B12413, Point Nuclear Power Station,Unit 1 Single-Loop Operation
| ML20237J876 | |
| Person / Time | |
|---|---|
| Site: | Millstone  |
| Issue date: | 12/31/1980 |
| From: | GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20237J824 | List: |
| References | |
| 80NED300, B12413, NEDO-24312, NUDOCS 8708260350 | |
| Download: ML20237J876 (27) | |
Text
-_ - . - _ _ _ _ - - - - - - _ . .
l l Docket No. 50-336 B12413 Attachment 2 Millstone Nuclear Power Station, Unit No.1 Single Loop Operation Analysis By General Electric l
l I
August 1987 ffjB260350e70817 p ADOCK 05000245 PDR l
)
"'Re's!' -
DECEMBE );30 i
MILLSTONE POINT NUCLEAR. POWER STATION, UNIT 1 SINGLE-LOOP OPERATION GEN ER AL h ELECTRIC
, NEDO-24312 80NED300 Class I December 1980 l
l MILLSTONE POINT NUCLEAR POWER STATION UNIT 1 SINGLE-LOOP OPERATION NUCLEAR POWER SYSTEMS DIVISION e GENERAL E LECTRIC COMPANY SAN JOSE, CALIFORNI A 95125 GEN ER AL h ELECTRIC
4 DISCLAIMER OF FIESPONSIBILITY This document was prepared by or for the General Electric Company. Neither the General Electnc Company nor any of the contributo s to this document:
A. idakes any warranty or representation, express or Implied, with respect to the accuracy, completeness, or usefulness of the information contained on this docu-ment. or that the use of any information disclosed in this document may not Ininnge privately owned rights; or B Assur~es any responsibility for Isability or damage of any kind which may result from the use of any information disclosed in this document.
. NE00-24312 CONTENTS Page
- 1. INTRODUCTION AND
SUMMARY
1-1
- 2. MCPR FUEL CLADDING INTEGRITY SAFETY LIMIT 2-1
! 2.1 Core Flow Uncertainty 2-1
, 2.2 TIP Reading Uncertainty 2-4
- 3. MCPR OPERATING LIMIT 3-1 3.1 Core Wide Transients 3-1 3.2 Rod Withdrawal Error 3-2 3.3 Operating MCPR Limit 3-4
- 4. STABILITY ANALYSIS 4-1
- 5. ACCIDENT ANALYSES 5-1 5.1 Loss-of-Coolant Analysis 5-1 5.2 One-Pump Seizure Accident 5-2
- 6. REFERENCES 6-1 iii/iv
, NEDO-24312 l
TABLES Table Titfe_ Page 5-1 Limiting MAPLHCR Reduction Factors 5-2 ILLUSTRATIONS Figure Title Page 2-1 Illustration of Single Recirculation Loop Operation Flows 2-5 3-1 Main Turbine Trip with Bypass Manual Flow Control 3-4 4-3 Decay Ratio Versus Power Curve for Two-Loop and Single-Loop Operation 4-2 5-1 Suction Break Spectrum Reflood Times 5-4 5-2 Suction Break Spectrum Uncovered Times 5-5 5-3 Suction Break Spectrum Uncovery Times 5-6 v/vi
- 1. INTRODUCTION AND
SUMMARY
The current technical specifications for the Millstone Point Nuclear Power l
Station, Unit 1, do not allow plant operation beyond a relatively short period of time if an idle recirculation loop cannot be returned to service. The Millstone 1 nuclear power plant (Technical Specification 3.6.H.2) shall not be operated for a period in excess of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> with one recirculation loop out of service.
The capability of operating at reduced power with a single recirculation loop is highly desirable, from a plant availability / outage planning standpoint, in the event maintenance of a recirculation pump or other component renders one loop inoperative. To justify single-loop operation, the safety analyses docu-mented in the Final Safety Evaluation Reports and Reference 1 were reviewed for one-pump operation. Increased uncertainties in the total core flow and TIP readings resulted in an 0.01 incremental increase in the MCPR fuel cladding integrity safety limit during single-loop operation. This 0.01 increase is also added to the MCPR operating limit. No other increase in this limit is required as core-wide transients are bounded by the rated power / flow analyses performed for each cycle, and the recirculation flow-rate dependent rod block and scram setpoint equations given in the technical specifications are adjusted for one-pump operation. The least stable power / flow condition, achieved by tripping both recirculation pumps, is not affected by one-pump operation. Under single-loop operation, the flow control must be in master manual, since control oscillations may occur in the recirculation flow control system under these conditions. Derived MAPLHCR reduction factors for single recirculation pump operation are 0.86 for both 8x8 and 8x8R fuel types.
The analyses were performed assuming the two recirculation manifolds are isolated from one another by closure of appropriate valves in the cross-tie (equalizer) line between the loops. The discharge valve in the idle recirculation loop is normally closed, but if its closure is prevented, the suction valve in the loop should be closed to prevent the partial loss of Low Pressure Coolant Injection (LPCI) through the recirculation pump into the downcomer degrading the intended LPCI performance.
1-1/1-2
NED0-24312
- 2. MCPR FUEL CLADDING INTEGRITY SAFETY LIMIT l Except for total core flow and TIP reading, the uncertainties used in the statis-l tical analysis to determine the MCPR fuel cladding integrity safety limit are not dependent on whether coolant flow is provided by one or two recirculation pumps. Uncertainties used in the two-loop operation analysis are documented in Table 5-1 of Ref erence 1 for reloads. A 6% core flow measurement uncertainty has been established for single-loop operation (compared to 2.5% for two-loop operation). As shown below, this value conservatively reflects the one standard deviation (one sigma) accuracy of the core flow measurement system documented in Reference 2. The random noise component of the TIP reading uncertainty was revised for single recirculation loop operation to reflect the operating plant test results given in Subsection 2.2 below. This revision resulted in
{
a single-loop operation process computer uncertainty of 9.1% for reload cores.
A comparable two-loop process computer uncertainty value is 8.7% for reload
! cores. The net effect of the revised core flow and TIP uncertainties is a 0.01 l
l incremental increase in the required MCPR fuel cladding integrity safety limit.
2.1 CORE FLOW UNCERTAINTY 2.1.1 Core Flow Measurement During Single-Loop Operation The jet pump core flow measurement system is calibrated to measure core flow when both sets of jet pumps are in forward flow; total core flow is the sum of the indicated loop flows. For single-loop operation, however, tne inactive loop jet pumps will be backflowing. Therefore, the measured flow in the backflowing f jet pumps must be subtracted from the measured flow in the active loop. In addition, the jet pump flow coefficient is different for reverse flow than for forward flow, and the measurement of reverse flow must be modified to account for this difference.
For single-loop operation, the total core flow is derived by the following formula:
Total Core Active Loop ""C "* P
-C Flow , Indicated Flow Indicated Flow 2-!
NEDO-24312 where C (= 0.95) is defined as the ratio of " Inactive Loop True Flow" to "Inac-tive Loop Indicated Flow", and " Loop Indicated Flow" is the flow indicated by the jet pump "singic-tap" loop flow summers and indicatora, which are set to indicate forward flow correctly.
The 0.95 factor was the result of a conservative analysis to appropriately modify the single-tap flow coefficient for reverse flow.* If a more exact, less conservative core flow measurement is required, special in-reactor calibration tests would have to be made. Such calibration tests would involve calibrating core support plate 6P versus core flow during two pump operation along the 100% flow control line, operating on one pump along the 100% flow control line, and calculating the correct value of C based on the core flow derived from the core support plate AP and the loop flow indicator readings.
2.1.2 Core Flow Uncertainty Analysis i
1 1
The uncertainty analysis procedure used to establish the core flow uncertainty i for one pump operation is essentially the same as for two-pump operation, except for some extensions. The core flow uncertainty analysis is described in Refer-ence 2. The analysis of one-pump core flow uncertainty is summarized below.
For single-loop operation, the total core flow can be expressed as follows ;
(refer to Figure 2-1):
WC"WA-WI where WC = total core flow; WA = active loop flow; and Wy = inactive loop (true) flow.
By applying the " propagation of errors" method to the above equation, the vari-ance .of the total flow uncertainty can be approximated by:
- The expected value of the "C" coef ficient is /0.88.
2-2
NED0-24312
~U U C sys 1-a y Arand 1-a frand where C = uncertainty of total core flow; WC Og = uncertainty systematic to both loops; Og = random uncertainty of active loop only; Og = random uncertainty of inactive loop only; O
C
= uncertainty of "C" coefficient; and a = ratio of inactive loop flow (W )Ito active loop flow (W )* A Resulting f rom an uncertainty analysis, the conservative, bounding values of og sys, ogA ,OgI rand and UC are 1.6%, 2.6%, 3.5%, and 2.8%,
rand respectively.
Based on the above uncertainties and a bounding value of 0.36 for "a", the variance of the total flow uncertainty is approximately:
1 l 2 2 2 = (1*6)2 1
(2.6) +
-(3.5)2 + (2.8)2 = (5.0%)2 ,
OWC 1-0.36 1-0.36 -
f When the effect of 4.1% core bypass flost split uncertainty at 12% (bounding case) bypass flow fraction is added to the above total core flow uncertainty, the active coolant flow uncertainty is:
2 Ojetive = (5.0%) +
1 2 (4.1%)2 = (5.0%)2 coolant which is less than the 6% core flow uncertainty assumed in the statistical analysis.
2-3
NEDO-24312 In s ummary, core flow during one-pump operation is determined in a conservative way, and its uncertainty has been conservatively evaluated.
2.2 TIP READING UNCERTAINTY To ascertain the TIP noise uncertainty for single recirculation loop operation, a test was per f ormed at an operating BWR. The test was performed at a power level 59.3% of rated with a single recirculation pump in operation (core flow 46.3% of rated). A rotationally symmetric control rod pattern existed prior to the test.
Five consecutive traverses were made with each of five TIP machines, giving a total of 25 traverses. Analysis of their data resulted in a nodal TIP noise of 2.85%. Use of this TIP noise value as a component of the process computer total uncertainty results in a one-sigma process computer total uncertainty value for single-loop operation of 9.1% for reload cores.
2-4
. NEDO-24312 l
l l
l CORE l
l
[
/ W C g A
WC = OTAL CORE HOW WA = ACTIVE LOOP FLOW W ,* = INACTIVE LOOP FLOW I' Figure 2-1. Illustration of Single Recirculation Loop Operation Flows 2-5/2-6
- 3. MCPR OPERATING LIMIT 3.1 CORE-WIDE TRANSIENTS Operation with one recirculation loop results in a maximum power output which is 20% to 30% below that which is attainable for two pump operation. Therefore, the consequences of abnormal operational transients from one-loop operation will be considerably less severe than those analyzed f rom a two-loop operational mode.
For pressurization, flow decrease, and cold water increase transients, previously transmitted Reload /FSAR results bound both the thermal and overpressure conse-
- quences of one-1 cop operation.
Figure 3-1 shows the consequences of a typical pressurization transient (turbine trip) as a function of power level. As can be seen, the consequences of the tran-sient during one-100p operation are considerably less because of the associated reduction in operating power level.
The consequences from flow decrease transients are also bounded by the full power analysis. A single pump trip from one-loop operation is less severe than a two pump trip from full power because of the reduced initial power level.
Cold water increase transients can result from either recirculation pump speedup or restart, or introduction of colder water into the reactor vessel by events such as loss of feedveter heater. The Kf f actors are derived assuming that both recirculation loops increase speed to the maximum permitted by the M-G set scoop tube position. This condition produces the maximum possible power increase and hence maximum ACPR for transients initiated from less than rated power and flow. When operating with only one recirculation loop, the flow and power increase associated with the increased speed on only one M-G set will be less than that associated with both pumps increasing speed; therefore, the K factors derived with the two-pump assumption are conservative for single-f loop operation. Inadvertent startup of an idle recirculation pump is not the limiting reactivity insertion transient. In addition, the restart of an idle pump would actually result in a neutron flux transient which would exceed the flow reference scram. The resulting transient with scram is expected to be less severe than the worst-case cold-water transient from rated power / flow.
3-1
)
The latter event (loss of feedwater heating) is generally the most severe cold water increase event with respect to increase in core power. This event is caused by positive reactivity insertion from increased subcooling of core inlet flow; there fore, the event is primarily dependent on the initial power level.
The higher the initial power level, the greater the CPR change during the transient.
Since the initial power level during one pump operation will be significantly lover, the one-pump cold water increase case is conservatively bounded by the full power (two pump) anal) sis.
From the above discussions, it can be concluded that the transient consequence from one-loop operation is bounded by previously cubmitted full power analysis.
3.2 ROD WITHDRAWAL ERROR The rod withdrawal error at rated power is given in the FSAR for the initial core and in cycle dependent reload supplemental submittals. These analyses are performed to demonstrate that, even if the operator ignores all instrument indications and the alarms which could occur during the course of the transient, the rod block sy9 tem will stop rod withdrawal at a minimum critical power ratio which is higher than the fuel cladding integrity safety limit.
During single-loop operation, correction of the flow-biased rod block monitor (RBM) equation (below) and the lower reactor power obtainable assures that the r
MCPR safety limit would not be violated during the postulated RWE.
One-pump operation results in backflow through 10 of the 20 jet pumps while the flow is being supplied into the lower plenum from the 10 active jet pumps.
Because of the backt' low through the inactive jet pumps, the present rod block equation was conservatively modified for use during one-pump operation, because the direct active-loop flow measurement may not indicate actual flow above about 35% drive flow without correction.
3-2
NEDO-24312 A procedure has been established for correcting the rod block equation to account for the discrepancy between actual flow and indicated flow in the active loop.
This preserves the original relationship between rod block and actual effective drive flow when operating with a single loop.
The two pump red block equation is:
1 RB=mW+(RB100 - m(100))
I The one pump equation becomes:
RB = mW + (RB100 - m(100)) - mow where AW = difference, determined by utility, between two-loop and single-loop ef fective drive flow when the active loop indicated flow is the same; RB = power at rod block in %;
m = flow reference slope for the rod block monitor (RBM);
W = drive flow in % of rated; and RB100 = top level rod block at 100% flow.
If the rod block setpoint (RB100) is changed, the equation must be recalculated using the new value.
The APRM trip settings are flow-biased in the same manner as the rod block monitor trip setting. The re f ore , the APRM rod block and scram trip settings are subject to the same procedural changes as the rod block monitor trip setting discussed above, 3-3
.1 _ . _ _ _ _ _ _
4 NEDO-24312 3.3 OPERATING MCPR LIMIT For single-loop operation, the rated condition steady-state MCPR limit is in-creased by 0.01 to account for the increase in the fuel cladding integrity safety limit (Section 2). At lower flows , the steady-state operating MCPR limit ia conservatively established by multiplying the rated flow steady-state limit by the Kr f actor. This ensurcs that the 99.9% statistical limit require-ment is always satisfied for any postulated abnormal operational occurrence.
l l
l l
l l
i i
l l
l l
l l
l l
3-4
NEDO-24312 1160 f
1140 3
1120 -
200 $4 m
5
.J
-t 5 a) l C 1100
= 8 Z
%._ x 100 g 5 1080 -
E a 'x
$ E e B A w 1060 -
f C 5
? 5 5 e b
E k 1040 -
E 1020 1000 l
980 -
RANJE OF EXPECTED P- +
MAXIMUM 1 LOOP POWER OPERATION I I I I 960 o 20 40 00 80 100 120 140 POr!ER LEVEL (% NUCLEAR BOILER RATED) l Figure 3-1. Main Turbine Trip with Bypass Manual Flow Control 3-5/3-6 1
- 4. STABILITY ANALYSIS l
l The least stable power / flow condition attainable under normal conditions occurs at natural circulation with the control rods set for rated power and flow. This condition may be reached following the trip of both recirculation pumps. As shown in Figure 4-1, operation along the minimum forced recirculation line with one pump running at minimum speed is more stable than operation with natural cir-culation flow only, but is less stable than operation with both pumps operating at minimum speed. Under single-loop operation, the flow control should be in master manual, since control oscillations may occur in the recirculation flow control system under these conditions.
4-1
\ _ - . _ _ _ _ _ _ _ _ _ _
NEDO-24312 1.2 l
i ULTIMATE STABILITY LIMIT '
1.0 - - - - - - - " " " " " " " " " " " " " " " * " " " " ~ " " " ~ ~ ' l j
- == == = SINGLE LOOP, PUMP MINIMUM SPEED
====== ======= BOTH LOOPS, PUMPS MINIMUM SPEED l
l 0.8 -
IC 3"
9 y 0.6 -
e
'4 NATURAL RATED FLOW d' CIRCULATION ! CONTROL 0 LINE /j LINE ff
/f 0.4 -
/
HIGHEST POWER ATTAINAB LE j// FOR SINGLE LOOP OPERATION 0.2 -
l I l 1 0 20 40 60 80 100 POWER (%)
Figure 4-1. _ Decay Ratio Versus Power Curve for Two-Loop and Single-Loop Operation 4-2
_ _ _ _ _ _ _ . )
=
NED0-24312
- 5. ACCIDENT ANALYSES 1
l l
~The broad spectru., of postulated accidents is covered by six categories of design l 1
basis events. These events are the loss-of-coolant, recirculation pump seizure, 1
control rod drop, main steamline break, refueling, and fuel assembly loading J accidents. The analytical results for loss-of-coolant and recirculation purp seizure accidents with one recirculation pump operating are given below. The I
results of the two-loop analysis for the last four events are conservatively I
applicable for one-pump operation.
1 5.1 LOSS-OF-COOLANT ANALYSIS 5.1.1 Break-Spectrum Analysis A break-spectrum analysis for single-loop operation was performed for Millstone Point Unit 1 using the model and assumptions documented in Reference 3. For the standard two-loop break-spectrum analysis, the most limiting break was a 100% severance of the recirculation suction line. This design basis accident (DBA) has a total uncovered time as shown in Figure 5-2 and a boiling transition time of less than 8 seconds for both 8x8 standard fuel and 8x8R fuel.
For single-loop break spectrum analysis, a boiling transition time of 0.1 second is conservatively assumed for all breaks, and the reflooding times and total uncovered times are similar to the times calculated for the two-loop analysis as shown in Figures 5-1 and 5-2. Therefore, the most limiting break for the single-loop analysis is also the DBA 100% severance of the recirculation suction line. The single-loop reflooding time is within one second of the two-loop reflooding time (Figure 5-1).
5.1.2 Single-Loop MAPLHCR Determination Since the limiting reflooding time for single-loop operation is similar to the reflooding time for two-loop operation, the procedure described in Section II.A.7.3 of Reference 3 is conservatively applicable. Reduction factors for l
the maximum average planar linear heat generation rate (MAPLHCR) were determined I for all 8x8 and 8x8R fuels.
i 5-1
NEDO-24312 The most limiting MAPLHGR reduction factors are for the DBA (100% severance of the recirculation suction line) and cre presented in Table 5-1. Single-loop operation MAPLHCR values are derived by multiplying the current two-loop opera-tion MAPLHGR values by the reduction factor for that fuel type. As discussed in Reference 3, single-loop MAPUICR values are conservative when calculated in this manner.
The analyses were performed assuming the two recirculation manifolds are isolated from one another by closure of appropriate valves in the cross-tie (equalizer) line between the loops. The discharge valve in the idle recirculation loop is normally closed, but if its closure is prevented, the suction valve in the loop l
should be closed to prevent the loss of Low Pressure Coolant Injection (LPCI)
) flow out of a postulated break in the idle suction line.
Table 5-1 LIMITING MAPLHGR REDUCTION FACf0RS Fuel Type Reduction Factors 8x8 0.86 8x8R 0.86 5.1.3 Small Break Peak Cladding Temperature Section II.A.7.4.4.2 of Reference 3 discusses the small sensitivity of the calcu-lated peak clad temperature (PCT) to the assumptions used in the one-pump opera-tion analysis and the duration of nucleate boiling. As this slight increase (50*F) in PCT is overwhelmingly offset by the decreased MAPLHGR (equivalent to 300* to 500*F PCT) for one-pump operation, the calculated PCT values for small breaks will be significantly below the 2200*F cladding temperature limit speci-fled in 10CFR50.46 5.2 ONE-PUMP SEIZURE ACCIDENT The one-pump seizure accident is a relatively mild event during two-recirculation-pump operation, as documented in References 1 and 2. Similar analyses were per-formed to determine the impact this accident would have on one-recirculation-pump operation. These analyses were performed with the models documented in Refer-ence 1 for a large core BWR/4 plant (Reference 4). The analyses were initialized 5-2
af M# NEDO-24312
.'from . steady-state operation at - the following initial conditio.a4 with the added condition of one inactive . recirculation loop:
L 1
thermal power = 75% and core flow = 58%,' and i
thermal power = 82%- and core flow = 56%. 1 These conditions were chosen because they represent reasonable upper limits 2
of ' single-loop operation within existing MAPLHGR and MCPR limits at the same
{.
maximum pump speed. Pump seizure pas simulated by setting the single operating 1 pump speed to zero instantaneously. ,
The anticipated sequence of. events following a recirculation pump seizure which f occura during plant operation with the alternate recirculation loop out of ser-l vice is-as follows:
- 1. The recirculation loop flow in the loop in which the pump seizure occurs drops instantaneously to zero. '
2'. ' Core voids increase which results in a negative reactivity insertion and a sharp decrease in neutron flux.
'k
- 3. Heat flux drops more slowly because of the fuel time constant.
- 4. Neutron flux, heat flux, reactor water level, steam flow, and feedwater flow all exhibit transient behaviors. However, it is not anticipated that the increase in water level will cause a turbine trip and result in a scram.
It is expected that the transient will terminate at a condition of natural cir-culation and reactor operation will continue. There will also be a small decrease in system pressure.
The minimum CPR for the pump seizure accident for the large core BWR/4 plant was determined to be greater than the fuel cladding integrity safety limit; therefore, no fuel f ailures were postulated to occur as a result of this analyzed event.
These results are applicable to Millstone Point Nuclear Power Station, Unit 1.
5-3
!J I zgoh*$ _.
0 2
1 0 -
1 1 .
1 -
P O -
P O O L 1 0
0 -
O E 1 L L O G W NS T
I s -
e m
i
- I 0 9 T d
- )
A B
D l
f o
o e
I 0N R 8O I T m C u
/ U S t r
F c
/ O e p
0T
/ I 7N E
C S
k
/ R a e
/ E P r p (
A B
/ 0 E n I
8R o
/ A K
i
/ A E
t c
u f R B S
/ I 0 5 .
1 5
e r
u
\ I 0 4
g
\ i F
/
I 0
3
\
- - - - - - - 0 2
0 0 0 0 0 0 0 ,
5 4 3 2 1 m 9 s 2 2 2 2 2 2 1 1 ,
5'"2gz58ah
~
V' $
i(ll
llI d .
O Mg.oFWr o t 0
2 1
0 I 1 1
- P O _
O P 0 O L I 0 O E 1
/ L L s _
OG e
/ W N6 m T S i .
T _
- / = I 0 9 d
, e
= r -
- * )
e v _
/ A B o c
D
/ 0N n
f 8O U
/ I T m _
C u U r S t l F O
c e
I 0T p 7
N S E
C k
/ R E
P a
e r
/ 0E
(
A B I
6R n A o i
/ K A
t c
/ E R
u
/ B S I 0 5 .
2 3
/
/ e J 0 r
u I
4 g i
F vi 0 3
- - - - - - 0 2
0 0 0 0 0 o 1
2 m w S
8 1
7 3
0 1 1 3 i{w5$*5 '<&S Y*
Ii \
I1\l1 1!i !
4
- @o " ew
~
_i M -
o s P
i s e O _ m P O i O L T O E L L _ y O GN r w e v
I T S _ )
A o B c
- i oD y n N U
- O I
m
- T C u U r S t F
c O e p
T S N
g oE k sC a R e E
P r
(
B A
E n R o A i K t A c g oE u s R S B
3 5
e r
u g g
, i F
N
\
\ \
\ \
\
0 o o 0 0 0 7 s s 4 3 2 j;sP $s>8z3 Ye lIll
- 6. REFERENCES
- 1. General Reload Fuel Application, General Electric Company, August 1979 (NEDE-240ll-P-A-1).
- 2. General Electric BWR Thermal Analysis Basis (CETAB): Data, Correlation, and Design Application, General Electric Company, January 1977 (NEDO-10958-A).
- 3. General Electric Company Analytical Model for Loss-of-Coolant Analysis in Accor dance with 10CFR50 Appendix K Amendment No. 2 - One Recirculation Loop Out-of-Service, General Electric Company, Revision 1, July 1978 (NEDO-20566-2).
- 4. Enclosure to Letter No. TVA-BFNP-TS-Il7, 0.E. Gray, III, to Harold R. Denton, September 15, 1978.
6-1/6-2
s i __
- , ; , i I
j' l 1.
1 i"'
.j
,J l
s) :
I l-l' l
G E N ER Al.$ E LECTRIC
_ _ _ _ _ _ . - . . - _ _