m^
l-
.,i
,. \\
h.' \\, -l/ \\ /
ss\\
r-
- fA..
?$.
'.?, L $ N I.. {~,
QJT. !
' #0 ^' # 5"* $
W j
f Y,~.ti; ikf J.~
i..'
-. \\~...
s
's -
\\ _, g.
/\\
N.
/ \\, \\ / \\ / / \\,/.s,.' i. -.-
i./. i.- f k.u. --.t '.
g i
\\/\\/\\./\\ \\/ i'.\\
s.
s u.--
- I I
I r' i/
'. I
~~i, i /\\. @
,/
s ss s/ \\/ Ls l.-
i
/\\
~
[
../,\\.,..'..i
/\\,,
./.\\;,
T g
./-
./\\
\\
/-
7 fg
'\\ /\\
/
4
\\
lNTEtt CA
1
. Y
".i
.f y/ \\i
'\\ /' i
/
1 j +' g l-;
. '\\ /'
,/ \\i
+
+
V K R j 3.:.
g i
c.
\\
~
poa>ut g.),5
_._ _ I c
).i N/ V---
i/
/
. ' _ ' - _1-
/ sI
)- '-
.l
'.,/.r 1..
j
+b--
'g 6
,/ -, \\.
- /\\
f\\
+
.e
'\\
/,ji
.p
)
.__\\_.'_\\ /\\
- 3.--y,
- . _\\.
7 4
- ?r /; X.
i \\
/ _7t\\
\\/
/
./
/
i
.i i
s g, e 1
g t..
~~
s
/.
~/:../ 7? \\ J f
3
~le f W g
T =
g v...v
....re
'. '/
V.-
w
/\\
\\ / \\ /]. /\\ ]\\ - \\
'/.,'\\,
\\ / \\ /. /\\. \\./ \\
./\\
G,,4f,qss pgag YN/toisgh
'I
__'hz;.l H 1' \\- 1 5Ei . l 'S_
~
i l X V
\\/ I'
[:
d V
s M.
\\
/
/\\
\\ /\\ /
unecock cA v'r Tid S
< > - ~
'. \\
.J' ZAREstvENt/G G. C'A u sTIL"S ALOMG-,,,. \\:
_j'_'
N' A f
9
\\
- ^'TA'^O 6.f/M NT j
f\\
\\ / m r =\\
\\ /. r i..
.-\\..,
\\:.
!... i '.:
. ?
/
k I'(
,N M N 2,--h N i '.
}
i1
)
~
er
~
4
- O.
\\..
.7. Pt.oD f%97ti FEEOsAdr ' GLock60 <kusr yQ A.r5 ftuy ThiaOogH i
. a-.
=
. j \\
!\\ l\\
\\, '.
.-l M
-(
f I (3eocg Eo cA v ET n
. : \\,
y
. =
7 l'
./
l-s.
.i[ ]
];
',, '. (
)< \\/
e
~5
/
..s
.f,
.s
+
- - =
/
.\\
es
/\\--
,\\ n.
/
/
<\\
\\
/3 f.\\,
~
}
f
./ - -.
N. g. K
. y
- s. e f.,
3
.;g I
j\\ /
f se N.,,.-,\\..
.\\.,,.-
..\\ /'
/\\
ei /..,/ \\
\\
/. /,\\.
N s
s,
\\/
e
- s.,
s s.
j
.g.
I gc\\/
<\\
/\\ /
/\\
/\\ /
2
.. /.
' G.
h "l
l/, h*.l
- ]'
.I,_;,;,_ ]f \\/.l4 3<
/.\\l' 3~
g/ i!
I" g
f V
s
./s /./\\ n f n
./
y.
<\\
,\\
.\\
\\
\\ f-s N
g s
1-c,7 r/
s t,
i s.
F1 f-l 11 1
F1 h
3
)
l p
g3 i
- 9. ~
.p. '
- p. '
T' s.
Sa, y,
Su i
I I:/ / / /\\/ ///./// /' / ////////////
i
~ fok F&oOR
~
(
\\
O h
h l
I I
l e
[
i.
1 i
g... j._..
i j-i i
i q-..
r-i
=
-~ -~
i i l l 1 i
i i
li I
I
(
i I
--]-
l i
l l
, q-- :-- -
i l
I!
i '
I I
i I
i i
I D; i
l I I
I i
i !
o l
l l
l ! i
.I i
- I G
I-j-
j j
l
- 2..
i i !
D f
k I.
l I
I
! i-1 I_W-
'i i
- r l
l
^l i:
i I
2' l l
O I6 I l
'I J.
p
.l-
.l-l I
~!
T ik I
. u.
i-i g
i' w
. i,,e m
t a
i !
vr i
.i
.I f
i 7
m
- l. -
el fl !
'I I:
A l-5.=.
=..+
.. if !
l F.s -
_l
.:- y
=
(,
k =+_
/
l C
@l
.:,Q i
. ?f 1
i
- . s;.
.+ ; m -4
/
F
- l:
rJ[
+
l' n]
,y 3
0 2
=
v'
- I l
.. 4
.. =
.; w
. f-.+
- 4 1
.r.
=
- 7
- p
. i.
w=
., N:
=
m z p c
..u c
.i^
l "l
h
[
! I
-A.
p.
l I
-l 6
a]. g:
aj v
2-l 1
i z
y' i
I k
.{
l !
I 4!
.i i
i i
i i
23
..(v.
i i
i.
i 11 -
I 7!
d
}
l 0 l'_Yo i "g
.I i
i l_
{
l
%3 J. __.
i i
i m.-
i o
I i
l l l
.n.
j l l l
i 4,
i_ t i
(
9 9
e-t p,
h
[
g es l
1 4
i
~'
O 9
9 9
e m
t 5 4 I
O,
=.
E
~
.s-f.
r 3y W
.%* o.
A eo y..
g
/ 6
. O 3
.c l
0 o
/
j m
/j a-
~
It
. =.
0
/
=n l
'Y S
O]
o g
/
_ o
.o i
01
. O n
, e o.
tg
/g
- b.
~
Pp. O
/yl 3
.e,-
o
.~
/
c o-I*
d
O O
/
- v 5
I xw a
/
11 3
i 8
0"o v:
/
g.
iYi i
i i
e i
i o
i o
a o
ar i
..n.}
s l
, e l
- -