ML20236V769

Rev 1 to Criticality Safety Assessment for Using Plasma Arc Torch to Cut Lower Core Support Assembly

ML20236V769
Person / Time
Site:	Three Mile Island
Issue date:	11/30/1987
From:	GENERAL PUBLIC UTILITIES CORP.
To:
Shared Package
ML20236V572	List: ML20236V570 ML20236V769
References
15737-2-N09-004, 15737-2-N09-004-R01, 15737-2-N9-4, 15737-2-N9-4-R1, NUDOCS 8712070081
Download: ML20236V769 (46)

Text

r-J.5737-2-N09-004 I criticality Safety Assessment.for Ucing the Plasma' Arc Torch to cut the Lower core Support Assembly q

i l

l-1 1 0712070001 871130 PDR ADOCK 05000320i~

P PDR l

Rev. 1

1 4 j l

i 1

l NO. I Nuclear 33737-2-so9-00c Title Criticality Safety Assessment for Using the Plasma Arc Torch PAGE 2 OF 46 l to Cut Lower Core Eupport Assembly '

I Rev.

SUMMARY

OF CHANGE 1

0 Original Issue 1 Base Case model was revised to reduce the restrictions on defueling operations. ]

)

l, 1

l i

i l

t  !

i l

l f

a troma suu _j

y s

LATEST LATEST LATEST LATEST LATEST LATEST LATEST REV. SHEET REV. SHEET REV. SHEET REV. SHEET M EV. SHEET RE V.

SHEET REV. SHEET 1 46 1 l

I l

l t

I i

I

. :7 4 -

_ .I '

)

~

E c

• - P;EVISIOff STATUS SHEET .0B15737 REV. )

'sTiC.NO.

e er. @rdSAsoB6Sp Saf ety Assessment f or 2-N09 004 1

l l

I 15737-2-N09-004 Table of contents

)

Section Pace Number 2

1.0 Introduction 6 1

1.1 Background 6

{

i- 1.2 Purpose 6 1 1.3 Scope 6 1.4 Criterion for Justification.for 7 ]

Use of Plasma Arc Torch 2.0 Plasma Arc Torch 7 l 1

2.1 System Description 7 I 2.2 Coolant System 8 3.0 Fuel Configuration and Arrangement 10 l

3.1 Quantification of Batch 3 Fuel 10 l 3.2 Fuel Variation Within 11 }

LCSA/ Lower Head j 3.2.1 LCSA Region . 11 3.2.2 Lower Head Region 12 4.0 l Criticality Safety Analysis 13 4.1 Background 13 4.1.1 Criticality Report for the 13 )

Reactor Coolant System  !

! 4.1.2 Report on Limits of Foreign 14 j l Materials Allowed in the TMI-2 '

l Reactor Coolant System During  !

Defueling Activities l 4.1.3 Conservatism Inherent in 15 i Previous Analyses 4.2 Base Case Model 16 1 4.2.1 Geometrical Considerations 16 4.2.2 Fuel Model 18

! 4.2.2.1 Fuel Enrichment 18 8 4.2.2.2 Fuel Burnup. Worth- 19 <

4.,7 2.3 Lattice Structure 20  !

l 4.2.3 Unborar,ed Coolant Volume 23 4.2.4 Conservatism 24 i 4.2.5 Base Case Model Results 25 4.2.5.1 Optimization 25 4.2.5.2 Base Case 25

• - "~~

• _ _ _ _ _ - - _ - _ _ _ - - _ _ _ - _ - _ - - - _ . - - - _ _ - - - - -- - - - - - - - - - -- J

15737-2-N09-004

4. 3' 'Quantification of Conservatism /26 4.3.1 Hydraulic Mixing Modelling 26 4.3.2 Effects of Stainless Steel 30 4.3.3 Effects of Impurities 29 4.3.4 Results 30 4.4 Benchmarking 31 5.0 Conclusions ~ 31 5.1 Operational Limitations .31 l

l 6.0 References 33 Rev. 1-L_____________________--_____________

15737-2-N09-004 1.0 Introduction

1.1 Background

As part of the effort to dismantle the lower core support assembly (LCSA), including the flow distributor head, a plasma are torch may be utilized. Adequate cooling of this torch during cutting operations is provided via a closed, circulating coolant systcm having an inventory of less than four (4) gallons. Testing of the torch has shown that Reactor Coolant System (RCS) grade or B-10 enriched borated water cannot he used as the coolant due to the high electrical conductivities of these fluids. Further testing has determined that the use of demineralized (i.e.,

unborated) water results in acceptable torch operation.

Consequently, unborated water will be used as the cooling fluid for the torch.

l 1.2 Purpose As the unborated water inventory in the torch coolant system exceeds the two (2) gallon limit established in Reference 1, it is the purpose of this report to demonstrate that the plasma arc torch can be used to cut the LCSA without causing a criticality safety concern ,

within the reactor vessel. I 1

1.3 Scope The evaluation presented in this report addresses the use of the plasms arc torch to cut the LCSA, including the flow distributor head. The use of the plasma are torch for j other purposes should not be considered bounded by this evaluation and will be addressed separately, on an as-required basis.  ;

l i

~6- Rev. 1

_]

= 1 1

15737-2-N09-004 i

1

'1 1.4 Criterion for Justification for Use of Flasma Arc Torch 1 4 The criterion used to establish the acceptability of using the plasma are torch for' dismantling the LCSA was that the I RCS neutron multiplication (k,ff) would not exceed 0.99 for all credible situations during torch usage. 'This j acceptance criterion is consistent with the previous licensing basis for the RCS during defueling'(References 1  ;

and 2). 1 2.0 Plasma Arc Torch i I

2.1 System Description

.1 A plasma arc torch has been developed for use in the '

dismantling of the lower core support assembly. The plasma are torch is a direct current, tungsten j electrode, metal burning device. An initial pilot are i I

will ionize the primary gas, nitrogen, to form a plasma j jet which will be focused on the material to be cut. )

The plasma stream reaches temperatures of approximately )

20,000-50,000 degrees F, and thus melts the material at which it in directed. A secondary gas, also nitrogen, l is used to aid'in flushing away the molten metal from  !

the cut and to provide thermal insulation for the torch i

head. A lo;. secondary gas purge flow (~5 scfm) will be provided whenever the torch is under water and not performing cutting operations. This flow is maintained j to keep the torch tip dry. A simplified schematic of j the plasma torch is given in rigure 1. l l

I The Automated Cutting Equipment System (ACES) will be used to position the plasma are torch. The controls of ,

the ACES consist of two computer electron 3c systems,.cne for the plasma process control, the other for position j Rev. 1 l

l

-~

15737-2-N09-004 control. The plasma process control selects, adjusts and sequences current, gas and coolant flows. The position controller operates a five axis servo motor driven system that por.itions the torch, thus controlling the location and speed of the torch. Computer software allows a torch trajectory to be preprogrammed and executed automatically upon command. The torch is limited in its range by the physical restrictions of the l

tracks on which it rides. This physical restriction prevents the torch from impacting on the reactor pressure boundary. The equipment however has the capability to perform cuts at any fuel assembly location if required to support lower head or LCSA defueling. A j torque limitation on the motor devices prevent the torch from being driven into, and embedded within, a j significant accumulation of fuel debris.

The plasma are torch will cut electrically conductive materials, such as stainless steel structures. As the fuel debris is mainly ceramic, which is not electrically conductive, prior to the cutting of any particular piece of stainless steel within the LCSA, any significant quantities of surrounding fuel debris will be removed.

2.2 Coolant System Because of the high operating temperatures of the torch, the metal components of the torch must be adequately cooled. This cooling is accomplished via a circulating water system (-4.5 gpm system flow rate). The cooling system consists of a standpipe, three water-to-air heat exchangers, a pump and associated hoses and fittings.

The standpipe, heat exchangers and pump are provided in a separate unit (HE-200) which will be located cn the north end canal platform. The maximum total water Rev. 1 l

4 15737-2-N09-004 1 inventory of the coolant system is less than 4.0 gallons.

l A schematic of the coolant system is provided in Figure 2. To ensu2e acceptable operating ,

characteristics of the plasma arc torch, the conductivity of the cooling fluid must be maintained below ~15 micro mhos. Whenever the conductivity exceeds this value, the process of starting the arc fails.

Testing of the torch with water of various boron concentrations (RCS grade down to -200 ppm) showed that the electrical conductivity of these fluids was too high< Th!!s , it was concluded that demineralized (unborated) water, starting with a conductivity of about 2 micro mhos, was the mort ideal cooling fluid that could be used successfully.

Periodic checking cf the fluid conductivity, flushing of the system and recharging the system with new l

demineralized water will also be required. The flushing tieain is shown in Figure 2. When a torch tip is damaged, the torch will he renoved from the reactor vessel for repairs. The cooling system will also be flushed to return the con @uctivity of the coolant to acceptab3e levels. Referring to Figurs 2, flushing is initiated by disconnecting the return line from the HE-200, and routing the line to the deep end of the canal. The HE-200 pump is switched on and the remaining coolant is discharged to the deep end of the canal.

Occasionally, the three heat exchangers will be flushed in a sisilar ranner. The system is then recharged by initiating a gravity flow from the fifteen gallon demineralized water tank, which is located on elevation 347'-6". The torch coolant system is then fully charged to a maximum capacity not to axceed four gallons. The torch is then reinstalled into the vessel.

Rev. 1

15737-2-N09-004 4 i

Flushing of the coolant system may also occur when the conductivity of the coolant becomes unacceptably high. '

In this eace the torch will either be located in its home position (i.e., two inches above the top of the grid plate) or be removed frev the vnsuul. If the torch is in its home position and a disthnee greater than one foot exiirts between the torch tip and the debris bed surface, the torch coolant system cay be flushed in a manner similar to that described above except that the torch will remain its home position. If flushing is performed with the torch in the vessel, no load handling activities are allowed in or over the reactor vessel during flushing, thus minimizing the potential for damage to the flush system. Based on t1:e mixing analysis for the torch coolant system, discussed in Section 4.3.1, from which it was concluded that mixing will occur rapidly, any inadvertent leakage of the coolant during system flushing will adequately mix with j the borated vessel water so as not to pose a criticality i

safety concern. If the one foot separation is not i available or if load handling activities cannot be suspended, the torch will be removed trom the vessel prior to flushing.

3.0 Fuel Configuration and Arrangement l

The original loading of the core included 56 assemblies of 1.98% (batch 1), 61 assemblies of 2.64% (batch 2) and j 235 60 assemblies of 2.96% (batch 3) U enrichment. The  !

loading pattern is shown in Figure 3. Eased on this ]

loading, the initial core average enrichment was 2.54%. l 3.1 Quantification of Batch 3 Fuel )

l Early visual inspections and sonar mapping of the l core indicated a significant number of the batch 3 Rev. 1 I l

i I

'15737-2-N09-004 fuel assemblies at the core periphery were still j standing. Some of them were full. length, while a-large number of these assemblies were less than full cross-section. Some of these assemblies were knocked down, cut and trimmed and then leaded into canisters during earlier defueling activities. Nonetheless, it has been determined that the total length of batch 3 fuel assemblies that remained standing at the time of-the beginning of fuel assembly stub removal.

corresponded to approximately 50% of the initial batch 3 fuel. It is expected that these assemblies were removed reasonably intact, with little fixing of this batch 3 fuel with other debris within the vessel. Thus it is expected that the majority (-75%)

of the initial batch 3 fuel will be removed from the reactor vessel prior to the deployLent of the plasma are torch. Consequently any fuel remaining in the-vessel should consist mainly of batches 1 aind 2 fuel.

3.2 Fuel Variation Within LCSA/ Lower Head 3.2.1 LCSA Region There are two types of matorial within the LCSA region.

In areas within the 30 inch radius fror. the reactor centerlino the material observed during the core Stratification and Sampling Program was granular debris and drilling shards. Most of this material is expected to have been generated by the core drilling operations.

Referring to Figure 3, this material is expected to be mostly of the lower two batch enrichments as the core drilling operations were limited to the central regions of the reactor vessel. The other type of material observsd in the LCSA is columns of material that resolidified as it was flowing downward from the core Rev. 1

15737-2-N09-004 region during the accident. Based on the information known to date, this material appears to be concentrated at the periphery of the LCSA. It is possible that this )

material is not from batch 3.and hence would have an i enrichment less than 2.96%. It is, however, plausible that portions of this material could have enrichments greater than that of batch 2 (2.64% unburned).

l 3.2.2 Lower Head Region i I

The material recently relocated to the lower head due to j the core drilling operations is expected to consist mostly {

of batches 1 and 2 fuel, for reasons similar to those discussed in Section 3.2.1. This applies to the fine vacuumable material that has been observed on the debris surface.

The rock-like material (up to -2 inch diameter) observed on the surface of the lower head debris during various inspections has been sampled and the results are provided in Tables 1 and 2 (Reference 9). Although credible enrichments as high as 2.6% were observed in the samples, there is significant variation across each sample. The '

average enrichment of all the samples was 2.3%, with the l average enrichment across any sample being no higher than 2.4%, except for sample 11-1-C. This sample had an average  ;

enrichment of 2.6%, but only 2 particles were analyzed.

l i The sub-surface material configuration is not fully understood at this time. It is, however, known that some l

l of this material consists of larger resolidified material, that most likely was relocated in liquid form, and a larger quantity of rock-like material. Having assessed the

! various possible mechanisms and points of relocation to the l

l l

Rev. 1

. l 1

1 15737-2-N09-004 i

I lower head, it is concluded that there are possibly large j i

chunks of material in the lower head with an enrichment greater than that of batch 2. However, it is improbable that the entire mass would have an enrichment naar this level.

l 4.0 Criticality Safety Analysis

4.1 Background

I 4.1.1 Criticality Report for the Reactor Coolant System l

j l

The Criticality Report for the Reactor Coolant System j (Reference 2) defined a boron concentration (i.e., I 4350 ppm) which would ensure that the RCS neutron ]

multiplication (k,ff) would not exceed 0.99 for all l credible configurations. In the model development.for l Reference 2, two conservative fuel models were considered.

These were the design basis model, also referred to as the lenticular model, and the spherical model. Since the lenticular model was three-dimensional, it was only 1

analyzed with the Monte-Carlo program KENO V.a (Reference 3). The spherical model however, because of radial symmetry, also allowed the use of the one-dimensional,  !

discrete-ordinates transport program XSDRNPM (Reference 3).

Comparisons between the lenticular model, with KENO V.a, I l

j and the spherical model, with XSDRNPM, showed that the i spherical model was slightly more reactive (i.e., ~0.3% 6k).

Additionally, comparative studies were made evaluating the I spherical model with both KENO V.a and XSDRNPM, which demonstrated that there was excellent agreement between the KENO V.a and XSDRNPM calculated results. For the RCS design basis model, the calculated value of k,ff was essentially 0.99, including a 2.5% Ak computer code 1 Rev. 1 L---------------------------------------------------

15737-2-N09-004-i 1

uncertainty bias, when the RCS boron concentration was set i at 4350 ppm (the minimum boron concentration allowed-by current technical specifications). To provide an adequate-operating margin, an administrative limit on the minimum operational RCS boron concentration was established at 4950 ppm.

1 4.1.2 Report on Limits of Foreign Materials Allowed in the THI-2 Reactor Coolant System During Defueling Activities ,

1' b j The 4350 ppm boron concentration established in Reference 2 j does not provide total protection against the potential j increase in the RCS neutron multiplication (k,gg) caused by the introduction of foreign materials into the RCS. In Reference 1, an evaluation was performed to assess.the' .

effects on the RCS reactivity which could be caused by such an introduction of foreign materials. In that evaluation, XSDRNPM was used to quantify the increases in k,gg.

XSDRNPM, rather than KENO V.a, was used since XSDRNPM results do not contain a statistical uncertainty and are therefore more amenable for the determination of small l reactivity effects. However, because of the geometrical limitations of XSDRNPM (i.e., one-dimensional), the ]

Reference 2 design basis (lenticular) model could not be used explicitly for the analyses. Instead, the one-dimensional spherical mods 1 was used.  !

l The conclusion of the Reference 1 evaluation was the j establishment of a two (2) gallon limit on the amount of unborated moderating material (i.e., a material that can become interstitially dispersed within the fuel) that can be introduced into the RCS such that k,gg will not exceed 0.99 for all credible situations. This' result was based on a RCS boron concentration of 4950 ppm, the lower Rev. 1

15737-2-N09-004 operational limit' permitted byLthe current-administrative

. procedures.

4.1.3 Conservatism Inherent inLPrevious Analyses The evaluations performed for both References l'and 2 i contained assumptions that were. considered overly conservative when applied to:the specific (activity of using the plasma arc torch to dismantle the LCSA. Consequently,-

the reduction of some of these. conservatism was considered.

necessary to allow the plasma arc torch analyses addressed in this evaluation to more realistically _model the conditions that will exist during the cutting of.the LCSA.

Justification for the reduction in these conservatism is addressed below, while specific assumptions used in'the i plasma arc torch analyses are addressed in later sections of this report.

First, the evaluations completed for both References 1 and

~

2 were performed with the intent that the results would be bounding during all credible situations that could'be encountered during the entire defueling process. No attempt was made to define assumptions for a'particular defueling activity or phase. However the' scope of this document limits the use of the plasma arc torch.(unless evaluated separately at a later date) to the cutting of the LCSA, including the flow distributor head (see Section-1.3). The assumptions and thus the criticality safety models developed for this evaluation, can be tailored to the specific activities and possible accident configurations associated with the cutting of the LCSA.

Second, at the time References 1 and'2 were developed, there was limited knowledge of the. spatial' distribution of fuel within the reactor vessel. However, current data available'from debris samplings, video inspections and Rev. 1-

15737-2-N09-004 defueling records, as well as a better understanding of the accident scenario, allow more realistic modelling of the fuel debris spatial distribution in the current criticality safety analyses. Finally, the previous analyses took credit for fuel burnup in the batch 3 fuel only. The rationale for this assumption was the small reactivity effect that was seen when the batches 1 and 2 fuel were added to the periphery of the batch 3 fuel.- Thus any credit for burnup of batches 1 and 2 fuel would essentially l have had a negligible effect on k,ff. This effect was l encountered since the previous analyses placed the entire initial inventory of the highest enriched, batch 3 fuel in the center of the fuel arrangement. However, with the placement of a smaller amount of batch 3 fuel in the central fuel region, as is done with the plasma are torch analyses, the reactivity worth of the other fuel batches increases. With the higher reactivity worth of the batches 1 and 2 fuel, the burnup worth of these fuel batches also becomes more important. Thus burnup of batches 1 and 2 fuel was included in the plasma arc torch criticality safety analyses.

4.2 Base Case Model 4.2.1 Geometrical Considerations Prior to the use of the plasma arc torch for cutting the LCSA, all significant fuel masses above the LCSA, with the exception of the fuel behind the core former plates, will be removed. However, the core former plates should prevent any significant quantities of this fuel from falling into a region in which the torch is operating. As the cutting of the various plates of ths LCSA proceeds, readily accessible debris will be removed. Also, the safety features inherent in the torch design prevent the torch from becoming Rev. I w____-_____-

15737-2-N09-004 i

embedded ~within the fuel debrisiduring normal operations

_(Section'2.1). Furthermore, as unborated water'is less dense than the borated water in:the reactor. vessel water,.

any coolant leakage would; tend to rise rather than sink-into the debris (Reference 5).

~

It has therefore been.

concluded, based on the above considerations,'that a bounding mechanism for the unborated water from the torch j l

coolant system to transport to the fuel debris would be for the water to intermix with the debris-pile at or near the surface'of the' debris accumulations. It is highly'unlikely 1 that any substantial amount of unborated water would deeply-intermix within the debris. It is concluded that the most likely geometry between the unborated coolant and the debris would be where the unborated-water forms a layer on, or slightly penetrates.into, the debris bed. However, for' conservatism, it was. assumed that'the entire volume of unborated water would be modelled'as totally submerged within the fuel. To maximize reactivity effects, the unborated region was placed in the center _(i.e., most

]

reactive location) of the fuel model. As it is assentiallyL impossible to make accurate predictions of the shape of any 4 I

unborated water region, a spherical configuration, which 1 minimizes the surface area to volume ratio, was assumed.

Based on the above discussion, the model shown in Figure 4 was developed for this safety evaluation. This model should be considered the plasma arc torch base case model.

The smaller sphere in Figure ~4 represents the unborated water mixing with fuel debris. The size of this region was-determined based on the volume (i.e., 3.0 gallons, see section 4.2.3) of unborated water that was assumed to: leak into the vessel. The outer and larger spherical fuel l volume is determined based on the balance of the initial .

fuel inventory, optimally moderated with the borated (i.e.,  !

4950 ppa) vessel water. The two' spheres are then l

Rev. 1

b 15737-2-N09-004 surrounded by a thickness of borated' water representing.an )

infinite reflector layer.

4.2.2 Fuel Model 4.2.2.1 Fuel Enrichment i I

J As was seen in Figure 4, two separate fuel zones'were considered in the plasma arc torch model, a smaller sphere comprised of unborated water and fuel,.and a q larger sphere containing borated water and fuel.- Due to the relatively small size of the inner sphere, it was recognized that unborated coolant could interact with a 1

suali localized fuel region. Thus an assessment was j performed to determine whether it was possible for the l coolant to leak:into any region of the vessel in which-significant quantities of batch 3 fuel (i.e., the J highest enriched) could potentially be located. Based on the current damage assessments, the initial core loading pattern and-the proposed plasma arc torch usage, it was found that torch usage could potentially occur'in the vicinity of batch 3 fuel. Consequently,'for conservatism, the enrichment of fuel that'was assumed to intermix with the unborated water in the smaller sphere of the base case model was that corresponding to burned batch 3 fuel.

As was noted previously (Section 3.1), prior to dismantling the LCSA most of the batch 3 fuel will have been removed from the reactor vessel. Additionally, most of the fuel in the LCSA is expected to be. fuel fines generated from the core drilling operation. As the core drilling operation was limited to the center 1of the core, whereas the batch 3 fuel.is on the periphery, it is not expected that much of the fuel within the LCSA Rev. 1 La___-_-___-___ _

15737-2-N09-004 will be batch 3. However, for conservatism, the fuel in the larger sphere was assumed to be a homogeneous mixture (core average) of the three fuel batches'. Using the initial enrichment and number of assemblies as shown in Figure 3, the core average unburned mixture enrichment was determined to be 2.54%.

4.2.2.2 Fuel Burnup Worth j In Reference 1 and 2, the additional reduction to k m resulting from including burnup of batches 1 and 2 fuel was conservatively neglected. However as has been discussed in Section 4.1.3, this burnup credit was considered for the plasma are torch analyses. To determine the burnup effect for batches 1 and 2 fuel, a procedure similar to that previously used for batch 3 fuel was adopted. A detailed discussion of this procedure is given in Reference 6.

Incorporation of burnup effects in the batches 1 and 2 fuel resulted in a net U 35 enrichment of 2.24% for the average fuel (i.e., the homogeneous mix of the three fuel enrichments). This enrichment is supported as conservative by the er.richment data determined from available fuel debris sample data (presented in Section 3.2.2), based on the following considerations:

o the majority of the batch 3 fuel has been removed from the vessel o debris in the LCSA is expected to be primarily batches 1 and 2 fuel (in relatively equal amounts) based on the initial loading patterns, earlier defueling activities and the current understanding of the accident scenario 1

Rev. 1

15737-2-N09-004 o average initial enrichment (i.e., without burnup) of batches 1 and 2 fuel is 2.31%

o the apparent lack of batch 1 fuel in the' sample data (i.e., only one data point has an enrichment lower than the initial batch 1 enrichment of 1.96%)

4.2.2.3 Lattice Structure As with the previous criticality safety analyses (References 1 and 2), the fuel was represented as a homogeneous medium for which the neutronic data corresponds to a dodecahedral lattice structure of spherically shaped fuel pellets. Whereas the References 1 and 2 analyses limited the maximum size of the fuel particle to the equivalent of a standard fuel pellet, for the plasma are torch analyses of this report, the l presence of melted fuel and thus larger pellets was considered.

Based on the relatively small size of the inner sphere,

it was assumed that the entire fuel mass that would mix with the unborated water would consist of batch 3 fuel.

According to the most recent damage assessments fuel melting was not initiated in any batch 3 fuel. Rather l

any dissolution of batch 3 fuel occurred as a result of melted batch 1 and 2 fuel flowing past the batch 3 fuel rods. Consequently, the fuel in any particles that are I larger than standard pellets is highly unlikely to be solely batch 3. If batch 3 fuel is present in the large particles, it is likely to be mixed with batches 1 and 2 j fuel. Additionally, based on the available sample data 1

Rev. 1 i

I 15737-2-N09-004 it is unlikely that any large fuel particles would be pure U02 S me f the impurities present in the sampled debris are quantified in Table 1. The effects of these impurities are evaluated in Section 4.3.3. Nontheless, to maximize the reactivity effects of fuel melting, an ,

i' optimum (i.e., most reactive) fuel particle size of pure UO 2 was used to represent the fuel within this region.

The optimum fuel particle size was determined by performing an extensive series of lattice cell calculations in which the particle size and fuel volume i fraction were varied until a most reactive particle j size and volume fraction combination was found. The j dodecahedral unit cell of the previous criticality analyses, spherical fuel particles surrounded by water, was utilized for these calculations. The 27-group END/B-IV Cross Section Library was applied in the SCALE ,

system (Reference 3) to provide resonance-shielded (NITAWL-S module) and cell-weighted (XSDRNPM-S) cross  !

sections. The optimum particle size was determined to have a diameter of 2.1 cm for the unborated region.

Similarly, a series of lattice cell calculations were performed to determine the optimum fuel volume fraction  ;

for core average fuel mixed with borated (4950 ppm) water. However the use of an optimum particle size for the outer fuel zone was considered unnecessarily conservative for the plasma are torch analyses. This conclusion is based in part on recent core damage assessments. These assessments indicate that a large percentage (->60%) of the debris in the LCSA/ lower head is either fines (less than pellet size) or large fused masses (greater than approximately 20 cm diameter).

Additionally, it is unlikely that any melted fuel particles would be pure UO as is assumed in the 2

optimization calculations. Furthermore as discussed in Rev. 1 l

l _

15737-2-N09-004-l Section 3.1, as mostiof batch.3 fuel will be' removed. l q

from the vessel. prior to plasma torch usage, this fuel region will consist mostly of batches 1-and-2 fuel. 1 Finally, the pure UO ' fuel Particle size range in whichi 2

the kco value. exceeds that for standard pellets is somewhat narrow. This is demonstrated by theydata presented in. Figure'5. Figure'5'provides'the-relative I relationship between kco and fuel particle' size for two different fuel enrichments (2.96%,.2.34%) (Reference 6). . j Although the boron 1 concentration used to develop the.

data for Figure 5 was 4350 ppm,.the general conclusions derived from this curve should not change for the boron-concentration of interest in this analysis (i.e.,

4950 ppm). Backup for this_ assumption is provided by j the 3.6 cm diameter optimum fuel ~ particle size shown in Table 3 for a 4950 ppm boron concentration.

An optimal fuel volume faction was utilized for'each of the different particle size calculations provided in the Figure 5. Pure UO2 particles were assumed for the analysis. The koo values presented in the figure were i 1

normalized to the koo value at the spherical diameter i corresponding to standard pellets (1.07 cm)'. This normalization was performed for the two enrichments analyr.ed. A review of the-figure shows that optimally moderated particles with diameters in the narrow range j of greater than the equivalent of standard pellets to.

less than.approximately 10 cm will=have"a kec; value-that exceeds the koc> value for standard pellets..

Consideration for the presence of impurities in the melted fuel along with the use of actual fuel volume fractions would. result in decreased values of k so for the melted fuel. (See Section 4.3.3)

Based on the arguments presented in the above paragraphs, it is concluded that the use of fuel Rev. 1

15737-2-N09-004 particles of a size corresponding'to the equivalent of standard pellets would be an appropriately conservative representation of fuel in the-outer fuel. zone.

4.2.3 Unborated coolant volume The' maximum.unborated coolantL inventory in the plasma  !

are torch cooling system.is, by design, less than four-(4) gallons. However, based on the physical characteristics'of the coolant system (e.g.,' system.is' vented to atmosphere),-it'is hydraulically impossible for the entire inventory to drain followingLa line break or torch tip blowout, whenever the torch is operating'in the reactor vessel.

To evaluate the maximum amount of unborated coolant leakage that would occur during torch operation, a drain down test was performed. In this. test, the system pump i

was permitted to operate throughout the duration of the.

test. In reality, a float switch (disabled in test) would shut off the pump on a low inventory level. The measured leakage from the test'was approximately'3.45 gallons. As the test was performed with-the hoses in air (to assist in measuring leakage quantity), this volume was reduced by the amount of the coolant inventory that would not drain because the torch will actually be immersed-in the reactor vessel

(-0.47 gallons). Thus, the maximum amount of-unborated.  ;

water that will drain from the-torch coolant system during torch operations is limited to less than 3.0 l gallons. This volume'(i.e., 3.0 gallons) was used as the volume of unborated water in the smaller. sphere of the base case model.

l 1

Rev. 1 l

- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ .__ _ _ - _ - . _ _ _ _ _ _ - __ - --- -- - - - _ _ - - - _ __ a

a1

'l

.)

15737-2-N09-004 4.2.4 conservatism. ,

I In the development of the plasma arc torch base case-criticality safety model, conservative assumptions were utilized. These conservatism include:

1 o no credit for presence of steel plates in LCSA-o no credit for.large amounts of structural or solid poison materials existing in debris (See Table.1) o optimized fuel particle size in unborated-fuel l region j o optimized fuel / moderator ratio in all fuel regions o no credit for mixing of unborated cooling water with borated vessel water 1 o minimum allowable boron concentration'of 4950 ppm is assumed in borated regions of model o unborated water region is placed in most reactive q configuration (center of fuel model)

Quantification of the reactivity worth of some of these conservatism is provided in Section 4.3.

It is recognized that isolated regions within the debris bed may have average enrichments that are greater or particle sizes that may be more reactive than'those used in the large sphere of the base case model. HowcVer, I considering the base case model'as a whole, including'the  !

inherent conservatism as outlined above, it is concluded that the base case model is a conservative representation of any. credible configuration that could be experienced j during use of the plasma arc torch to dismantle the LCSA, l and thus is appropriate for use in this evaluation.  !

Rev. 1

.I.

C___________________.__________._________..__.._ . _ _ _ _ _ . _ _ . _ _ _ _ . . _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . . _ _ . _ _ _ _ .- _________.__...__._._-__._____-_m___ _ _ - _ _ _ - _ _ _

15737-2-N09-004 )

i 4.2.5 Bas.o Case Model Results 4.2.5.1 Optimization An extensive series of calculations were performed to-determine the optimum fuel particle size and  !

corresponding optimum fuel-volume. fraction:for thel various boron concentrations of interest.. The results of these calculations-are given in Table-3. During preliminary investigations it was found.that the optimum size and volume fraction.were mainly a-function of boron level and that a change in enrichment had little affect  !

on these parameters. Consequently,' optimization was not performed at every combination of; enrichment and boron concentration, but rather at one enrichment-for each boron level of interest. j I

4.2.5.2 Base Cass.

The results of the base case model using both XSDRNPM and KENO V.a were provided by Reference 10. Using-XSSRNPM to an analyze the base ~ case model, k,gg_was determined to be 0.9582. Typically XSDRNPM analyses-are performed to provide added confidence to the values predicted using KENO V.a. Generally the results predicted using the two codes for TMI-2 criticality safety analyses have agreed well. Howeverl the first KENO V.c. run performed using the base case model predicted k,gg to be 0.9663 10.0010. The agreement between the results was not as good as that experienced in previous criticality safety analyses:for TMI-2.

Because of this difference, the analysis was further investigated. As a result of the invcatigation it was concluded that the difference was most likely due to'the statistical nature of KENO V.a (amplified by the presence of an unborated central region in the model Rev.'1

c j

. t j 15737-2-N09-004 i

geometry). To confirs,this conclusion, nine' additional ENO V.a runs were.made with the. result reported-in q Table'4 (0.9599 i 0.0011) being the mean value of-[thejl0

~

j runs. The results of'all the RNO runs-ara providedLin Table 5.. Neither the U NO V.a nor-the X5DRNPM>results stated above include an analytical. uncertainty bias.

Applying a 2.5% Ak bias-(see section 4.4)Lthe E NO V.a'-

result becomes 0.9849.' This value is considerad to'be the base case result. This result meets the acceptance criterion as outlined in Section 1.4.

4.3 Quantification of Conservatism To quantify the effects on k,gg as a result of some of the I conservatism inherent in the base case model, as described in Section 4.2.4, additional analyses were performed. 1 These analyses are provided-to demonstrate that there.is a.

large degree of conservatism in the base case model.

l 4.3.1 Hydraulic Mixing Modelling' An evaluation was performed to determine the extent of a local boron dilution, rather than a local boron displacement as assumed-in Reference-1, resulting from a.

postulated break in the placaba are torch cooling hoses' or from a blown torch tip. The entrainment of the unborated j cooling water was calculated using empirical correlations j for the mixing of turbulent water jets into large quiescent j water systems (Reference 4). No credit was taken for the )i other mixing mechanisms that would be present (e.g., i turbulence created by the. torch operation,-the gas purge,-

interaction with debris or.the normal vessel convection .

currents). Based on the correlations'from Reference 4, , ~

which indicate that mixing is essentially a function of.the l break area, an analysis was performed to determine:the-average boron concentration of the fluid entrained in.the- .

I Rev. l' ,

I

)

1 19737-2-N09-004  !

I jet at various distances from the postulate'd cooling line break, as a function of the assumed break area (see Figure 6). This analysis was performed for two types of breaks: (a) a circular break with both hose. ends discharging, and (b) a slot break. The results of this )

analysis showed that mixing occurred less quickly with  !

larger break areas. Thus,.the mixing rates for the maximum break areas were used to incorporate the mixing phenomena J into a criticality safety model. I I

i To assess the effect that mixing would have on_k,ff it was ]

arbitrarily decided that the boron concentration for the j analysis would be defined using two mixing regions. The f

limiting boron concentration between the two regions was also arbitrarily selected to be 2000 ppm. Based on the

)

ce, d h a at d ume, tw c h -

average boron concentration increased to a level above 2000 ppm for (a) a full area guillotine break of the ,

coolant hose and, (b) a 1.0 square inch riot break. This I is the first step in the process to include mixing in the criticality safety model. A 1.0 square inch slot break was .

considered to be the maximum credible size c.onsidering the '

l hose used, the planned operating procedures, and fluid conditions existing withirt the hose. The full area break bounds all other credible breaks including a torch tip j blowout. The arbitrary selection of the 2000 ppm boron

]

level was appropriate since the sole purpose of the j l analysis was to demonstrate that mixing would occur rapidly and that there is a larger degree of conservatism l I

associated with neglecting mixing in the base case model. j l

In both scenarios considered, mixing occurred very rapidly  !

with the volume of water, both borated and unborated, l entrained in the jet, prior to the 2000 ppm distance (See Rev. 1 I

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . _ _ . _ _ ')

15737-2-N09-004 Figure 6) being less than 0.25 gallons. For use in the criticality safety model this volume was assumed to be-0.25 gallons of unborated (0 ppm) water. This water was then optimally mixed with optimally sized batch 3 fuel particles in the innermost region of theLaodel (See.

Figure 7). Next, a region containing 4.61 gallons of.

2000 ppm borated water was optimally mixed'with'the optimal batch 3 fuel particles. The 4.61 gallons was used to simulate the mixing of the additional 2.75 gallons of-unborated coolant with 1.86 gallons borated vessel l (4950 ppm) water. Outside this region was'the balance of the full ccre fuel inventory, optimally mixed with:the burned core average fuel described in Section 4.2.2.

Finally, an infinite borated water reflector was placed external to the fuel regions. Conservatism inherent in

- this hydraulic mixing model include:

o the effect caused by unborated water being less dense than borated water and thus tending to rise, rather

,. than sink into the fuel, has been neglected

• o all water in the jet with boron concentrations < 2000 ppm was assumed to be unborated o all water in the jet with boron concentrations 2 2000 ppm was assumed to have a 2000 ppm concentration o significant mixing mechanisms were neglected (turbulence created by torch operation, gas purge, interaction with debris and normal vessel convaction currents) o

~

unborated water placed in center (highly reactive) location of model Rev. 1 L___ _ _ _ _ . _ _ . . _ _ _ _ _ _ _

i 15737-2-N09-004:

i A more elaborate criticality. safety model, in which there were more fuel regions, containing a finer beron concentration distribution, could have also' born used.

This model should result'in even a smaller. calculated k,ff -

value. i 4.3.2 Effects of Stainless Steel j l

l Stainless steel occupies a large portion of the' volume within the LCSA region of>the reactor vessel. All' steel .j has been conservatively neglected in the 'evelopment d of the j base case model. The largest piece of steel within the l LCSA, the grid forging, was used as the basis for a model.

developed to assess the reactivity worth of this stainless' steel. The grid forging is a steel plate, approximately 13.5 inches thick, drilled with approximately 6.5 inch.

diameter holes in a' lattice as shown in Figure 8.- j A model of the grid forging was developed to perform the stainless steel sensitivity analysis, however each of the holes was assumed to be only (6) inches in diameter.

Additionally, the size of the grid forging'was assumed to

]

be infinite in the radial direction and fourteen (14) ]

inches high axially. Each hole was assumed filled with an )

~

optimum mixture of unborated water and fuel.-'The fuel-used  !

in this' case was optimum sized fuel' particles, with an enrichment of 2.3%. On the top and bottom of the steel was j an infinite thickness of borated water reflector. -i l

It is recognized that the dimensions used in.this criticality saftty analysis differ slightly-from the actual- i grid forging dimensions. However,. based on the extremely l low values of k,gg seen for this analysis (see section. )

4.3.4), the effects of thase differences will not affect' j the overall conclusion that the presence of significant l l

Rev. 1 )

l i

q

.. , , ,cl q

'15737-2-N09-004L .:

'i a

amounts of stainless steel has a nagative-effect'on the neutron ,

q multiplication.- g i

'J 4.3.3 Effects of Impurities To assess the reduction in reactivity due to the ,

presence of impurities in the melted fuel, another

]

series of lattice. cell calculations were performed.

these calculatier.s the average impurities identified in:

In.

]

Table 1 were assumed to be' mixed.with optimally sized,. 1 i

1 burned batch 3 fuel. 'Unboratied water.was used-as a b fi moderating material. .The mixture particleLsize andLfuel d l

volume fraction were varied until a ma d um?k g value.- .

was determined. The potential'depleti w if' neutron I poisons (e.g., Blo) .was not i::onsidered in~ t'h is analysis.

4.3.4 Results j l

~

The renults of these additional cases are given in Table 4. The main conclusion to be drawn from'this 4 table is that there is a large degree of conservatism associated with the base case model; For example, the l results of the simplified mixing-modelishow a nor.inal' k,gg of 0.924 i 0.001. This corresponds.to approximately a 3.6%4 k reduction from the base case

'l analysis. Additionally by virtue of the extremely low-

.l calculated value of k,gg, the results1of the stainless' .l steel model indicate that'there is a'large negative j effect on k,ff when credit is taken for..the significant quantities of stainless' steel that are present in the  :

LCSA. The effects of impurities.on the reactivity of ..

l the melted fuel can be seen by a comparison of the= l optimum k , value for pure UOg and the optimum value considering the impurities. The'value consideringLthe .

impurities was less than 0.8, while the pure UO2Na ""8' ]

, 1 j

30 R e v . : 1<. 1 I- j

J 15737-2-N09-004

.I 1.37. The negative reactivity effect of the. impurities I would be decreased if the' potential depletion of the neutron poisons were considered. Additionally, the l relative worth of the impurities would decrease if l ;

borated water were assumed to be the moderating material. Nevertheless, the calculated difference in the kco values demonstrates the conservatism associated )

with neglecting the presence of inpurities in the ]

melted fuel.

l In conclusion, the results of these additional analyses l demonstrate that there is a large degree of conservatism associated with the. base case model. l i

4.4 Benchmarking

]

1 l

.l i

.In Reference 2, an analytical uncertainty bias of 2.5%A k, l including the KENO V.a statistical uncertainty, was established as )

j an appropriate value for the borated systems being investigated in that report. Uncertainty values reported in the literature for unborated systems have been shown to be somewhat lower than this valva. Consequently, the 2.5% A k value is considered conservative l

for the plasma are torch criticality safety analyses provided in  !

l this report. This bias is considered applicable for both KENO V.a j and XSDRNPM analyses since previous analyses (Reference 2), as well l as Table 4 demonstrate the good agreement between the results j l generated by these codes.

l l

l 5.0 conclusions l

l Based on the evaluation presented in this report, it is concluded that the plasma arc torch, with a maximum coolant system inventory l of four (4) gallons of unborated water, can be used to dismantle the LCSA, including the elliptical flow distributor head, without developing a criticality safety concern within the reactor vensel.

31 Rev. 1

. - _ _ _ - _ _ _ _ _ _ _ - ~

15737-2-N09-004' -!

5.1 Operational Limitations j i I The above conclusion is based on the following operational limitations: l q

o The plasma arc torch will'only be1used to cut the LCSA.

i o All standing fuel. assemblies must be removed from the co e-region prior to.the use of plasma' arc torch in the' reactor vessel.

o A maximum of four (4) gallons of unborated water is  ;

permitted in the plasma arc torch coolant system with a  !

system configuration such that a maximum of three (3)  ;

t' gallons can drain following a line rupture or torch tip blowout with the torch operating in the reactor vessel. i o Following the loss of coolant inventory, the torch must be removed and repaired before refilling the torch cooling i system. l j

o If in-vessel flushing of the torch is being performed, no  :

load handling operations (heavy or light) are permitted in i or ebove the reactor vessel.

l o Flushing of the plasma arc torch coolant system with the torch within the vessel can only occur if there are no known leaks in the coolant system, the torch is in its home position, there is at least_a one-foot separation between the torch tip and significant debris quantities,.

and the gas purge is operating. Otherwise, the. torch must  !

be removed from vessel prior to connection of the flushing tie-in.

o The maximum inventory cf unborated water permitted in the flush system storage tank ils fifteen (15) gallons. '

32 nev. 1

i 15737u?-N09-004 i

6.0 References j l

j

1. Report on Limits of Foreign Materials Allowed in the TMI-2 Reactor Coolant System During Defueling Activities, Rev. 1, 15737-2-N09-002, September 1985.

l

2. Criticality Report for the Reactor Coolant System, Rev. O, i 1

15737-2-N09-001, October 1984. ]

I

3. SCALE: A Modular Code System for Performing Standardized j Computer Analyses for Licensing Evaluations, Vols. 1-3, NUREG/CR-0209, Nnclear Engineering Applications Department- (1982, Revised June 1983, December 1984).

4. Perry's Chemical Engineering Handbook, 6th Edition, 1984, i

5. Hazards Analysis: Potential for Boron Dilution of the Reactor l Coolant System, Rev. 2, September 1985.

6. TMI Criticality Studies: Lower Vessel Rubble and Analytical Benchmarking, GEND-071, May 1986.

t

7. Letter, R. M. Westfall (ORNL) to D. S. Williams (GPUN), August 11, 1987.  !

l

8. Letter, R. M. Westfall (ORNL) to D. S. Williams (GPUN), August l 24, 1987. ]

9. TMI-2 Technical Bulletin 85-21, Rev. 4, " Additional Results of Lower Head Debris Samples"

10. Letter, C. V. Parks (OkNL) to D. S. Williams (GPUN), November 6, 1987.

11. Letter, C. v. Parks (ORNL) to D. S. Williams (GPUN) , November 12, 1987.

Rev. 1

15737-2-N09-004 i

l 1

1 TABLE 1 i AYERAGE ELEMENTAL CONCENTRATIDW OF THE LOWER HEAD DEBRIS j Core Component / Particle / Concentration (wtt)

~

{

ETesient , 7-1-B 11 A l l C 11-2-C ll-4-B 11-4-D 11-5-C 11-6-B ll-7-C' 1 J

Fuel U 65.3 66.8 64.2 63.2 65.1 65.6 64.0 69.5 62.3 I Zr 12.0 12.8 13.0 12.9 15.0 12.1 12.2 11.7 12.6 Sn -* -* -* . -a ..a ..a ..a ..a ..c Control Rod Ag 0.22b ..a ..a .a ..a ..a ..a ..a ..a Cd --a --a --a 0.025b --a --a --a --a 0.065b In ..a ..a ..a ..a ..a ..a ..e ..a ..a Burnable Poison Rod A1 J -C -C .C -C -C -C -C -C B 0.094 0.12 ..a 0.071 0.065 0.12b 0.077 0.36 0.096 Gd ..a ..a ..a ..a ..a ..a ..a ..a ..a

$tructural Material Fe 2.4 1.88 2.28 2,48 2.90 3.70 2.04 1.83 2.41  :

Cr 0.95 0.59 0.6 0.79 0.99 0.96 0.65 0.58 0.61

--a Ni --a 0.24b --a 0.26b --a 0.20b ..a 0.24b J Mn --a 0.057 --s 0.068 0.085 0.089 0.065b 0.068 0.062b Hb ..a ..a ..a ..a ..a ..a ..a ..a ..a l 51 ..f ..c ..c ..c ..c ..c ..e ..c ..c )

1 Mo 0.14 0.19 0.12 0.093 0.13 0.21 6.11 0.21 0.12 I Cu 0.46 ..a ..a 0.21 -.a ..a .a ..a ..a l

Total wt% of '

sample d 81 82 80 80 84 84 81 84 80

a. Below detectable concentrations.

b. Some concentrations for this particle were below the detection limit. .

l They have not been included in the listed value.

c. Results are not included as the samples were contaminated with these elements during dissolution or handling.  ;

d. The remaining percer.tage is currently attributed to oxygen by the laboratory.

Further analyses continues.

, 34 Rev.,1

i 15737-2-No9-004

(.

\

4 T1.81.E 2

[_

U-235 ENRICle4ENT OF THE LOWER HEAD DESRIS SAMPLES Sample Number /U235 Enrichment (wt 1) (b) 3 4 5 ,6 7 8 9 Particle (=) ,1 2 .

7-1 -8 2.2 2.3 2.3 ' 2.4 2.6' 11 A 2.0 1.8 11 C 2.3 2.9 (c) ,

11 C 2.4 2.6 2.2 2.2 2.6 2.5

. 11-4-B 2.2 2,4 2.3 2.6' '

11 D 2.6 2.3 2.2 11-5-C 2.2 2.1 2.3 2.2 2.2 3.1(c) -- 2.2 2.2 11-6-8 2.5 2.3 2.4 11-7-C 2.6 2.5 2.1 l

6) The first number indicates locations the samples were taken from. The second number is a sequential sample number. The letters signify subdivisions of the sample. The samples numbers identify particles taken from each sub-division for analysis. Locations 7 and 11 are on the south and southwest sides of the reactor, respectively. -

(b) fypical uncertainty associated with results is about + 10%. ,

(c) 1.arge associated uncertainty.

~

35 E*V' 'l

15737-2-N09-004 Table 3: Optimiza': ion Results Fuel Particle Boron Diameter Optimum Fuel Concentration Enrichment (cm) Volume Fraction (ppm) (%) k oo -

3.0 0.67 4950 2.57 0.9675 3.2 0.67 4950 2.57 0.9679 3.4 0.68 4950 2.57 0.9681 3.5 0.68 4950 2.57. 0.9682 3.6 0.68 4950 2.57 0.9682(a) 3.7 0.68 4950 '2.57 0.9682 3.8 0.68 4950 2.57 0.9681 2.5 0.54 2000 2.67 1.1150 2.7 0.55 2000 2.67 1.1154 l 2.8 0.55 2000 2.67 1.1155(a) 2.9 0 56 2000 2.67 1.1155 3.0 0.56 2000 2.67 1.1155 2.0 0.33 0 2.57 1.3722 2.1 0.33 0 2.57 1.3724(a) 2.2 0.33 0 2.57 1,3723 2.4 0.34 0 2.57 1.3720 (a) optimum values for noted boren concentrations l

l l

36 Rev. 1

15737-2-N09-004 Table 4:-Results i

Case (a) Computer Code k,fg knax ID) -

~

1 Base Case KENO V.a 0.9599 10.0011 IC) ~0.9849

'XSDRNPM 0.9582(d) 0.9832 j l

Mixing Case KENO.V.a 0.924 1 0.001 0.949 Stainless Steel KENO V.a 0.794 i 0.002- 0 819 (a) See Sections 4.2 and 4.3 for descriptions of cases.

(b) k max = k,ff + 2.5% 6k uncertainty bias (c) Value is the mean of 10 KENO V.a runs, each using approximately 200,000 histories.

(d) XSDRNPM convergence is 10 ~4 s

I i

1 I

I 37 Rev. 1 1

-15737-2-N09-004 Table 5: Results of KENO V a Analyses for Ease ~ Case Model Run I")

k,gf (c) 1 0.9663'i 0.0010

-2 0.95SO i 0.0011 3 0.9612 1 0.0011 -

4 0.9648 1 0.0011 5 '0.9608 i O.0012 i

6 0.9574 i 0.0013 7 0.9584 1 0.0011 8 "

0.9576 1 0.0012 9 0.9566 i 0.0011-10 0.9610 1 0.0011 Mean 0.9599 1 0.0011 ID)

(a) Cases dit'fer only in change in- random' number. There were 200,000 histories per case.

(b) Mean = k =' = 0.9599 (c) Results do not include any uncertainty bias.

1 i

l 1

1 j

l 38 Rev. 1 ,

j

15737-2-N09-004'

. . l

'1 1

e STAtteNo 8L85mA GA5 G'E j J

l l '

l I

e-. PLAlwaramC i - r.:

Cun4 TNT SCus cs j i,

P

.N N

\ g N TWNCETEN PetCT AAC ILECTaCD4 NON FREG

% N "' CURRENT g s sounct j N N .

j Nj '

i f k l

\ N N N N N -_

i N N , Vl#?iCAL , l g D#*vtMOTOR '

' I N N l A N ,

q g . AllWwfM . . CCNIROL j W Darvg weten PANEL r  ;

, TC4cw a A0tak ,.

• ClitsoNtm l t \ -

I

\ -

weexpites 1

PLASMA ARC CUTTING 5YSTEM SC'W. MAT 7C Figure 1 1

1 39 Rev. 1 i i-

.'}.,l.u.<.:

. . :' ~ N09-004 15 mu.es

< > N" s a sys, on

-Oc

< > aw'.c tamanL WwrEA wh# _

jo 1.p. Mc6E W

W g #g$/g* WS. Ac7tcas4T W MS.4Jup 1

l 1

~ '~~"-

, i ev g TTWs'~~'

, 7r ,

• 5 = > w~

l g l eut sa  %

l

• so.m Pws 1

i lE4/h. Ol lr _ __

ndcar i j

8 J i L ER.  ;

)

= = __ _ _;

I

__s#PLv astvan posta 88 L 818 99w., Pestr 70 8 99#ety vaut M.ge, g ,. ' _

./ % me saww f I8 Route Return Line pyg,g pes,. ass MC L9.

gg gg To Deep End Canal m When Flushing gg j 17 MWs8E N

W 'finnad W U 1

Figure 2 Plasma Arc Torch Coolant System Schematic 40 Rev. 1 C____ Z-_ _ - --- _ * ~ ~' ' ' ~ ' ' ' ~ ~ ~ ~

1

.i - m Figure 3 TMI 2 CORE ENRICHMENT PATTERN A B C D E F****** G H K L M N O P R {

.  : I i  !  ! l f V V V 7 i ! !

15 *

! 5 5 5 l: AV A WA WA AA h * *  !

".' Ifj> [.

c .'.'., c.

1 -

..., r

• r '

.'..~. y 83;  :  : i

• 'I '- *' " " t 13 .

. V ,. 22 :. :..- .

k..:.,ss::"-.. e"' . ' .i ::: . ' . .- :..:.

e nio .

.~c., .. %. . ' . ~ . 'f l

.~.~.' c~L

• M ' .~. ~ . ' , . ***..', f r ,j

,4%. .. -'. * . * . . ,,.,., ,.! . * * * . * , .f .*.*.*<

$.e.[$w$

...  :. l jg

. /;

f .::

', ,(y,,_(

y.~

m w

e,. r ,

>,*m.

s. n,.v

.n ... .n,4 , -  :*.. : .. 2

• j

. ' . ~ . - g,y,4;y: c. .~. g> .~ic, ug .~..<

• y/

.~.~.' w

...-7/r

-.4::~ ' . */

l W,l,+A: ::: o  ::: y,, p,a3l ?.*.*, '

fr L4** 'bliW4'* Wo .v siOY4'*

.*: Qg@+i A 11 * $45 *4% %k *O*l t$ 4: *- , i wm .::: ..- c. . a. ...

W. j/j,.

1 l . . .- ,

1  :::  ;-  :::

10 ,/ m,v.s, a t.e. . Q,

.a.i.,$_< %W.:..::. / .::

.: ... . . .m.

"  ?:

• M,s,;:f 6 *: *-#.,g,W;.e 9

.. ;:* W..y^t,a :  :*. w 9- ,,/ m.+ . ..

MF~...'.

.w .

. @~ ..:...

. 9,,, . .:

. .. ~. ....

M.km@,K.:'.

, ... .'u*WA

... \

[*[*[%,[

8- 7 J hhi((*[*k{*[*[*% )%M {*((*g  :<4

[j 7-Y/. 5ted! NACO *$$5+

.:.:. 9;.w..:.n.e ;; .. .-

W4* %;maY::O:'. . ..

bEO::x.Q ..

O*

4.....::.Mik  ::: &~

... AC.7W y, n .:.*.*

. .m. .x ..., ,

~. . ....

4,..e ,y,/

c. , ..

5

/ ,~ ::.*

,f: y'.a**@

- ~ /

6--/ w$ri T

M~W/

N O*f L $.]  :@*.*

. Qg;40 j/ r/j hH  ?::. " ^'y?:. *. ' -::. QM Wit ..* :.v o~ +.

...:. v *::.

...:., h$v%;pe.

.//

...:,f 6 %M,y)a  ?::. .

/

j:'y:;'.,,

a-? .~.

ff '*.=:.

. ,S iti *::. M *::. %g:% *

. :. * < W>;i .. , <

( .

v

?::. . j,[*:::. .

' n *::.. *::.<

~ y .,* ,ff"

? f )

A  ? >

Q* # :n 4 ?::.

' <(? ,- =. :.::.'

? ::::.< f) l / *.*::.' . N. n ?::. .

a::.- * .. f,f y

~

,7 Jf7/ - . .. . , h'Mya . ...

4

) ....  ?. . . ( .

3 h <O:< _ .. t <O:V :O:.Q M

// A':0: <hw.H.:

/ /) ?m.hhfj h:w'I y) y)

1

' y y w K

1 V' fg 1 ENRICHMENT No. OF ASSEMBUES  :

l 1.98% 56 O:*

i

' 61 i 2.64 % 4 E,. 2.96 % 60 l

41 Rev. 1 i

)

,- ,*c 15737-2-N09-004 J o Unborated Water (3 gallens) i

~

o Burned Batch 3 Fuel (2.6'7%)

o Optimum Fuel Volume Fra tion (VF=0.33) o Optimum Fuel Particle ire (d=2.1 cm) o r = 15.9 cm o Borated Water (4950 ppm) o Core Average E ichment with Burnup (2.24%)

o Standard Pellet ize Fuel Particles ,d=1.07cm) I o Optimum Fuel Volu Fraction (VF=0.66) Infinite Borated o r = 150.1 cm Water Reflector Figure 4: Base Case LCSA' Criticality Safety Model

___-_-___._____-___-__----_-.n____-___=___-________-____-___________--_--_-_-

.. ..~

15737-2-N09-004

.;_ 3A R.. C .. S Z

_ . L

_ _3..S.

BORON-4350 PPW 1.03 1.025 -

+.

l 1.02 - +

y 1.015 -

+

0 W +

0 J 1.01-I +

r 0

2 1.005 - +

1-i i

0.995-l i

0.99 i i i i i i i 1 3 5 7 -9 SPHERICAL. DIAWETER (cm)  ;

U(2.96)02 + U(2.34)02' .l 1

Figure 5

.]

1

, .:e 'i 15737-2-N09-004  !

1 l

.s 4

4950 ppm water -

t t

mixing region 0 ppm water Y

(2000 ppm . ) 2000 ppm (torch cooling water (assumed to be0 ppm) (assumed to be 2000 ppm)

/

t

~l break location 2000 (boron concentration = 2000 ppm) i

\

i l

~I Figure 6: Hydraulic Mixing Modelling l l

i 44 Rev. 1- l I

15737-2-N09-004 p

4 i

. 1 j

3:

Infinite borated water reflector I L

'2 o Unborated Water

-(0.25 gal) 1 o Burned Batih 3 fuel 12.67%).

o Borated water ~

~7.0 cm (4.61 gal, 2000 ppm) oo *3 opt=imum Fuel Particle o Burned Batch Size (d=2.1 cm) 3 fuel ( .67%) o 0ptimum Fuel Volume or=y 21. cm Fraction (VF=0.33) o Optimum Fuel Particle Size (d .8 cm) o Borated Water (4950 ppm )

o Optimum Fuel Volume Fraction o Fuel Enrichment 2.24i (VF=0.55) o r 3= 150.1_cm o Standard Fuel Pa'rticle Size (d=1.07 cm) .

o Optimum Fuel Volume 'i Fraction (VF=0.66)

Figure 7:

Criticality Safety Model'to Assess Effects of Mixing I

l l

45 Rev. II d i

.. j

_.___-__-__._--__--______.-_-___:____-----_--. - - __ - . _ _ . - - - _ . _ . _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ J

,,as

. 15737-2-N09-004 i

1 l

l i

l l

L o Unborated Water d \ o Unburned Actuni Dimensions r I Enrichment = 2 57%

d = 6.5 inches '

J p = 2.1 inches .

P Stainless Steel Analysis perfomed using infinite lattice of above arrangement  !

with d = 8.0 inches, p = 2 inches l l

Figure 8: Criticality Safety Model Considering Presence of Stainless Steel within LCSA j

'l i

l I

46 Rev. 1 l

_ ___o