ML20041B125
| ML20041B125 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 02/16/1982 |
| From: | Withers B PORTLAND GENERAL ELECTRIC CO. |
| To: | Clark R Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8202230297 | |
| Download: ML20041B125 (12) | |
Text
,
sy t im f 9 f
mumme Portland General ElectricCompany
- q y-.n,,.s.~g
.m.
FEB 2 2 Iggga. :
an D.vom Vm F Eme.t h
h Q
4 tu February 16, 1982 Trojan Nuclear Plant Docket 50-344 License NPF-1 Director of Nuclear Reactor Redulation Attn:
Mr. Robert A. Clark, Chief Operating Reactors Branch No. 3 Division of Licensing U. S. Nuclear Regulatory Commission Washington, DC 20555
Dear Mr. Clark:
Overpressure Mitigation System
~.
This letter is in response to the requests contained in your letter of January 8,1982 on the Trojan Nuclear Plant's overpressure mitiga-tion system. The information that you have requested is contained in the attachment to this letter. We hope that this information will result in a timely and favorable review of our proposed Technical Specification changes (LCA 52, dated May 1, 1976).
Should you have my questions concerning these responses, please contact us.
i Sincerely, Bart D. Withers Vice President Nuclear i
Attachments l
l c:
ric. Lynn Frank, Director State of Oregon Department of Energy
(
8202230297 820216
/
p'9l l
PDR ADOCK 05000344 l
0 P
ppg S:
(1 121 S W Samon Gmt Prand. O egen 972C).2 l
~
R. A. Clerk Trojan Nuclsar Plant February 16, 1982 Docket 50-344 Attachment A License NPF-1 Response to Request for Information Dated January 8, 1982 The following are the responses of the Portland General Electric Company.
to the NRC's request for additional information submitted by letter dated January 8, 1982. The NRC requests.are restated in their entirety, followed by the PGE response.
1.
Branch Technical Position RSB 5-2 requires the Overpressurization Mitigation System (OMS) to meet single active failure analysis when the initiating cause of the event is not considered as the single active failure.
In addition to your analysis submitted in your July 21, 1977 letter, consider as the initiating event the failure of a DC power bus which results in the isolation of letdown flow and also fails closed one of the pressurizer PORVs. A postu-lated single failure (f ailing closed) of the other PORV would fail mitigating systems for this event. Discuss your plant's provisions to mitigate such a transient.
' Response There are four basic Reactor Coolant System (RCS) conditions which must be evaluated for the event postulated above, where the OMS is inoperable coincident with a pressure transient. These four states are:
a.
RCS cooldown from hot standby to cold shutdown prior to placing the Residual Heat Removal (RHR) System into service, b.
RCS cooldown from hot shutdown to cold shutdown with the RHR System in service, c.
RCS heatup from cold shutdown to hot shutdown with the RHR System in service, and d.
RCS heatup from hot shutdown to hot standby after the RHR System is removed from service.
The postulated scenerio is applied to each of these situations. A discussion of Trojan's ability to mitigate the pressure transient follows.
In the process of maneuvering the Plant from a hot standby condition -
to a cold ahutdown condition, the RCS temperature and pressure are maintained in the acceptable region of the limit curve found in Trojan Technical Specifications (Figure 3.4.3).
Pressurizer' level is maintained by the pressurizer level control system at no-load level
(#30 percent). This provides a vapor space of approximately 1260 3
ft. Should the postulated event occur prior to placing the RHR System in service, this vapor space provides a buffer against over-pressurization of the RCS.
R. A. Clark February 16,.1982 Attachment A 2
For a loss of letdown and a maximum charging flow of 130 gpm, and assuming no operator action for 10 minutes after receiving a variety of alarms associated with the loss of a de bus, the vapor space would 3
be reduced by approximately 175 f t. This would result in a pressure increase of roughly 18 percent. A pressure rise of this magnitude would not pose a serious challenge to the Appendix G limits for Trojan.
In addition to the alarms associated with a loss of a de bus, the operator would also receive a +5 percent level deviation alarm for the pressurizer. Sufficient time would still remain for the operator to take action in terminating this transient before a serious over-pressure situation exists.
Since this pressurizer level is maintained until the RHR System is placed in service, a vapor space of sufficient magnitude to mitigate a potential overpressure situation exists until the RHR System can be used to relieve ' pressure transients.
RCC<450 psig.
The RHR System is placed in service at T RCS RHR suction valves MD-8701.and M0-8702 from the RCS (Loop 4) are opened. The RHR is then available to mitigate a solid plant pressure transient in the following ways:
a.
A normal RHR flow path would still be available to the Chemical and Volume Control System (CVCS) downstream of the closed letdown isolation valves CV-8152 and CV-8149A, B and C.
Flow would be directed either to the volume control tank or one of the holdup tanks. This would provide the operator with sufficient time to diagnose the source of the pressure transient and to carry out remedial actions
- and, b.
The RHR is also designed with safety valve PSV-8708 which has a setpoint of 450 psig and a relief capacity of 900 gpm (discharges to the pressurizer relief tank). In addition, three other safety valves are available on the RHR System, PSV-8856A, PSV-8856B and PSV-8709. Each of.these has the capacity to relieve 20 gpm to the pressurizer relief tank at a setpoint of 600 psig.
During Plant heatup from cold shutdown to hot shutdown, the RHR System is also available to mitigate pressure transients as described above. The RHR System is removed from service after a bubble has been drawn-in the pressurizer, and presserizer level has been stabi-lized at the no-load level of 30 percent, and the pressurizer level control system is in AUTO.
R.!- A. [ Clark
. February 16, 1982 Attachment A 3
3 Because _ of the existence of a large vapor space. (1260 f t ) in the pressurizer prior to removing the RHR. System from service,.
there is always adequate protection against'an overpressurization event resulting from a loss of letdown with maximum ' charging flow and OMS unavaileble. -
During the telephone conversation held between PGE and the NRC on January 27 of ? this year, a point of concern was raised by.the NRC over;the ability of the vapor space to mitigate an overpressure
~
transient. It was the contention of'the NRC that since Trojan's Technical Specification 3.4.4-allows plant operation in Modes 1 through 3, with 92. percent pressurizer level, that PGE must analyze pressure mitigation based on an 8' percent pressurizer vapor volume.
PGE does not t,elieve.this to be warranted for the following reasons:
a.
Except.during abnormal transient conditions, in Modes 1, 2 and 3, pressurizer level is maintained between 30 percent -
(no-load level) and 61.5 ~ percent (full load).
b.
During cooldown,' the pressurizer level control system is in AUTO and maintaining a pressurizer level of -30 percent.
In addition, because of-shrinkage - of RCS inventory the pressurizer level tends to fall, not rise.
c.-
During heatup,_the Plant procedure specifically directs the operator to draw a. bubble in the pressurizer and stabilize pressurizer level at a no-load level of 30 percent prior to removing the RHR1 System from the RCS.
d.
At an indicated 92 percent pressurizer level, greater than 8 percent of the pressurizer is vapor space. The percentage 3
is closer to 17 percent or 306 ft.
In conclusion, it is PGE's contention that during heatup or cooldown there are two mechanisms (not necessarily concurrent) which are available to mitigate a pressure transient when the OMS is unavailable.
These are either:
a.
A large vapor space in the pressurizer which provides the operator with sufficient time (>10 minutes) in which to evaluate and to terminate the event before a serious over-pressure condition exists, or b.
The availability of the RHR System and its overpressure relief capacity.
2.
The branch position requires an alarm to alert the operator that a pressure transient is occurring. You state that a high pressure alarm is received in the control room'on the plant 250 computer.
B.-A.
Clerk February 16, 1982 Attachment A 4
a.
How is the alarm annunciated?
b.
Must the alarm be acknowledged to silence 'the annunciation?
c.
What visual indication associated with the alarm condition is available?
d.
When is the alarm enabled when you are cooling the plant down from hot standby to cold. shutdown?
e.
How of ten are the setpoints checked and alarm action tested?
f.
Is there a backup alarm available during high vulnerability periods (low temperature and water solid RCS) when the '
plant 250 computer is removed from service for maintenance (you stated that maintenance is normally performed during low temperature operation)?
Response
a.
During solid plant conditions, several high pressure alarms are received in the Trojan control room on the P-250 computer.
The three alarms are from PT-400 set at 425 psig, and "A" and "B" RRR' pump saction pressures..Each of these alarms is annunciated by a short, sharp ring..While no operator action is required-to secure the audio portion of the computer alarm, a pressure transient will result in several of the setpoints discussed above being reached, and their alarms annunciated.
Each of the computer alarms results in a visual indication also. This visual annunciation is a flashing alarm message on the CRT screen. It requires operator acknowledgement' to terminate.
b.
See Part (a) above.
c.
An RCS pressure transient may result in the actuation of the following normal control room annunciators:
- 1). CAUTION - APPROACHING OVERPRESSURIZATION
- 2) OVERPRESSURE PCV-455A TO OPEN
- 3) OVERPRESSURE PCV-456 TO OPEN.
These are found on control room panel K-13.
1 Each of these visual annunciators require operator action to - terminate the alarm.
d.
The OMS is enabled during cooldown from hot standby to cold shutdown. Two pressure alarms annunciate to alert the f
f
= - -,
.-,.n-,
--,,y
,---e.
R. A Clark February 16, 1982 Attachment A 5
operator that the RCS pressure has decreased sufficiently to allow the low pressure mode to be instituted without inadvertent actuation of ;either PORV.. The setpoint for these alarms is 375 psig. These annunciators-will remain lighted until the RCS pressure is increased above the setpoint of the alarm bistable.
e.
A functional test of the OMS is performed prior to placing-the Plant in a water-solid. condition. This functional test checks the setpoints of the system and corresponding alarms.
f.
The three alarms discussed in Part (c)
- 1) CAUTION - APPROACHING OVERPRESSURIZATION,
- 2) OVERPRESSURE PCV-455A 10 OPEN, and
- 3) _0VERPRESSURE PVC-456 TO OPEN i
are not computer alarms and would remain ~ unaffected by 4
a P-250 computer outage.
3.
The branch position requires an alarm ta alert the operator to enable the OMS at the correct Plant condition during cooldown. You rely on l
a pressure' actuated alarm to perform this function. How do you ensure that the Reactor Coolant System temperature does not fall below-the allowable temperature corresponding to the above alarm pressure set-point, thus violating Appendix G limits?
Response
i
~
Examination of the RCS pressure-temperature limit curve indicates that below RCS pressures of 500 psig, the acceptable region'is almost independent of temperature.
Furthermore, for the Trojan Nuclear Plant, the maximum RTNDT is 77'F after 15 EFPY. Since component cooling water to the inlet of the RHR I
heat exchanger is limited by operating procedure' to a minimum tempera-ture of 70*F, it is very unlikely that RCS temperature could approach even the conservative RTNDT value of 77'F, especially if decay heat and pump heat are considered.
1 As discussed in our response to Question 1, prior to enabling the OMS, a pressurizer vapor space of sufficient volume is available to miti-gate the pressure transient and provide the operator with sufficient
~
i' time to evaluate and terminate the event prior to reaching an over-pressure condition in the RCS.
If the RRR System is in service, which is likely, it has the capability.of relieving 900 gpm at 450 psig to mitigate a pressure transient prior to enabling the OMS.
Based upon the above considerations, violation of Appendix G limits is highly unlikely.
i
. ~.
..m_
. _.,. _ -,. _, -.... _ _ _~ ~.-.,._._ _-..,. _ _,
R. A. Clark Fabruary 16, 1982 Attachment A 6
4.
Deleted from NRC request letter.
5.
The Branch Technical Position requires the OMS to be functional through an Operating Basis Earthquake (OBE). You stated that the air accumulators met Seismic Category I requirements but failed to mention the rest of the system.
a.
Identify all system components that are not designed to function during an OBE, and give the basis for these components not meeting this requirement.
b.
Analyze the situation if the components that you identify in "a" above do not function.
c.
Are the PORV operators qualified to operate through an OBE?
Response
The OMS installed in the Trojan Nuclear Plant is Seismic Category I with the exception of two components. The accumulators, air piping to the valve operators, solenoid valves and associated circuitry are all Seismic Category I.
The PORV operators and their associated air regulators have not been specifically qualified for operation through an OBE. The PORVs are designed to withstand seismic loadings equiva-lent to 3.0 g in the horizontal direction and 2.0 g in the vertical direction. The PORV operators were not originally procured nor analyzed as Seismic I components. Since at the time the OMS was implemented, there was no clear seismic requirement tor this system, the cost of qualifying the operator was not warranted.
PGE is currently investigating the requirements for qualification of these valves and their cir reg '.stors.
If the Trojan OMS is inoperable due to an OBE, a similar situation to that postulated in Question 1 exists and PGE's response is similar.
6.
Additional information is requested on the following items concerning RCS and RHR relief valves and safety valves.
a.
What is the basis for the backpressure assumed for the PORVs in your transient analysis?
b.
What is the relief capacity sensitivity to the PORV backpressure?
c.
Is the relief capacity of RHR safety valve PSV-8708 sufficient to mitigate the overpressure transients that were analyzed for the OMS?
d.
Does the valve position indication of the PORVs and the PORV block valves indicate a control circuit condition or the actual position of the valves?
e.
Do the PORVs and RRR safety valves have control room temperature indication on the piping downsteam of the valves?
~
R.-A. Clark Fabruary 16, 1982 Attachm:nt A 7
Response
a.
A backpressure of 10 psig was used in the Trojan pressure transient analysis. ihis.value is based upon a nitrogen cover gas pressure of 3.psig in the pressurizer relief tank. While PORV cycling will cause a subsequent increase in pressurizer relief tank pressure, the value chosen for the analysis allows a margin for this backpressure increase.
Dynamic backpressure on'the relief valve was considered negligible. This is because the expected discharge flow rate from the relief valves is relatively small for the size of the discharge lines and relief tank when compared to the design flow rate from the pressurizer safety valves.
b.
In " Pressure Mitigating Systems Transient Analysis Results",
prepared by Westinghouse Corporation for the Westinghouse Owners Group, relief capacity. sensitivity to a' typical relief valve was addressed. Figure 2.2.1 of that report is included as Attachment B.
Since this study was done to envelope all Westinghouse plants, this data -is not expected to be significantly different for the Trojan PORVs.
RHR safety valve PSV-8708 is capable of relieving.900 gpm c.
at a setpoint 'of 450 psig. With this relief capacity it is capable of mitigating a pressure transient resulting from:
- 1) Isolation of letdown,
- 2) Inadvertent start of a centrifugal charging pump; (CCP) or positive displacement pump (PDF),
- 3) Inadvertent start of a safety injection pump (SIP).
While PSV-8708 is not capable of ~mttigating pressure transient resulting from the starting of two CCPs,:two SIPS, or a combination of a CCP and SIP, the Plant proce-
. dure requires operators to lock out both SIPS and a CCP-prior to going solid. This reduces the number.of potential sources of a pressure-transient. The operator is also advised against the start of an RCP during solid plant con-ditions, and warnings and precautions are included in all pertinent instructions.
d.
.The valve position indication of the PORVs and the PORV block valves at Trojan are direct positive indication, not ta' ken from the control signal to the valves.
Temperature indication is available in the Trojan control e.
room for piping downstream of the PORVs. No temperature indication is available for piping downstream of the RHR safety valves.
R. A. Clark Esb ruary 16, 1982 Attachment A 8
7 Provide the following additional information concerning Trojan operating procedures:
a.
Have appropriate caution or warning nc;es been included in all applicable procedures to make operators aware of the situations that could possibly lead to pressure transients?
b.
Provide some examples of these caution notes in Trojan procedures.
Response
Trojan operating procedures include appropriate caution and warning notes to make operators aware of situ tions that could possibly lead to pressure transients. Some examples of these precautions and warnings are:
G01-4, " Plant Shutdown from Hot Standby to Cold Shutdown" Section II:
Cooldown to Cold Shutdown Part B:
Precautions and Limitations Paragraph 24.0 "Do not isolate the RRR inlet line from the reac-tor coolant loop unless there is a steam bubble in the pressurizer or the charging pumps are stopped.
This assures there is a relief valve when the reactor coolant system is at low pressure (less than 500 psig) and solid."
Paragraph 26.0 "When solid with the reactor coolant pressure being maintained by'the low pressure letdown control valve, changes to the flow rate through the residual heat removal loop by throttling of valves or starting and stopping the residual heat removal pumps will result in changes to the reactor coolant pressure l
(ie, stopping of the residual heat removal pumps may cause an increase in the reactor coolant pressure of between 100 and 140 psig)."
and in Part C:
Instructions Paragraph 6.1 "When the reactor coolant system pressure decreases to <375 psig, unblock the pressurizer powe operated reliefs and verify isolation valves MO 8000 B&A are open.
I I
R. A. Clark F bruary 16, 1982 i
Attachmint A 9
Note:
Exercise caution during solid system operation to avoid overpressurizing the reactor coolant system."
8.
In your April 8, 1977 letter (Appendix B), you provided some informa-tion on the training that you conducted on the overpressurization incidents; provide the following additional training information:
What overpressure training have you performed since mid-19777 a.
b.
How do you ensure that a continued emphasis is placed on possible overpressurization situations in your licensing and retraining programs?
c.
How is this training and LER review documented?
Response
Training in overpressurization is not a specific subject area.
It is addressed in a variety of areas including safety limits, compliance with Appendix G limits, pressurizer level and pressure, and alarms and actuating signals. Precautions and warnings pertaining to over-pressurization, in addition to OMS operating instructions, have been an integral part of Trojan's procedures since 1977. Overpressuriza-tion is covered during the licensing and retraining classes which are held on a regular basis.
Regarding LER reviews and OMS, Trojan has implemented an Operational Assessment Review (OAR) program in accordance with NUREG 0737.
Information on situations which are identified by the OAR Program as applicable to Trojan are integrated into the training program.
Training subjects are documented through lesson plans which are under continuous review and update, and by personnel records which reflect an individual's training.
9.
What is the present status of the Trojan OMS?
Have all permanent OMS installations and modifications been a.
completed?
b.
Have all proposed temperature detectors been installed on the secondary side of the steam generator to provide primary to secondary differential temperature indication?
c.
Where does this AT indication read out?
d.
What alarms are associated with this AT system?
R. A. Clark F bruary 16, 1982 Attachment A 10
Response
The OMS installations and modifications have been completed.
Currently, the Trojan Nuclear Plant utilizes temporary thermocouples-attached to the shells of "B" and "C" steam generators at approxi-mately the tube bend height. These are used in conjunction with the cold-leg wide-range RTDs to provide the necessary temperature dif-ferential. Local readout is provided within Containment.
PG3 is currently evaluating the need for a permanent secondary side temperature instrument at Trojan.
Should an affirmative decision be made regarding this subject, the earliest installation date would be during the 1983 refueling outage. Output from this instrument would be displayed by the P-250 plant computer and would provide the operator with secondary side temperature in the control room. Using this data and data from the cold-leg wide-range RTDs, a primary-to-secondary temperature differential could be determined.
PGE does not believe that alarms should be a necessary part of the primary-to-secondary temperature system. Trojan operating instruc-tions require the operator to determine the AT across the steam generator prior to RCP start.
An alarm system would be unjustifiably costly and its benefit limited.
l GET/4jr9A5
K. A. Clark Trojan Nuclear Plant February 16, 1982 Docket 50-344 Attachment B License NPF-1 FIGURE 2.2.1 E _.-
_a O
REFERENCE RELIEF VALVE un'_=4can=
i- _._ _. _.
t= =.
C 50
=
r=_
v
_. _ - _f1_
p_.;., : n-f _2_;=
=$EEQM;_ - _,
2_ m - - - -
n. =.: - -- u = =:. = -
' _.: = = -- - -
--=2.--
-/=:=
= = =
=:===_====._,._.-.=.-:.=__
= -.E :== = : = _ =,.___.
. n_b== =......= __ _ : : = ==== c _y_- I. _ __I ___ _ _2 r
- :n
.-i-----
- =:. a r =- r r==-
u.
..12, um =
5'== =
=.f= -/.t = -- ;.-
- = =. = t '
b =_ ~- 4 = :r- -r-
- = : -- - - -. -r-
. =p =__,__:
t== = =-- :
== 1 =._ :
=
- ===-~-=
- -.--g==
BACKPRESSURE = 10 psig :- 1==:. W:L.:. : K _
.= T=: :_- rt - -.
.=,_
__t r =
... c --- :==
- =.- - - --. _ c. _1.
n =-== =x.-.
=
_ = s_:.r:-n_ = x===:=...-_
- ==-~.--
7 f-f:==== - ; -- *
=.=n
=c. : y =-
_7.._..i.______ g M-* = =.= =j===:r-
t r : : 1. = = = -==; : = ;=
2
.n p --- 9, _.
= = - -
= = _
= - - -
.......r==._
..:2__
MJg_- - f=..- -.*7.-.__---i--
I.f-
~
.-...-_p&W -._ _ -..2. ". -]~=*'.),--_-<
C ^. = 7
{,.. _.. _t=.Z_
j-T -3E EE im{==:fE = =.=.9s g===-7_j_gE-]=.s.E!= -
- _ _c_!_
__ -.,..g;
-v_,...-..
.- - f f.-
_f._.-.;.,
~
t 2:=--
-- - - - - - *
L._.
.- if ;f_ - = & - --=~_*,:m_--.
.5.
-h. ):C l'E' o ' C2f'i hh[_IE MI'.==."===:*EEtEE
==---
r
........._/.f'.___/-.=[=%. -.
. - ~ _..... ~....
=
3 '#W.- nN -" =. !=
=u==. _.
=====:==--
=.
.2
==
==
=t n =. := n; x.
, =. : =
3:
in t:=
u-/_./: =_/-.=
- .g =. :t..
.!L - =n.. u.=:. c :
-r = =- - - = = = = = = = = =, =
- .v = =: c = t =- = : =._..__
=s==___
== r=
-t-
- s=i=p--+l___
-;=
= $^ --
=q -- =yx-f=q n%.r.-- =:
=-
- - ' - - = -
- _i. _
.---: =
..=
i%m. :=_ =.,_1s.:. =_ = E. _=iE.. E_l.-f,W_h.gf._+-_- +_ -
=.~-t- _. _ _ _ = _ _. =._ t.__E E__ H..E.
-' =
-= :t
=
-_y -- =/=f=/.=a
.-.._t= =r=
_ a- -
- - +
m
- m., =v= =-
f =d p
n==r =__ w=
,
;:
W
== dei dE ME%ggggg@ BACKPRESSURE = 50 psig ga.=JE. ggi
_ 8'. "=i.=.._ &$/ fiE["4Ei=.3 EEE-EE!E =E EEiI=: =E EdM
-= r. --
,- ___,.=====
i_
2]/ y /.
-s p:...=}==
- ==
_a_-=2===:==
===,
== t= :
== 2 3 2_
rs=
a===.. -
m :... _==-..=[
=
1==1=
===t.-
i=t==:1===- - -
:1====--
.f:r:W =:;f=j -- 3--
=
. ;. =
== =_
=
.- 2=
- - - + -
_ _ _.~. _ _ = _ _
=;-
p.--
,=..__=_. _ -
. _ _==,._
7_ _,_
. p.=. - =..=.. _=.. =g. ;._ t-
=_t _=_.
w
. -. ~, =. _ -
~_ =....b..
=,.
t L_
. _ = _ = = _ _
=
= =.. =.
1.
1
- u nt=
f
~
._ ~--
f 1== tit g BACKPRESSURE = 100 psig =-
':'tW.= E;.
f=
l =
I=
n f.-i d= -a
- c:=== u n==
f
1 -
- . "n*
7 %== =. ;
L'._h. *:.-
- ~ ~
..-t-- ::-- =j
--L
- =
-1
- !.n g: _ ___..
.=
=
- 6::=.y
=
.; a-- -
l-j.
SW. =
p;.
!;p--
=: -
=___._. _ _ -
-~~
[,
'n=
. = =--_=._. -
-._===-
t==. Q=
=
1=
i
- 1. r..--
==
.=--=
-
- 2. _
i
- :. : t.=...
=.. -:.___.. n --t,-- -
t-
. _ _~.-5
_nn.
.---g..~
G--'* :-',-*:.='~5 l'
- h.~_~,*.
- A :
- =
C.k.'
- _Z$ --*-*,':*.j,. 1T. ;,-*"- * * * * *
' - - A
t
1_. _3
- -+
=
CI'.
.__==q
--* r*-'
47+T-t ! "*'
1
~[_
- F9 ~ " "'-~~"IU.0"0 I '~~ ~ ~~E'i
._3,_
- 5. h k M.3_._
Q'._ _~~T~$0.
.sv._4.,__
~~
- - =_-e--
==
=
-.-.-- a g d rie 5 5R e4xits'lg =,.__
j
. _ _ _= d==
T
==
_.,_,.~..-t.
=:-
m.
t
m....
..........._=. _=. _
__._.__~r,_...--+-.
3 **
--.+
--+.. !. =. - -. -...
q,...
+
. i l
l