ML20085D052
| ML20085D052 | |
| Person / Time | |
|---|---|
| Site: | Saxton File:GPU Nuclear icon.png |
| Issue date: | 12/20/1962 |
| From: | SAXTON NUCLEAR EXPERIMENTAL CORP. |
| To: | |
| Shared Package | |
| ML20083L048 | List:
|
| References | |
| FOIA-91-17 11397, NUDOCS 9110150215 | |
| Download: ML20085D052 (23) | |
Text
.~
6 e
Nhh l
4
% of complianca SAFEOUARIG REPORT FOR PHASE 10F SAXTON NUCLEAR EXPERIMENTAL CORPORATION FIVE-YEAR RISEARCH AND DEVELOPMENT PROGRAM ADDENDUM NO. 1 December 20, 1962
.].;
Js
,5 2/
9110150215 910424
//
f7 D
DK 7
.r..,,,n
= _.
l l
6 I
l TABLE OF CONTENTS l
Page No.
i I.
INTRODUCTION 1
A.
S c ope a nd Ba c k gr ou nct.....................................
B.
Development Objectives of the Proposed Core Modifications -
2 l
1.
Operation with Special 9-Rod Subassemblies..........
2 2.
Operation with h-Rod Subassembly....................
3.
Operation with Hollow Tube Subassembly..............
3 C.
Tentative Schedule of Proposed Tests......................
3 r
II.
CORE MODIFICATIONS A.
Description of Special Fuel Subassemblies I
1.
Mechanical Design -
a.
Nine-Rod (3 x 3) Subassembly No. 1............
h b.
Nine-Rod (3 x 3) Subassembly No. 2............
5 l
c.
Four-Rod Subassembly No. 3....................
7 i
d.
Hollow Tube Subass embly No. h.................
7 8
f 2.
T he rmal a nd Hydr a ulic De s ig n........................
10 l
3.
N d. ear Design......................................
l III. ACCIENT ANALYSIS 11 A.
General...................................................
11 B.
Bor on R emov al A c c id e nt....................................
C.
L e s s of Flow A c c ide nt.....................................
13 6
TABLE II-1 PHYSICAL DATA..,......................................
TABLE II-2 THEPM.L AND HYDP.AULIC DA1 A............................
9 12 l
TABLE III-1 ACGIDENT ANALYSIS PARAMETERS.........................
FIGURE 1 - 3 x 3 FUEL SUBASSDIBLY - SCHEMATIC CROSS SECTION FIGURE 2 - 3 x 3 TOP PLATL IATCH AND SUPPORT ASSEMBLIES FIGURE 3 - 2 7. 2 FUEL SUBASSEMBLY, SCHEMATIC CROSS SECTION FIGURE h - 2 x 2 TCP PLATE, LATCH, AND SUPPORT ASSEMBLIES FIGURE 5 - HOLLOW TUBE SUBASSEMBLY - SCHEMATIC CROSS SECTION FIGURE 6 - FLUX OSCILLATOR ASSEMBLY VERTICAL SECTION FIGURE 7 - DNB RATIOS FOR 2 x 2 SUBASSmBLY (20 kw/ft) AND SPIKED ASSDiBLY (16 kw/ft) FOLLOWINO LOSS OF PUMP POWER APPENDIX A - JUSTIFICATION FOR USE OF THIN CLAD STAINLESS STEEL "EL PCDS IN SAXTON
... ~ - -. - -._
4 I.
INTRODUCTION A.
Scope and Background Tne limiting safeguards considerations governing the design and operation of a special spiked fuel subasaembly in the Saxton core have been analysed previously in the Safeguards Report for Phase 1 of the Saxton Research and Development Program, submitted on January 5,1962 as Amendment No.10.
In order to augment the previouc record with regard to changes in the test proEram and subassembly design, the present report describes and evaluates the effect on reactor safety of the following additional experiments.
(1) Reactor operation with two special 9-rod subassetblies containing Various enrich-Zircaloy clad fuel rods and thin stainless steel clad fuel rods.
fuel pellets are to be used in these subassemblies and a few of the ments of UO2 pellets in the stainless steel clad rods in one of the subassemblies will ultimstely be operated at a power density as high as 16 kw/ft.
(ii) heactor operation with a special h-rod subassembly with stainles steel clad rods of larger than normal diameter and pitch, and operation of this subassembly at power densities up to 20 kw/ft.
(iti) Reactor operation with a non-fuel bearing sucassembly, consisting of nine instrumentation thimbles for the purpose of evaluating miniature, direct reading flux detectors and conducting flux oscillator experiments.
The foreEoing program represents a departure from that described in the previous amendment in a number of significant respects. The following summary is presented to clarify these changes l
(iv) The Zircaloy clad 9-rod fuel sabassembly mentioned on pages lh and f
23 of the Safeguards Report for Phase 1 of the Research and Development Program will not be inserted in the reactor initially but may be used later in the I
Phase 1 program.
(v) The 9-rod Zircaloy subassembly mentioned above and the special stainless steti clad 9-rod subassemblies discussed on pages lh and 15 of the Phase 1 Safeguards Report (subassemblies designed to produce 16 kw/ft at 20 MW and 23.5 M4 reactor power levels) will be supplemented by two new 9-rod sub-assemblies containing four corner rods clad with Zircaloy and the five remaining rods clad with thin stainless steel.
The two stainless steel clad 9-rod sub-assemblies will not be inserted in the reactor initially but may be used later in the Phase 1 program.
(vi) Operation with the b-rod subassembly with stainlass steel clad -
rods at 20 ku/ft and the non-fuel bearing thimble subassembly are new provisions in the program.
(vii) It is still planned to operate with the special stainless steel clad 63-rod hollow assembly which is mentioned on page lh of the Phase 1 Safeguards Report and has fuel enrichments shown as case (1) on page-21 of the Phase 1 Safeguards Report.
(viii). A new test schedule provides for insertion of the special assenblies and spiked assemblies, with the exception of the two all stainless steel 9-rod
.--.m...._...
~.
2 subassemblies and the 63-rod hollow assembly mentioned above under item (vii) p.ior to chemical shim operation.
The schedule shown in Phase 1 Safeguards Report provided for approximately three months of chemical shim operation with a uniformly enriched core prior to inserting the specially enriched fuel assemblies.
The effects of operating a slightly spiked 9-rod subassembly in a central lattice position, a condition characterized by a power density of 16 kw/f t and a steady-state reactor power of 23.5 MWt, have been analyzed and reported in
- the Phase 1 Safeguards Report.
Neither of these criteria will be exceeded during the phase of the program covered by the present amendment.
Initial operation of one of the new 9-rod subassemblies which will be enriched so that it produces 16 kw/ft in the central core position at a steady-state reactor power level of 23.5 MWt will take place in a peripheral pos,ition where the local flux level is well b dow the design value. After visual inspection following operation in the peripheral core position, this element will be moved to the central core position for oper?+. ion at a power density as high as 16 kv/f t.
The other new 9-rod subaisembly aill have fuel enrichments da 'gned to produce no more than 13 3 kw/ft when located in the central core pc ion at a reactor power level of 20 MWt.
Therefore, insofar ae the operation with the two 9-rod test subassemblies is concerned, there is no operating condition or potential accident more severe than those analyzed in the safeguards reports submitted to date. This report will demonstrate that no undue hazards will result from operating the h-rod sub-assembly in a peripheral locat, ion at the power level specified.
I In future phases of the development program, further changes in core I
make-up, including repositioning of the subassemblies in question, are con-templated which will equal and ultimately exceed the operating conditions here-tofore evaluated.
At such times as it is proposed to extend these limits, appropriate analysis will be performed and submitted in support of such changes.
B.
Development Objectives of the Proposed Core Modifications 1.
Operation wity 7pacial 9-Rod Subassemblies It is planned that after about one year of operation in the initial l
core locations, the two 9-rod subassemblies will be removed for nondestructive I
testing and vi.aual examination to determine general. performance characteristics.
In addition, two of the thin stainless steel clad rods may be removed from each subassembly for destructive examination.
I If the results from the visual examination are favorable, these sub-assemblies will' be put back in the reactor, and the higher enrichment subassembly will be operated at its rated power density corresponding to 16 kv/ft to provide comparative data at the higher thermal output and burnup level.
2.
Operation with h-Rod Subassembly One of the objectives of the Saxton development program is the investigation of oxide fuel performance at higher power output per unit length To this end it is planned to operate the speci'al h-rod and higher burnup.
subassembly ultimately at power densities of the order of 1.5 times the present core maximum. The large rod diameter provided -in this subassembly enables a
3.
power density increase of this magnitude to be made without exceeding the present After irradiation at these higher power densities, range of surface heat flux.
the subassembly will be visually examined and one rod may be destructively tested.
3.
Operation with Hollow Tube Subassembly The hollow tube subaseembly will serve as a facility for special in-core instrumentation evaluation. The scope of this development program includes testing of various types of flux detectors, including miniature fission chambers, Performance gamma chambers, and fission-sensitive thermometric type detectors.
and reliability of the detectors and The associated electrical cables and hard-ware will be studied for both fixed and noveble devices.
The program makes use of some of the thimbles initially installed for the Saxton flux wire system in addition to this special subassembly described in this report.
Apart from the detector development program, it is intended to utilize one tube of the special subassembly for a flux oscillator. The purpose of this experiment is to evaluate flux oscillator technique as a means of measuring the reactivity characteristics of the core at any time in life. The in-core device, which is described in greater detail in Section II.A.l.d. of u
this report, is a meana of producing cyclic or random variations in neutron absorption over a localized region of the core.
By controlling the amplitude and frequency of these perturbations, and by feeding the resulting flux signal to a cross-correlation computer, it is possible to obtain the transfer function of the core.
From this information a reactivity analysis can be performed which will be useful in the evaluation of boron hideout and other aspects cf themical shbn control. The technique will also be employed to study the spectra 1 charac-teristics of reactor fluctuations.
C.
Tentative Schedule of Proposed Tests The following is a timetable of the proposed experimental program as now conceived:
(il Initial power tests -
Now under way and due to be completed by February 15, 1963.
(ii)
Insertion of special test subassemblies -
Completed by middle of March 1963 (iii) Rodded core operation with special test subassemblies -
4 Completed by end of March 1963 (iv) Chemical shim operation with special test subassemblies -
Power levels up to 23.5 MWt, April through October 1963 Insertion of test 63-rod hollow assembly and 9 od spiked subassembly in central core location, November 1963 Power levels up to 23.5 MWt, December 1963 through July 196h.
I,
~_.
h.
II.
CORE MODIFICATIONS A.
Description of Special Puel Subassemblies _
l.
Mechanical Design The standard Saxton fuel assembly is described in Section 203-B of It was noted in that description that certain of the Final Safeguards Report.
the assemblies, designated as Group 2 assemblies, contain removable subassemblies, It was further comprising the center 3 x 3 fuel rod cluster of the 9 x 9 array.
stated that the removable subassembly is encased in a 0.020-in. thick perforated etsinless steel rectangular channel, whose end caps are designed in such a way that the coolant encounters a flow path similar to that of the Group 1 or Standard 72-rod assembly.
Except where differences are noted in the following description, the featurcs of the special fuel subassemblies to be inserted in Group 2 assemblier under the provisions of this amendment are the same as those of the removable 3 x 3 clusters for which they are to be substituted.
Nine-Rod (3 x 3) Subassembly No.1 a.
This subassembly is to be inserted in the Group 2 assembly in that location designated by the numeral I in the core cross-section diagram, It consists of nine fuel rods, Figure 203-2 of the Final Safeguards Report:
j four of which are clad with Zircalcy-b and five of which are clad with Type 30h stainless steel. As shown in the cross-sectional sketch, Figure 1, the four Zircaloy rods occupy corner positions.
All of the stainless steel rods, except the central rod are provided with specially shaped end plugs which This design facilitates removal protrude through the subassembly top plate.
of these rods from the subassembly by means of long-handled tools when the sub-assembly is in the fuel storage rack.
The subassembly support tube, which provides a means of inserting and withdrawing the subassembly through the reactor head port, is specially modified to permit remote unlatching.
Separation of the subassembly from the support tube will be performed only in the storage well, in conjunction uth visual inspection and rod replacement.
The support tube incorporates a pitot tube and thermocouple assembly for measurement of channel exit coolant velocity and temperature. A hollow cylindrical column, surrounding the pitot tube, serves as a stop to prevent the removable fuel rods from moving upward with respect to the subassembly when the reactor is in operation.
The support tube, latch, and top plate of the 9-rod subassembly are shown in Figure 2.
The five stainless steel fuel rods in subassembly No. 2 are of normal (5.69 w/o U-235) Sarton core enrichment, and will operate at essentially the same power density and heat flux as the subassembly which they replace.
l The cladding of these five rods will be thinner than that of the standard rods, however, and will rely on the enclosed pellets for support against coolant The maximum strain of the cladding occurs during initial hot, pressure.
zero-power pressurization, when there would be a maximum diametral cleartnce between pellet and clad in the unstressed conditica. Elastic collapse of the cladding onto the pellet will occur, with possible localized permanent strain of up to about 0.h% yield.
The mechanism of collapse is characterized by deformation to an elliptic cross section with line contact between pellet and
.~ -.-
$. j Expansion of the clad, and not by the formation of a longitudinal " wrinkle."
pellet as the reactor is brought to power will reduce the compressive stress in the cladding, and, in the erLreme case, produce a tensile stress, whose maximum (27,700 psi) is well below ths yield strength (6b,000 psi) at full-po An analysis showjng the temperature.
yielding because of strain hardening of the material. adequacy of The tubing used for fabrication of the thin walled clad is Properties corre-10 mil, welded and cold drawn, Type 30h stainless steel.The specifications for sponding to 10% work hardening are used for design.
quality control and inspection are the same as those used for the procurement of normal Saxton fuel tubes as well as those for the Yankee and Selni reactor The specification provides for chemical snalysis, determination of room temperature and elevated temperature tensile properties,100% eddy-current inspection, and dimensional inspections including inside diameter, wall thickness, Following assembly and end closure welding, each ovality of I.D. and camber.
thin clad fuel rod is dimensionally inspected, visually examined under 30X magnification for ipdications of cracking in the veld zone, and helium leak tested to a 1 x 10-o ec/secleakrate.
The four Zircaloy-b clad fuel rods are the same as the ones described in paragraph $ of Supplement No.1 to Amendment No.10 to the Saxton The Zircaloy cladding is designed to be free standing; license application.
that is, to sustain maximum external pressures with only elastic deformation and no pellet support. The quality control and meterial specifications for the l
Zircaloy cladding are the same as those employed in the procurement of the CVTR Inspections include complete ultrasonic and dimensional checks, a helium fuel.
leak test, and radiography of final welds.
The physical data describing 9-rod subassembly No.1 is included in Table II-1, along with comparative data for a standard Saxton subassembly.
b.
Nine-Rod Subsssembly No. 2 This subassembly is identical to that described in a. above, except that the average enrichment of the five stainless clad rods is increased to provide a higher power density when the. subassembly is ultimately moved to This subassembly will initially be inserted in i
l lecation III in the core.The enrichment schedule for the rods in this subassembly location V in the core.
is as follows:
Enrichment, w/o U-235 Pellet Region lb pellets (bottom of stack) 5.69 2 pellets (next above) 7.30 3 pellets (next above) 6.81 12 pellets (center of stack) 6.h6 3 pellets (next Lbove) 6.81 2 pellets. (next above) 7.30 1h pellets (top nf stack) 5.69 The fcur Zircaloy clad rods in this subassembly are the same, including the fuel enrichment of 6.1 w/o U-235, as the Zircaloy rods in sub-assembly No. 1.
TAEE II-1 PHYSICAL DATA
]
(All dimensions nceinal) 4 i
9-Rod h-Red Standard Subassembly Sutassembly Subassembly Stainless Clad Zircaloy Clad g inless Clad No. of rods per subassembly 9
5 4
4 Active fuel length,.in.
36.6 33 9 36.0 36.0 Rod outside dia., in.
O.391 0 3705 0 391 0 5995 Clad thickness, in.
0.015 0.0095 0.0237 0.0235 i
Pellet dia., in.
O.357 0 357 0 337 0 54h l
0.004 0.004 0.0065 0.008 Cold pellet-to-clad gap, in.
Pitch, in.
O.580 0 535 0 535 0.746 6.1 83 Enrichment,v/oU-235 5.69 Center subassembly enrichment = 5 69; peripheral subassembly enrichment given on page $.
i 4
4 i
i 9
1 7.
3 Four-Rod Stgassembly No. 3 c.
This subassembly which will be placed in location IV in the core contains four stainless steel clad rods in a 2 x 2 rectangular array.
The diameter of these rods is 50% larger than the standa-d rod. Although the fraction of the channel cross section occupied by fuel rods is the same as in the standard fuel assembly, the net area for coolant flow is slightly reduced by the addition of a 3/16-in. 0.D. flux wire thimble which extends for the full length of the fuel cluster along its longitudinal axis.
Figure 3 is a cross-sectional sketch of the b-rod subassembly.
The fuel rods are secured by a spring clip grid assembly and surrounded by a 20 mil stainless steel rectangular channel. The channel for this assembly is closed, rather than perforated, to prevent mixing and thus allow a more precise heat balance to be made on the basis of exit flow and The subassembly support tube incorporates a pitot tube and enthalpy rise.
thermocouple probe, and may be separated from the fuel subassembly by means of a remote latch similar to that employed on the 9-rod subassembly.
A sketch of the end plate, latch, and support tube is shown in Figure h.
The stainless steel cladding is fabricated from Type 30h welded and cold-drawn tubing procured to the same ouility control and inspection specifications as the thin valled and standard cladding used in the remainder of the fuel. It is designed to be free standing, and at maximum reactor pres' u e and temperature, deformation will not exceed the elastic limit.
Table II-1 includes the detailed fuel dimensions for this subassembly.
Enrichment of the h-rod subassembly is set at 8.3 w/o U-235, and the enrichment is uniform for all pellets in the subassembly, d.
Hollow Tube Subassembly No. h This subassembly, which contains no fuel-bearing rods, is designed to accommodate a variety of special instrument probes and an oscillator rod. This subassembly will be inserted in core location II. The cluster consists of one 1/2-in. I.D. thimble, four 1/h-in. I.D. thimbles, two 3/16-in. I.D.
thimbles, and two 1/8-in. I.D. thimbles, arranged as shown in Figure 5.
A
=
schedule of specific instrument probce to be inserted in these thimbles 10 not firm at this time; however, the following list of proposed items for evaluation is typical:
(1) Ionization chambers, without guard ring construction, containing U-235 activation material, and connected to semiflexible coaxial cables.
(ii) Ionization chambers, sin 11ar to (i), but containing no uranium, to determine gamma sensitivity.
(iii) Cable for (i) and (ii) without-detectors, to determine effects of radiation on cable characteristics.
(iv) Ionization chambers with guard ring construction, attached to semiflexible triaxial cables.
1
_(
8.
(v) Neutron sensitive thermocouple-type detectors, which operate on the principle of differential heating of two identical alloy strips containing uranium of different enrichments.
(vi) Movable probes, consisting of any of the above detectors provided with f3exible cables.
Samples of each type of equipment purchased from a variety of manufacturers will be tested in sufficient numbers to provide statistically meaningful data.
The thimbles are fabricated from Type 316 seamless, annealed stainless steel tubing, with welded Type 30h stainless steel end plugs.
Each tube is eddy-current tested for material defects and subjected to a 3750 psi external hydrostatic test pressure.
The tht:ble cluster is supported in the core region by Type 3Oh stainless steel grid plates and is surrounded by a perforated channel similar to those of the 9-rod fuel subassembly.
Sections of Zircaloy-2, forming an annular filler of about 0.093-in, wall thickness, are inserted in the channel surrounding the tube cluster to reduce the water cross section and thus prevent undesirable flux peaking. The thimble, extend upward through a support tube to the adapter port and are scaled to the adapter seal plug by brazing. The entire subassembly may be removed or inseric9 through the adapter port without removing the vessel head.
- The 1/2-in. central thimble will normally be used for the rod oscillator. A sketch of the oscillator probe is shown in Figure 6.
It consists of two concentric tubes to which are affixed equally spaced sleeves of Ag-In-Cd absorber. A change in the degree of shadowing of the inner absorber bands by the outer bands, brought about by vertical motion of the inner tube, produces a change in the reactivity worth of the probe. An electrodynamic drive system outside the reactor is used to produce this motion.
The nominal reactivity input of which the system is capable is expected to be 2 x 10-h S k/k. If measurements indicate the actual worth is greater than 4
5 x 10-ho k/k, the maximum stroke of the oscillator drive will be limited to obtain a value less than 5 x 10-uts k/k. A reactivity change of this magnitude does not measurably affect the control of the reactor.
Because of the minor reactivity effect of the oscillator and the absence of heat or fission product generating materials generally, there is no nuclear safety problem associated with operation of the hollow tube subassembly.
2.
Thermal and Hydraulic Design The principal calculated thermal and hydraulic parameters relating to the operation of the test fuel subassemblies and the peak values representing the normal fuel are listed in Table II-2.
The values given are based on the reactor operating conditions corresponding to the maximum power level of 23.5 MWt with chemical shtm.
In Section III of the Phase 1 Safeguards Report, a considerable body of experimental-information was cited in support of operating a 72-rod
i
~
TABLE II-2 THEFMAL AND HYDRAULIC DATA 520 F 23 5 FMt Chemical Shim Operation; T inlet Spiked 0-Rod Spiked E-Fod Subassembly Subassembly Unspiked Region (Central Location)
(Feripheral Location)
Maximum themal output, kv/ft 14.1 16 20 j
Msnrimm surface heat flux, Btu /ft - hr h70,000 533,000 hkh,000 Hot - Chanael Factors F
L92 3 31 2.76 q
F 2.U 232 1.82 y
Minimum DNB retic by W-2 Correlation:
23 5 FM (100%), 2000 psia q-DIGR = 2.8h q-DNBR = 2.27 q-DIGR = 2.81 28.2 MW (120%),18'00 psia q-DIGR = 2.07 q-DIGR = 1.65 q-DIBR = 2.08 i
i 1
5
10.
11.
This in-pile experimental evidence indicated that a result of high power density test assembly at 16 kw/ft.there is no known instance of fuel damage a Justification of sustained operatic' d on the same evidence, reinforced
/ft during limited operations up to 2h kw ller, and that the flux wire and of the b-rod subassembly at 20 kw/ft is baseby the fact will provide assurance that the actual It is therefore concluded that the exit pitot tube-thermocouple probe ble step in the approach to higher limit.
power density dees not exceed the designthe proposed op power density performance evaluation.
l during normal The potential for burnout of the b-rod subassemb yf the spiked 9-The (DiB) was determined for steady state operation is substantially less than that o DNB co{ relation, known as W-2,This corre margin to departure from nucleate boiling and the overpower transient using the new developed at Westinghouse Atomic Power Division.d to the AEC in conjun i likely described in previous hazards analyses submitteAs indicated in Table 1 The corresponding to be incurred in the,h-rod subassembly at overpo the Yankee reactor.
W-2 correlation.
d 1.65 for the spiked 9-rod the burnout criterion defined as q-DNB in thesafety marg subassembly operating in the center of the core.
t under credible The relative safety margins with respect to burnou i
III.
accident conditions are evaluated in Sect on Nuclear Design loyed in the 3
The multigroup diffusion theory techniques as were empibed in S f power distribution in nuclear design of the Saxton core, and descr Safeguards Report, were applied to the determination o I
Spiked region corrections were made conservative the test subassemblies.that the peak power densities used as the bas sth i
r tests with h
evaluation ere believed to be higher t anConfirmation of the i
initial powe t
these subassemblies, using the in-core ins ru (1/2-in.I.D.) thimble t
The reactivity effect of floodii g the largestbe less than 0.00h% a6 k/k All other thimbles and all in the hollow tube subassembly was calculated tothe It has been "looding coefficients.
resented in Section III of this other core locations would exhibit smaller i
shown in the boron rer. oval accident analys s p cepted without core damage.i n on the p report that a step input of 0.33% a6 k can be acIt is con l flooding effects.
the thimble subassembly based on accidenta i
by L.S.Teng, lWCAP-1997 nNew DNB (Burnout) Correlations," Revised Edit on l
H.B.Currin, A.G. Thorp II, September 1962.
l i
=<
v__
i,v*y y-v----
p.-
,w
-,e--,-e-y ve y-w-,ry-
,y-i w--
w,
-=-wye t
-,y--
w--'
- + + - iwee TWr y-g e
y-e-
12 l
-l TAILE III-l AOC: DENT ANALYSIS PARAMETERS i
I Spiked 4-Rod Average Ai,sembly Subassembly Core 16kv/ft 20 kv/ft f
Steady ntate inlet mass velocity, 10 lb/ft hr 1.003 1.003 0 925 Maximum heat flux, 10 Btu /ft hr 0.214 0 534 0.44h Average heat flux, 10 3tuffg hr 0.153 0 378 0 318 6
2 Equivalent diameter, based on vetted perimeter, ft 0.0588 0.0588 0.0392 Equivalent diameter, based 1
f on heat transfer perimeter, ft 0.0588 0.0588 0.0728 i
Head loss coefficients, based on flov area in fuel region Bottcn end plate 5 70 5 70 2.27 Spacers, total of four 2.13 2.13 3 76 Top end plate 0.45 0.45 3 04 Steady-state core pressure drop, psi 2.61 2.61 2.61 l
..~
~
t w 13e O.
Loss of Flow Accident The loss of coolant flow due to failure of pump power was also analyzed using the W-2 correlation for the b-rod subassembly operating at 20 kw/ft.
As described later in this section, the closed channel formed by the nonperforated enclosure prevents inflow during the bulk boiling phase of the accident, causing the DNB ratio to continually decrease during the later stages of flow coactdown.
It was shown by censervative calculation that even if DNB occurred throughout the subassembly at 6% of the initial flow (where the minimum DNB ratio is calculated to be at least 1.82) there would be no clad melting.
Should such an accident occer, the h-rod subassembly would be examined before returning it to service, or an analysis would be made based on actual flow coastdotn characteristics applicable at the time of the accident, to ascertain that no defonmation of the fuel rods could have resulted.
In the lose of coolant flow accident, the DNBR at any point in the hot channel normally passes through a minimum during the ceastdown and then rises as heat flux and exit enthalpy are reduced. The lvwer set of curves in F1gure 7, showing DNBR versus time l'or the 16 kw/ft spiked assembly, illustrates this behavior. The rise in DNER after 5 seconds is due in part to cross flow between channels, adding to the mass velocity of coolant in regions of greatest heat flux and high quality. In the closed channel of the h-rod subassembly this effect is absent, and the resulting trend in the DNBR relationship is downward, as shown in the upper set of curves in Figure 7.
Thus, the critical period with regard to DNB is not passed during the time of substantial coastdown flow, but may be reached aftar natural convection flow.
Because the prediction of coolant flow is relatively uncertain in this area, ths quantitative expression of safety margin in terms of a DNB ratio is not justified. However, since much of the residual heat is removed from the pellets during the flow coastdown period following scram, it is possible to show that even if DNB did occur at a con-servatively selected point in the coastdown interval, no hazard would result.
Specifically, it was postulated that DNB occurred at all locations along the h-rod channel when the coolant mass velocity is still 6% of the initial value.
By calculations, the actual DNB ratio at this point in time is 1.82 at the channel exit and 3.h8 at 50% of the core height, the latter position not yet having reached bulk boiling.
If DNB were to occur at that instant, the maximum possible clad temperature would be about ih70 F at the center of the core, the 0
ccrresponding value at 90% of the core height is about 10000F.
It is expected that if DNB actually occurred, it would take place much later in the transient, so that the peak temperature would be substantially lower due to the-greater removal of residual pellet heat.
4
^-
N'I B
C I
f m
B \\
A B
(C C
B i
1 i
j J
(
A - Thin-Wall Stainless Clad, Fixed Rod i
B - Thin Wall Stainless Clad, Removable Rod C - Zircaloy-4 Clad, Fixed Rod i
i l
l I
FIGURE I 3 x 3 FUEL SUB.',SSEMBLY - SCHEMATIC CROSS SECTION
k 4
4,6
-a-A a
As=,nr.
A a4 t
1 gH5 j fy )
m-e gg.
ga r m
5 g g s
- QmH Il, m, >g g l
m:
a 9
4 o.
A if H
.M8
.)
1 o
m..
f A
A A
gH 8m NH N
iNb I f
/j/
l 3
7
/
N V E
a i
s i
a M.
- i c
y*
.{iil g
E w
l
- k.. _ _.s f
{1 G
9
~;,F.' L_
t s
i i
e j
d yt i
^
)'
Q i t'g
- g--
s y ("
v i,
es 3
l-g a
a: H
[
W r, ; " 7 s
pa Ei N
s.
s m
gm
' f, *,
l h \\
g i
a
<Q ti l 1
i' s a
s
- \\,
lr n
1 i.
,1 G
~
e s
I f~
i s
y<
t k +; T w
y/[~,' _'~~
' ' 'i 5
- i.,e I.'N d
5
'r.w d
i echl Q4 7 Z
- 9 4
(f'.-l-3q3 D 4))!
"'e'
,- 1 4
((. j I
)!
t n.m
!4 8
a) rs I
n
-o c= [ Q='
' ;g /p u t
u m
t' 3
,o
]
l i
'gi-(p a E
l rm GT A:
)
8
[/, /../'
m M'
mM i
- /.
l n i
'/,
-s O g f /, /..,
,,./. '.
D,.,(
en v
5W P
Y,/$k/.
c.<:.
@h...y f g tg (
!. 7> 44 m
S k
$ \\,
i t
- g
'w'f m...
u i.
m j',j'<:7 k g
gQO ' h ):
C. \\('
! { hyf 9'/
g l'k ' 'J 1h! #: M;/.('3.,
e?
Q 5
i.
6 y
2
@,y,
-c u v
I,:
9
!!'~ : C I
'tGb.'
t v
n:,;
'e,'
\\
a, ' -
i b
f/
l i it j,
3
' // 1
- '.l t
!/
\\'/,/4 C', '-}
\\
\\
, /,
\\
i i.
1
(
3 A
B G
J A
B L
)
I A - Stainless Clad Fixed Rod l
B - Steinless Clad Removable Rod C - Flux Wire Thimble l
l l
l FIGURE 3 1
2 x 2 FUEL SUBASSEMBLY, SCHEMATIC CROSS SECTION
.u.
e-
.a
--e..
4m
,-.M--.------A.-%
~.
e c.w_A.-
.e
=
.a.a.
_-a
.--.,.m--
- - - = - -
h e
eb
.g
-3 4
-s 12
((
,v,( )
a.
( (
k.*
'([ [gV ll 1(y o
4 (j
e 4
a n. o-
. tR m,,,.
j
~t m
i n.
Q p
&4
- ot (H,.%
- i y
h, s
/
> +4 9 l9 i
i y
/
. i j'!
l.l N
f
//
f, l. _.
n e -.
t, k f Q.
- t -
a li l
J ol-l m
l N,
O F4
- q
)
2 m
f SB i
~4
)
R\\ j l
4 i
l lN N
i
.5 l
41 g
4: ;.
4 1'
w..
~
)
- ~
', ~
/
i.
/
ja
.. ( a ';.
- )
L i'
l f
y:
my V'{
$g i'
) ',
N f
- l, 1,,
['
w.
3
/
l
. S.L.I s
t n.
m a f,
g NJ dt r
g,b, (4
. N 3
4 f
i
~/ ::]
[
~
~
m
/;
3 n
j q
I!
J w,
'y w
/
d i '?
,n yt b
2 s \\p q f e
r
..e
,s g; 3 e,
s g
f I
}i v
.\\
3, n
3 y
g, g.
t
/.
6t:l(, /,
['
! y
\\.
9
.y l
(,
., ) h. 's,, p.
6 I f
' //,
f b
N f3 l
lit
-'i
)%,
'O D.,,
i f
4 i.
[
v? /;l q[..
dr $.
a l
j' I.
lL J;i ~f j
i,!
9 i,
'/.
3 ll
/
11
( ks h.
_ []
l,'
-[-
f
's.
H if I
Ws
L 4
S i
4
(
l l
B I
i B
i U
^
n i
B l
l B
p's v
///////////////<,,
1 A - 1/2 In. l. D. Flux Oscillator Thimble l
B - Il4 In. l. D. Test ThimMe C - 3/16 In. l. D. Test Thimble D - Il8 in. I. D. Test Thimble E - Zircaloy Filler FIGURE 5 HOLLOW TUBE SUBASSEMBLY - SCHEMATIC CROSS SECTION
4 4
30 I
l 8
.5 2
Y
.4 s
/
g
\\/
8 cv
/
2 x 2 Subassembly 20kv/ft
- A qDNBR' "
NDNB M2.0 w
qDNBR
\\
j
\\
/
l f
Spiked Assemb,,y-16kw/ft
/
^
a M
Scram jgp33,;
Completed I
i' l
1.0.
2 4
6 f) 0 Time from Ioss of Pump Power, Seconds Figure 7 - DNB Ratios for 2 x 2 Subassembly (20 kw/ft) i and Spiked Asceably (16 kv/ft). Following Loss of Pump Power f
- 9--t-
--w++ew-m,r,,_y,,__
-~.--*-~.-.,m..._,,,,,,,,,,_;
r APPENDIX A 1
JUSTIFICATION FOR USE OF THIN CLAD STAINLESS STEEL FUEL RODS IN SAXTON INTRODUCTION As part of the over-all Westinghouse fuel development effort, it is very desirable fron a neutron eccnomy viewpoint to operate with thin clad stainless steel fuel rods. To help achieve this end, it is proposed to include 10 mil stainless steel clad fuel rods in the special assemblier to be irradiated in Saxton during the chemical shim operation.
l t
Previous Westinghouse reactor plant designs have utilized either Zircaloy clad or free-standing stainicss steel clad fuel rods.
At present, stainless steel and Zircaloy are still the major materiale considered for use fuel r:dr, and a product development program is presently as cladding for UO2 under way to evaluate the respective economies of each of these materials relative to their initial cost, fabrication costs, and perfornance.
in the case of stainless steel, the most feasible approach to economy is reduction of the wall thickness. This results in a departure from the free-standing design concept (i.e., where the tube wall is thick enough to withstand the external pressure of the coolant without collapsing) to the proposed case where the clad tube requires support from the fuel in order to prevent collapse when subh:ted to operating c0nditions.
APPLIED MECHANICS CONSIDERATION Out-of-pile tests :2 indicate the manner of collapse for a thin stain-l less steel elad fuel rod to be a function of the ratio of clad thickness to diameter, the ratio of clad thickness to cold gap, and the strength of the cladding material. Reduction of the_ values for these factors results in the l
method of collapse shifting from one of a slightly eval shape or line contact, where the cladding leans on the UO, to the formation of a wrinkle.
These out-2 l
of-pile tests indicate that the forming of a wrinkle can be minimized or eliminated for a given cladding matcrial with a fixed thicknass to diameter ratio by the reduction of the initial diametric gap.
I The clad and pellet dimensione proposed for use in Saxton are as I
follows; L
i 0 6 in.
Cladding inside diameter:
i"' -
Cladding wall thickness 0
7 in.
Pellet dianoter.
4 As the rvact: pre?sure and temperature are in:reased for the first i
time, the stresses it. the cladding at pressure and temperature following initial
l 2.
At the reactor operation with the thin clad fuel elene"te will be as follows:
The zero power condition, the clad will be collapsed upon the pellets.
pressure was taken to be 2000 psi and the temperature (coolant and clad) was taken at 5300F. The maxinum stress will be localized on the inside surface of the clad and will be approximately 80,500 psi compression. This stress was deternined using elastic methods and is a combination of hoop stress and bending The yield stress of the material at this temperature is approximately stress.
69,000 osi.
Thus, there will be sone plastic yiciding, localized, and the actual stress will be belcw B0,500 psi.
It is estimated that the total permanent strain at the zero power condition will be approximately 0.h%.
As the reactor is brought to power, the pellets will expand faster than the clad because of their higher temperature-rise.
The pellets will therefore push the clad back towards a circular shape and, in the extreme case (maximum The temperature, worst tolerance condition), the clad will be stretched.
maximum tensile stress will be 27,700 psi.
The yield stress of the material at the higher temperature (approximately 6720F mean clad) experienced during power operation is approximately 6h,000 psi so that no yielding of the clad will occur.
As the reacter is taken from full power to zero power and up again, the stresa in the clad will cycle from 27,700 psi tension to something less than 60,500 psi compression and no further yielding will occur be:ause of tnc strain hardening of the material.
Based on the Navy Code, the maximum stress range for a fatigue life of 500 cycles is 132,000 psi (66,003 x 2). The total stress range i
which the clad will experience will be 27,700 + 80,500 = 108,200 psi. Therefore, the clad rhould be capable of experiencing more than 500 cycles of reactor operatier between zero power and full power.
Taking the reactor frem full power or zero power to zero pressure and room temperature will be a less severe cycle en the clad than the zero power to full power cycle described ebove.
l FAILURE ANALYSIS OF THE THIN CLAD UNDER ELAS'fIC PLUS ah% PLASTIC YIELDIN3 Although various modes of failure of the thin wall stainless steel cans are conceivable, the' most serious and probable case is tnat of strain fatigue failure ender cyclic power conditions.
This problem is caused by the alternate cycling of the clad, where collapse due to the coolant pressure, and then expansion due to the thermal dilation of the fuel is experienced.
The over-all stress changes occurring.to thp clad is amenable to mathematical analysis.
Experiments by L. F. Coffin 3:4 over a range of thermal-and mechanical cycling conditions have indicated that the number of cycles to failure, N, may be related to the plastic strain range from maximum compression to maximum tension, a g p, through a relationship.of the form -
E C
N A6p_
=
uhere K and C are constants related to the type of materials and manner of 0.63.
0.5 and C
straining.- Tor this particular case, K
=
=
For A151 30h stainless steel, an appreciable amount of out-of-pile data in agreement with the abcVe equation shows that the material can with-etand a plastic-strain of 2% (compared to the 0.h% mentioned earlier) for
J 3
e l
Since this plastic strain rani;e is apprecially higher i
failure at 1000 cycles.than the plastic strain anticipated in thin clad fuel rods experinents in Saxton, no failures are anticipated.
Other potential modes of fatlure such as thermal ratchetinC, fretting corrosion and stress corrosion have been considered but are too 1
4 warrant further discussion.
i 3
REFERENCES _
Large Closed Cycle R&D Program, Progress hcport for the Period Se t
1.
1961 to Novcaber 30, 1961. WCAP-3707, 2 Pbel Results of Irradiation Tests of Thin Walled Stainless Steel Clad 2.
ANS Transactions, Vol. 5,1962, p.235, Rods.
A Study of the Effects of Cyclic Thermal Stresses on a Dactile Material, ASME Transactions, Vol. 76, 195h, p.931.
3 L. S. Coffin.
The Problems of Thermal Stress Fatigue in Austenitic Steol at Devated ASTM. STP 165, June 195h, p.31.
b.
Temperatures, L. S. Coffin.
?
I 5
l'
.. _.,