ML16279A520

Rhode Island Atomic Energy Commission Tenth Response to the April 13, 2010, Request for Additional Information Regarding License Renewal (Redacted)

ML16279A520
Person / Time
Site:	Rhode Island Atomic Energy Commission
Issue date:	07/15/2011
From:	Rhode Island Nuclear Science Center
To:	Office of Nuclear Reactor Regulation
Boyle P
References
Download: ML16279A520 (68)

Text

RHODE ISLAND ATOMIC ENERGY COMMISSION RESEARCH REACTOR LICENSE NO. R-95 DOCKET NO. 50-193 RESPONSES TO NRC STAFF REQUEST FOR ADDITIONAL INFORMATION FOR LICENSE RENEWAL REVIEW REDACTED VERSION*

SECURITY-RELATED INFORMATION REMOVED

• REDACTED TEXT AND FIGURES BLACKED OUR OR DENOTED BY BRACKETS

4.0 W Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 100413 RAI Tenth Response 4.14 Table 4.5 gives a maximum total power peaking factor of 3.06 for grid position D6. Explain how this power peaking factor accounts for localized power peaking that could be caused by a flooded experiment located in or adjacent to the core. (See RAI 14.65)

All of the neutronics analysis was done with the assumption that the central flux trap was filled with water. Consequently, power peaking under this condition is already taken into account.

10.2 Section 10.2.2 discusses administrative controls in place to limit draining of the reactor pool via the through-port. The text states that the through-port should not be opened for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> following reactor shutdown. The LOCA analysis presented in Section 13.2.3 of the SAR does not analyze pool drainage through the open through-port, nor does it provide a comparison of pool drain time for a closed and open through-port. Provide justification for the statement that opening the through-port 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after reactor shutdown is conservative.

The LOCA Analysis presented in section 13.2.3 of the SAR has been revised and is entitled "Section 13.2.3 Loss of Coolant Accident (LOCA) of the Safety Analysis Report of the Rhode Island Nuclear Science Center Reactor, Submitted May 3, 2004". In this analysis, the LOCA model is one in which an eight inch beam port extension is sheared off, and water drains through six sharp edged round half inch diameter round holes in the pool wall. It was conservatively assumed that since the drain lines of all of the beam ports are tied together by a common drain line, it would be possible for the drain line to back up and allow the un-damaged beam ports to fill with water, in which case each beam port would act as a drain path to confinement. Administratively, the area of each beam port that is open to confinement has been limited to a one half inch diameter hole. Consequently, the drain model considered a system which has six, half inch diameter holes, which corresponds to one for each beam port. The through port was not considered.

In response to this RAI question, the drain model for this analysis has been refined to include the through port experimental facility, and to provide a more realistic and less conservative representation of the experimental facility drain piping system. This analysis is entitled "LOCA Analysis Addendum. In this analysis, it is shown that each experimental facility drain, including the through port has a one half inch diameter orifice plate welded into the drain line. In each case, this empties into a one inch diameter drain line. All of these drain lines empty into a common two inch drain line that is at an elevation below all of the experimental facilities. The common drain line empties into a five inch line, which is reduced back into a two inch line that opens to atmosphere.

1

I , ,

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 Experiment Drain System Since the drain line diameter gets progressively larger, and it opens to atmosphere, it is not possible for the drain line to get backed up. This fact, coupled with the fact that the elevation of the common drain line is below the lowest elevation of any of the experimental facilities means that for the Design Basis Accident in which one beam port is sheared off, there are only two drain paths for the coolant water:

1. Damaged Port Drain

2. Area of Port Open to Confinement Shearing off a through port is not considered to be credible because there is virtually no access to the through port from the top of the pool. As shown in the photograph below, this port runs across the back wall of the pool, underneath the thermal column extension:

2

El Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 Consequently, the drain model in the Addendum has been changed from a tank in which six, one half inch diameter holes are at the elevation of the bottom of an eight inch beam port, to a tank in which there are two, one half inch diameter drain holes at the centerline elevation of the common two inch drain line. Under this scenario, the amount of time that it takes for the reactor pool level to drop from the pool level scram set point, to the top of the grid box is 19.34 hours3.935185e-4 days <br />0.00944 hours <br />5.621693e-5 weeks <br />1.2937e-5 months <br />.

The analysis done in "Section 13.2.3 Loss of Coolant Accident (LOCA) of the Safety Analysis Report of the Rhode Island Nuclear Science Center Reactor, Submitted May 3, 2004" shows that if the pool level does not drop below the elevation of the bottom of the eight inch beam ports before the decay power fraction is 0.827% after infinite reactor operation, then the fuel cladding will not be damaged. For the drain model in the addendum, it is impossible for the pool level to drop below this point due to shearing an experimental port because the through port is inaccessible, and the eight inch ports have the lowest drain level.

The revised analysis from 2004 includes a reference that provides data for the amount of time that it takes for decay power to reach various power fractions under various operating histories (Stillman). The analysis in the addendum indicates that the amount of time that it takes for the power fraction to reach 0.827% after infinite reactor operation is 162332 seconds (4.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />).

3

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 All of this means that it takes 4.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> for the decay power to reach the point at which the fuel cladding will not become damaged, given that the pool level is not below the elevation of the bottom of the eight inch beam ports. If a Design Basis Accident occurs, it will take 19.34 hours3.935185e-4 days <br />0.00944 hours <br />5.621693e-5 weeks <br />1.2937e-5 months <br /> for the pool level to drain to the top of the grid box, which is well above the elevation of the bottom of the eight inch beam ports. As a result, the power fraction will decay to a harmless level by the time that the pool level reaches the top of the core box. As part of the addendum analysis, the maximum allowable drain area between an experimental port and confinement was determined to be 1.48 in2 .

The through port is at a lower elevation than the beam ports. As stated earlier, a catastrophic failure of the through port is not considered to be credible. This is why the Maximum Credible Accident for the facility has always been the catastrophic failure of a beam port rather than the through port. Therefore, if the through port were to develop a leak, the pool drain time would be no greater than the time calculated in the LOCA Addendum analysis, provided that the area open 2

between confinement and the through port is no greater than 1.48 in2.

Section 2.3.3 of the Safeguards Report for the Rhode Island Open Pool Reactor (4 April 1962) indicates that the original criteria for using the through port was that both ends of the port would have gate valves that could be closed in the event of a leak. Like the beam ports, the through port also has a one inch diameter drain line that can be used to isolate the port from the experimental drain system. Consequently, provided that the gate valves are in place, it is possible to close off all of the potential pool water drain pathways associated with this experimental facility.

Since the drain time estimation in the LOCA Addendum is on the order of hours for the pool level to reach the top of the core box, there is sufficient time to perform mitigating actions. In the event that it were deemed to be worthwhile to reopen the through port after it has been isolated due to a leak, it would be possible to move the core to the opposite end of the pool, and isolate that section of the pool from the end with the through port so that the high power section of the pool with the ports could be drained independently form the low power section of the pool where the core is positioned. This action makes the twelve hour delay time prior to opening the port, irrelevant.

13.11 The calculation of pool drain time in Section 13.2.3 makes assumptions about the design of and administrative controls for use of the beam ports and through-port. Propose TS requirements for the design and operation of the beam ports and through-port that are consistent with the assumptions made in the analysis of a LOCA, or provide justification for not including such TS requirements.

The Addendum LOCA analysis shows that as long as the area between each individual experimental port and confinement is no greater than 1.48 in2 , then 4

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 there is sufficient pool drain time to allow for decay power to reach the point at which the fuel cladding cannot be compromised. However, this assumes that the water level will not drop below the elevation of the bottom of the eight inch beam ports. The elevation of the bottom of the through port is below the elevation of the bottom of the eight inch beam port, and an analysis for a LOCA in which the fuel is completely un submerged has not been performed. The answer to RAI question 10.2 shows that administrative controls on the use of the through port will prevent conditions from occurring that could lead to a LOCA that has not been analyzed. Therefore, the administrative controls will be set conservatively to say that:

1. Each beam port shall have no more than an area of 1.25 in2 open to confinement during reactor operation.

2. When the reactor is in operation, the drain valve to the through port shall be closed.

3. When the through port is in use, gate valves shall be installed on the end(s) of the port that will be used for access.

4. When the through port is not being monitored for a leak condition, the ends of the port shall be closed.

The bases for these specifications will be that:

Specification 1:

The LOCA analysis shows that as long as the pool level does not drain through an area greater than 1.48 in2 to confinement, then there will be sufficient time for decay power to drop to a point which will not damage the fuel cladding, provided that the pool level does not drop below the elevation of the bottom of the eight inch beam ports. It also shows that if any single port has a catastrophic failure, the un-damaged ports do not become pool drain pathways. Consequently, limiting the areas of each experimental port that is open to confinement to 1.25 in2 is conservative.

Specification 2:

The through port has three potential pool leak pathways. The first is the through port drain. By keeping this drain closed during operation, that potential leak pathway is blocked, and the potential for an unnoticed pool leak though this experimental facility is prevented..

5

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 Specification 3:

If the end(s) of the through port that will be used for access have gate valves mounted to them, then in the event of a leak, the port can be easily isolated so that the leak is stopped.

Specification 4:

The LOCA analysis has shown that the amount of time available for performing mitigating actions in the event of a pool leak is on the order of hours. Consequently, as long as reactor personnel will become aware of a pool leak though the through port reasonably quickly, and the gate valves are in place, the consequence of the leak can be mitigated quickly by closing the valves.

14.153 The second paragraph of the bases for TS 4.3.a appears to be a description of the pool level detection system, not the bases for the proposed surveillance requirements. In accordance with 10 CFR 50.36, provide bases that explain the reasons for the requirements of TS 4.3.a.4, 4.3.a.5, and 4.3.a.6.

Technical Specifications 3.3 and 4.3 have been re-written in order to make them more consistent with ANSI 15.1. The following table provides a summary of how the specifications have changed:

Original Specification New Location Location 3.3.a.1 Primary pH Note I 3.3.a.2 Primary Conductivity 3.3.1.1 3.3.a.3 Primary Radiological Analysis 3.3.1.2 3.3.b. I Secondary pH Note 2 3.3.b.2 Secondary Radiological Analysis 3.3.2 4.3.a. 1 Primary pH Surveillance Note I 4.3.a.2 Primary Conductivity Surveillance 4.3.1.1 4.3.a.3 Primary Radiological Analysis Surveillance 4.3.1.2 4.3.a.4 Pool Level Scram Test 4.2.6.6 4.3.a.5 Pool Inspection - Primary System Inspection 4.3.1.4 4.3.a.6 Pool Level Verification 4.3.1.3 4.3.b. I Secondary pH Surveillance Note 2 4.3.b.2 Secondary Radiological Analysis Surveillance 4.3.2.1 N/A Secondary System Inspection 4.3.2.2 Note I ANSI 15.1 recommends that either pH or Conductivity be monitored.

pH and conductivity are related, so monitoring them both is redundant. CO 2 dissolves into the pure pool water, which pushes the equilibrium pH down to an average of approximately 5.6.

Consequently, conductivity is a better measure of water quality.

6

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 RINSC proposes to use a conductivity measurement rather than pH to monitor water quality.

Note 2 The purpose of measuring pH and conductivity is to reduce activation products in the coolant, and to minimize corrosion. Activation products are not an issue on the secondary side of the cooling system because the coolant is not exposed to a neutron flux. Corrosion on the secondary side of the cooling system is no longer an issue because the aluminum piping has been replaced with PVC piping, and the cooling towers are made of non-corrosive materials as well. RINSC proposes to remove this surveillance.

3.3 Coolant Systems 3.3.1 Primary Coolant System 3.3.1.1 Primary Coolant Conductivity Applicability:

This specification applies to the primary coolant.

Objective:

The objective of this specification is to maintain the primary coolant in a condition that minimizes corrosion of the fuel cladding, core structural materials, and primary coolant system components, as well as to minimize activation products produced as a result of impurities in the coolant.

Specifications:

The primary coolant conductivity shall be < 2 gmho /

cm when averaged over a quarter of a year.

Bases:

Specification 3.3.1.1 is based on empirical data from the facility history. Over the lifetime of the facility, primary coolant conductivity has been maintained within the limit specified, and no corrosion on the fuel cladding, core structural materials, or primary coolant system components have been noted.

7

f.

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 3.3.1.2 Primary Coolant Activity Applicability:

This specification applies to the primary coolant.

Objective:

The objective of this specification is to provide a mechanism for detecting a potential fuel cladding leak.

Specification:

Cs-137 and 1-131 activity in the primary coolant shall be maintained at levels that are indistinguishable from background.

Basis:

Specification 3.3.1.2 provides a mechanism for detecting a potential fuel cladding leak by requiring that periodic primary coolant analysis be performed to test for the presence of Cs-137 or 1-131. These isotopes are prominent fission products. Consequently, if either of these isotopes are detected in the primary coolant, it may indicate a fuel cladding leak.

3.3.2 Secondary Coolant System Applicability:

This specification applies to the secondary coolant.

Objective:

The objective of this specification is to provide a mechanism for detecting a potential primary to secondary system leak.

Specifications:

Na-24 activity in the secondary coolant shall be maintained at levels that are indistinguishable from background.

8

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 Bases:

Specification 3.3.2.1 provides a mechanism for detecting a potential primary to secondary system leak by requiring that periodic secondary coolant analysis be performed to test for the presence of Na-24. This isotope is produced by the activation of the aluminum structural materials in the primary pool, and a small concentration of it is present in the primary coolant during, and immediately following operation of the reactor. If this isotope is found in the secondary coolant, it may indicate a primary to secondary system leak.

4.3 Coolant Systems 4.3.1 Primary Coolant System 4.3.1.1 Primary Coolant Conductivity Surveillance Applicability:

This specification applies to the surveillance of the primary coolant.

Objective:

The objective of this specification is to provide a periodic verification that the primary coolant conductivity is within prescribed limits.

Specification:

The conductivity of the primary coolant shall be tested monthly.

Basis:

Specification 4.3.1.1 requires that the conductivity of the primary coolant be tested on a monthly basis. ANSI 15.1 recommends that this be performed on a weekly to quarterly schedule. Specification 3.1.1.1 sets a limit on the average conductivity when averaged over one quarter of a year. Consequently, a monthly measurement falls within the ANSI recommended 9

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 schedule, and allows for a running average based on three data points per quarter.

4.3.1.2 Primary Coolant Activity Surveillance Applicability:

This specification applies to the surveillance of the primary coolant.

Objective:

The objective of this specification is to provide a periodic verification that the Cs-137 and 1-131 activity in the primary coolant is not significantly above background.

Specifications:

Cs-137 and 1-131 activity in the primary coolant shall be measured annually.

Basis:

Specification 4.3.1.2 requires that the Cs-137 and 1-131 activity in the primary coolant be tested on an annual basis. This schedule is consistent with the schedule recommended by ANSI 15.1.

4.3.1.3 Primary Coolant Level Inspection Surveillance Applicability:

This specification applies to the surveillance of the primary coolant.

Objective:

The objective of this specification is to ensure that the coolant level is at an adequate height above the core during reactor operation.

Specification:

The primary coolant level shall be verified to be greater than or equal to the Limiting Safety System Setting 10

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 value prior to the initial start-up each day that the reactor is started up from the shutdown condition.

Basis:

Specification 4.3.1.3 requires that the primary coolant level be inspected prior to the first reactor start-up of each day. A float switch system is used to monitor the pool level 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> per day, 7 days per week. This system is tied into the facility alarm system, which is monitored by an offsite alarm company. In the event that the pool level reaches one inch greater than the LSSS, the automatic pool fill is started. If the pool level drops to the LSSS, then a scram occurs, the operator receives an alarm, and the alarm company receives an alarm. A daily verification of the pool level prior to starting the reactor up provides adequate assurance that the float switch is working to maintain the pool level.

4.3.1.4 Primary Coolant System Inspection Surveillance Applicability:

This specification applies to the surveillance of the primary cooling system components.

Objective:

The objective of this specification is to provide a periodic verification that there are no obvious defects in any of the system components.

Specifications:

The components of the primary coolant system shall be inspected annually.

Basis:

Specification 4.3.1.3 requires that the primary coolant system be inspected on an annual basis. This schedule is consistent with the historical inspection schedule for the facility.

11

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 4.3.2 Secondary Coolant System 4.3.2.1 Secondary Coolant Activity Surveillance Applicability:

This specification applies to the surveillance of the secondary coolant.

Objective:

The objective of this specification is to provide a periodic verification that the Na-24 activity in the primary coolant is not significantly above background.

Specification:

Na-24 activity in the secondary coolant shall be measured annually.

Basis:

Specification 4.3.2.1 requires that the Na-24 activity in the primary coolant be tested on an annual basis. This schedule is consistent with the schedule recommended by ANSI 15.1.

4.3.2.2 Secondary Coolant System Inspection Surveillance Applicability:

This specification applies to the surveillance of the secondary cooling system components.

Objective:

The objective of this specification is to provide a periodic verification that there are no obvious defects in any of the system components.

Specification:

12

Rhode Island Nuclear Science Center Tenth Response to NRC Request for Additional Information Dated April 13, 2010 The components of the secondary coolant system shall be inspected annually.

Basis:

Specification 4.3.2.2 requires that the primary coolant system be inspected on an annual basis. This schedule is consistent with the historical inspection schedule for the facility.

14.154 The "Bases" section of TS 4.3.b appears to be a description of how secondary coolant chemistry is controlled and how secondary coolant radioactivity is monitored, not the bases for the proposed surveillance requirements. In accordance with 10 CFR 50.36, provide bases that explain the reasons for the requirements of TS 4.3.b.1 and 4.3.b.2.

Technical Specifications 3.3 and 4.3 have been re-written in order to make them more consistent with ANSI 15.1. See the answer to RAI question 14.153.

13

Rhode Island Nuclear Science Center Complete Set of Responses to the Nuclear Regulatory Commission Request for Additional Information Dated 13 April 2010

21 Chapter 2 of the SAR contains multiple section headings with no related information. Provide the omitted information.

Ninth Response Submitted February 24, 2011 2.3.1 General and Local Climate The average annual temperature in Rhode Island is 50'F (10°C). At Providence the temperature ranges from an average of 28'F (-2°C) in January to 73°F (23°C) in July. The record high temperature, 104'F (40'C), was registered in Providence on 2 August 1975; the record low, -

23'F (-31°C), at Kingston on 11 January 1942. In Providence, the average annual precipitation (1971-2000) was 46.5 in (118 cm); snowfall averages 37 in (94 cm) a year.

2.3.2.3 Humidity Rhode Island has a humid climate, with cold winters and short summers.

The humidity varies depending on wind direction and ocean temperature.

2.4.3 Sanitary Sewer System The Rhode Island Nuclear Science Center is connected to the Narragansett / South Kingstown municipal sewer system, which has a final outflow to the ocean.

2.4.4 Ground Water Ground water and site drainage flows directly into the Narragansett Bay.

The original site study report performed by General Electric cited this as being one of the reasons that the reactor was built at this site.

2.2 Section 2.4.4. Provide a discussion of potential impacts of the RINSC on groundwater, or the lack thereof, including the potential for neutron activation of groundwater, leakage from the reactor pool and primary coolant system, and leakage from contaminated water systems at the facility.

Second Response Submitted August 6, 2010 The major potential impact that the facility could have on groundwater arises from the fact that there is a small amount of tritium production in the reactor pool water. If the pool were to have a significant leak, this would be released to the ground or sewer system. The tritium concentration in the pool has been measured to be 3 X 10-4 pCi / cc. 10 CFR 20 Appendix B Table 3 indicates that the concentration limit for the release of tritium to the sewer system is I X 10-2 gCi / cc. Consequently, the concentration in the reactor pool is two orders of I

magnitude less than the release limit. This indicates that there is no significant potential facility impact on the groundwater.

4.1 Figures 4-5, 4-6, 4-7, and 4-8 were omitted from the SAR. Provide these figures.

Seventh Response Submitted December 14, 2010 The figures of Chapter 4, "Reactor Description", were incorrectly referenced throughout the chapter. The following table shows each reference, which page it is found on, the figure it references and the page they can be found, the corrected reference in an updated version of the SAR, and a description of the figure.

Original Text Page Original Figure Page Current Figure Page Description 4-1 4-3 4-1 4-25 4-1 4-4 14 Element Core 4-1 4-3 - - 4-2 4-5 Core Assembly 5-3 4-6 5-3 5-I1 4-3 4-9 Cutaway View of Flow Channels 4-1 4-7 4-4 4-28 4-4 4-11 Stan-up to Equilibrium Cores 4-6 4-7 4-4 4-28 4-4 4-11 Start-up to Equilibrium Cores 4-6 4-8 4-4 4-28 4-4 4-11 Start-up to Equilibrium Cores 4-7 4-14 - Hot Channel Fuel Surface Graph 4-8 4-16 - LEU core Flow vs DP Graph 4-I 4-21 4-2 4-26 4-5 4-26 17 Element Core 4 4-21 4-1,4-2 4-25,4-26 4-1.4-5 4-4,4-26 14,17 Element Cores 4-5 4-5 4-3 4-27 4-6 4-27 Power Peaking Factors in Thermal Hydraulic Calculations The "Hot Channel Fuel Surface Graph" and "LEU Core Flow vs DP Graph" figures are part of the new analysis being performed by Argonne National Labs.

See the updated analysis for this data. The "Power Peaking Factors in Thermal Hydraulic Calculations" figure is also outdated.

2

14 Fuel Element Core Figure 4-1 3

Reactor Core Assembly Figure 4-2 4

Cutaway view of Flow Channels Figure 4-3 5

Startup-to-Equilibrium Core Configurations with Remaining Uranium (grams of U235)

Figure 4-4 6

17 Fuel Element Core Design Figure 4-5 7

Power Peaking Factors in Thermal Hydraulic Calculations E D 0 E D 0 a G G G G 8

- -

0.62) 0,44 662*3 0.007 0830 0.807 1.781 1.168 1700 1." 1.1*1 1.766 1.586 1 .$; 1.16 1.536 1538 1G06 7 2192 1 57 2192 2.M* 1.727 2.2*

59% 6.6% __5._ 4A% 6^6 4J%

1011 1251 1011 11* 1.,11? 1.052 1.6*

1.8*1 1.,72 1630 15 1630 1an3 1M 1536 6 1.56 1.5C8 1.,1j 2M31 3SM 2531 2.105 .179 2.435 72% 9.0% 72%  % 7.7% 6.2%

1.285 1.236 1151 1.151 111 s 165A 5 1M S 1SO 1,5,6 1.5,6 2.925 2M26 3.122 $1,22

&2% 8.2% 7.j% 7.3%

1009 1256 1009 1.134 1.418 1.1*4 11.6 11M 1.832 4 1.331 1.500 1,53 1526 1 a3 16M 2530 3019 2.530 2.M7 1,256 2.770 9o0% 7.2 61,% .2% 6.7%

77%

0.818 091 0.018 Clow 1.110 0.191 1749 1312 1749 3 1..4 1410 1.654 15,6 103 1.5364 1.3*5 1.A8M 11.510 2.197 1 "S 217 2.83" 2.506 2.458 5.8% 6.9% So%

Rog, 0 2 Rod Rod 0.53 O-182 052M 2228 I*7) 2.228 1.686 1.563 1.536 G G G 1.600 11317 1.400 111% 3.55 &1%

Startup Core Startup Gore with 3 more Fuel Elinmer3 Radial Pealig factor lwiw psior9g facor Axial Oe6ia factor Total pialmg fator ci rofa toiw 0r in ellemera Figure 4-6 4.2 Section 4.2.1. Provide a summary description of the fuel development and qualification program for the fuel type used at the RINSC. The description should include the fuel characteristics and parameters important to safe operation of the reactor (e.g., power density, power rate change limits, bum up, etc.). Verify that 8

those parameters important to safety are included in or bounded by the requirements in the technical specifications (TS).

First Response Submitted June 10, 2010 Fuel qualification for reactor operation at 2 MW was established during the conversion from HEU fuel to LEU fuel. NUREG 1313 has a total review of the fuel qualification for research reactors, provided that the core in question meets the parameters under which the tests were done. These parameters were taken into consideration when the current safety analysis was performed. This qualification is valid up to 100% fuel bum-up.

4.3 Section 4.1 states that the fuel composition is U2Si2, while Section 13.2.1 states that the fuel composition is U3Si2. Clarify which composition is correct for the RNSC reactor fuel First Response Submitted June 10, 2010 U3Si2 is correct.

4.4 Section 4.2.2. Provide a summary description of the program for shim safety blade and regulating blade inspection and replacement.

First Response Submitted June 10, 2010 RINSC performs an annual inspection of the shim safety blades by raising each blade to it's full out position and making a visual inspection of each blade. Drive speeds and blade drop times are measured in order to verify that there is no indication that blade motion is hindered. An annual control rod calibration is performed to determine that the reactivity worth of the control blades provide sufficient shutdown margin to meet RINSC Technical Specification Limits. The current control blades have been in use since 1967. At present, there has been no indication that the control blade worth have been diminished significantly.

Consequently, there is no program or plan to replace control blades over the next twenty year licensing period.

4.5 Section 4.2.3 states that the graphite reflectors are designed for expansion "from an integrated flux of 2X10 2 1 nvt (expansion based on a more than two-year, full-power operation factor)." Given that the TS do not explicitly limit the duration of full-power operation, provide a discussion of the methods used to ensure that the graphite reflectors will not be exposed to an integrated neutron flux greater than the expansion design basis (e.g., calculation of integrated flux, surveillance programs, etc.). The discussion should include consideration of current integrated flux and integrated flux during the period of the renewed license.

9

Sixth Response Submitted December 7, 2010 The graphite reflector element is a block contained in a 3-inch square aluminum can with handles to allow remote handling. It should be noted that graphite has been used for many years as a reflector and crumbling or other catastrophic failures of graphite pieces have never occurred. AGOT reactor-grade graphite is used to avoid boron contamination. Graphite2 is known to undergo several changes when trace exposed to neutron irradiation:

• Dimensional change due to neutron induced swelling

• Elastic modulus change as measured by the impulse excitation technique

• Coefficient of thermal expansion change

" Thermal conductivity change

" Electrical resistivity change

• Irradiation-induced creep SAR Section 4.2.3 discusses the expansion of graphite due to irradiation and gas evolution. The design allows for a maximum increase in graphite dimensions of 1.1% due to irradiation growth and gas evolution. The SAR suggests that the graphite will be fine up to an integrated flux of 2 X 1021 neutrons /cm 2 . Our maximum flux at the center of the core is estimated to be 1013 nrcm 2 -s.

Therefore to reach 2 X 1021 nvt:

[2 X 1021 n/cm 2 ] / [1013 n / cm 2-s] = 2 X 108 seconds or about 5.6 X 104 hours0.0012 days <br />0.0289 hours <br />1.719577e-4 weeks <br />3.9572e-5 months <br /> of operation. At 7 hour8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />s/day operation, this amounts to 8,000 days of full-power operation or about 22 years of full-power operation. It should be noted that the graphite reflectors are now on the periphery rather than at the center of the core (LEU core configuration). The neutron flux at that location is at least an order of magnitude less than that at the center of the core. Thus, a conservative estimate of the time needed to reach the integrated flux would be 80,000 days or more of full-power operation. That amounts to about 219 years or well beyond the requested period of the renewed license.

2. Marsden, B.J., Preston, S. D., Wickham, A. J., "Evaluation of graphite safety issues for the British production piles at Windscale," IAEA-TECDOC-1043, September 1997.

4.6 Section 4.2.5. Describe the design characteristics of the reactor that ensure the control blades will fully insert despite motion of the core support structure (e.g.,

shaking of the core due to an earthquake). The response should include tolerances between the control blades and the control blade shrouds that prevent binding of the blades within the shrouds.

Second Response Submitted August 6, 2010 10

The core is suspended from a bridge that is mounted over the top of the reactor pool. General Electric Drawing 198E299 shows how the suspension frame holds the control rod housings and core grid box together. The reactor pool sits on a military gun pad. The pool is constructed of a large mass of reinforced concrete.

Consequently, in the event of an earthquake, the pool, bridge, and core are expected to move as a unit. The shim safety control blades fit inside a shroud, which is part of the core grid box. When the shim safety control blades are not fully inserted into the core, each blade is suspended by an electromagnet which holds it in its withdrawn position. When fully withdrawn, the ends of the blades remain inside the shroud, which prevents misalignment on release. A significant earthquake would likely shake the shim safety blades free from the magnets.

However, the reactor is fitted with a seismic scram device, which scrams the reactor upon detection of an earth tremor. General Electric Drawing 197E647 shows that the total spacing between the control blades and the shrouds is 0.125 inches.

4.7 Section 4.2.4. Provide a discussion of design features of the neutron startup sources that allow for reliable operation and replacement of the sources. The discussion should include calibrations, source checks, interlocks, and risk of damage to the sources. Include a discussion of any design features and/or administrative controls that reduce the potential for damage to the sources. The discussion should also describe whether improper operation or damage to the sources could potentially lead to instrument error or mislead reactor operators. If the potential exists for damage to the neutron startup sources from operation of the reactor, propose TS requirements to ensure there will be no damage to the sources, or provide justification for not having such TS requirements.

Second Response Submitted August 6, 2010 There are three neutron sources that are available for use as a start-up source.

The first is a pair of PuBe sources that are stored together in a common container, the second is an SbBe source, and the third consists of the Be reflectors in the core. The reactor Start-Up channel has a neutron count interlock of 3 cps, which is the minimum neutron count rate that must be present in the core in order to start the reactor up. Any one of the available neutron sources may be used as a start-up source, however given the typical reactor operating schedule, the Be reflector elements are generally used as the neutron start-up source. Gamma decay from fission fragments interact with the Be to produce a sufficient level of photo neutrons that the external sources of neutrons are generally not needed in order to have a neutron count rate of at least 3 cps in the core. It is not anticipated that there is risk of damage to the sources. The PuBe sources are leak tested every six months. The Be reflector elements are inspected as part of the fuel element inspection program.

4.8 Section 4.2.5. Provide justification for the design of the core support structure as to its ability to support the weight of the core and its ability to withstand radiation 11

damage, mechanical stress, and chemical degradation over the period of the renewed license.

Sixth Response Submitted December 7, 2010 Relative to commercial power reactors, the RINSC reactor operates at very low power, temperature, and pressure. Consequently, damage to the core support structure due to radiation exposure and thermal aging will be significantly less than the damage typical of power reactors. No reports of significant radiation damage to core components of small research reactors have been published.

Since the power industry does damage studies to show that their facilities can continue to operate safely with extended lifetimes, it is reasonable to assume that research reactors can safely operate within similar lifetimes.

The reactor core support consists of a suspension frame which is bolted to a moveable bridge, operated by a hand crank, which can relocate the entire core plus core support structure to various positions in the reactor pool. The four comers of the structure are occupied by the suspension posts. These comer posts connect the grid plate to the reactor bridge. The core suspension system includes a locating plate, made of heavy steel that spans the upper end of the suspension frame to provide support and location for the control blade drive mechanisms.

The control blade drive guide tubes are flanged to the bottom of this locating plate. Core elements are contained in a grid box that is enclosed on four sides to confine the flow of cooling water between elements (See Fig. 4-2 of the SAR).

The grid box assembly, including the drive mechanisms, is supported by the suspension frame. The elements that make up the core sit on a 7 x 9 grid plate with the four comer positions occupied by the suspension frame comer posts.

This core support system was designed to support the weight of the core plus control and cooling elements. The design has satisfactorily supported the weight of the components for over forty years and there is no credible reason why it should not continue to function as designed. No appreciable deterioration of any components of the support structure has been seen during inspections.

The core support structural materials are predominantly made of 606 l-T6 Al. In order to minimize corrosion of the aluminum, reactor pool water pH and conductivity (resistivity) levels are measured weekly to verify that the values are within the RINSC Technical Specification limits (pH between 5.5 and 7.5; resistivity greater than 500 kQ/cm). As described in Section 5.5.1 of the SAR, make-up water for the pool passes through a five micron filter, an activated charcoal filter, two mixed bed demineralizers and a one micron filter before entering the pool. The pH and conductivity are measured weekly to verify that the water in the pool is within the specification limits.

The core fuel cladding material is also 6061-T6 Al. We perform an annual fuel element inspection that would provide another indication of whether or not 12

aluminum core materials are beginning to suffer from corrosion, radiation damage or thermal stress.

4.9 Section 4.4. Discuss the ability of the biological shield and pool liner to continue to meet their design bases during the period of the renewed license. Include considerations of radiation, chemical, and thermal degradation. Describe any surveillance programs in place to detect degradation of the biological shield and pool liner.

Sixth Response Submitted December 7, 2010 Relative to commercial power reactors, the RINSC reactor operates at very low power, temperature, and pressure. Consequently, damage to the biological shield and the pool liner due to radiation exposure and thermal aging will be significantly less than the damage to similar structures typical of power reactors.

No reports of significant radiation damage to biological shields or pool liners of small research reactors have been published. Since the power industry does damage studies to show that their facilities can continue to operate safely with extended lifetimes, it is reasonable to assume that research reactors can safely operate within similar lifetimes.

The reactor pool is surrounded by thick (minimum 10 ft) reinforced concrete.

The inner surface of the concrete is lined with 1/4-in. thick aluminum. The liner is the primary pool water containment vessel. For the same reasons given in response to RAI 4.8, corrosion of the aluminum liner is expected to be minimal over the extended lifetime of the reactor. The same monitoring (pH and conductivity) used for the aluminum fuel clad would also alert operators to any corrosion of the liner. Required annual inspections of the pool supplement weekly water monitoring to confirm the integrity of the pool liner. Any significant degradation of the liner, whether from chemical or mechanical causes, that could lead to pool water leakage would be detected by routine monitoring of the makeup water system.

The combination of the water pool and the surrounding concrete provide a biological shield for facility workers that keeps the dose rate below I mrem/hr at all points above and outside the pool area (SAR, Section 4.4). Water level monitors and radiation monitors ensure that the water depth is sufficient (approximately 24 ft) to shield personnel near the top of the pool. During routine operations radiation surveys are performed to monitor dose rates (SAR Ch. 11).

The radiation attenuation properties of the pool water are based on the nuclear properties of the water and the attenuation level will not change over time as long as the water level is maintained. The aluminum liner is a minor contributor to the biological shielding relative to the water and concrete, but as explained above it is not expected to deteriorate over the lifetime of the facility. While concrete is susceptible to thermal and radiation damage, the low power and low 13

temperature of the RINSC reactor will not lead to any degradation of the concrete over the lifetime of the reactor.

A survey of aging effects on concrete was performed at the Idaho National Laboratory 3. According to this report, for conditions of radiation flux up to 2 x 1019 nvt (thermal) and temperatures to 120 'C, radiation damage to the type of concrete used in our facility was insignificant, while other types show considerable loss of strength (specifically high alumina cement concrete). All effects on concrete due to radiation, per se, were too slight to reliably measure because of the gross effects from the increased temperature during exposure.

Generally speaking, the threshold of degradation in the concrete is approximately 95 0 C.

The neutron flux at the core end of the beam ports (basically the inner surface of the concrete shielding) is approximately I to 4 x 1012 n/cm 2-s. Assuming 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> of operation per week, ten years of full-power operation would just approach the 1019 nvt threshold for the most susceptible type of concrete (not used at the RINSC). However, as noted in the referenced report, radiation damage is basically not measureable compared to temperature effects. The safety limit for the pool water temperature is 130 'F (54.4 °C). This is substantially below the 95 'C threshold value for thermal damage to the concrete. Based on these considerations it can be concluded that the biological shield will not deteriorate over the extended lifetime of the facility.

3. Literature Review of the Effects of Radiation and Temperature on the Aging of Concrete, INEELiEXT-04-02319, September 2004 4.10 Section 4.5. Describe and justify the methods used to determine the reactor kinetics parameters found in Table 4-3. Provide the names of any codes used and a description of the modeling process, if applicable.

Fourth Response Submitted September 8, 2010 The reactor kinetics parameters were recomputed for the equilibrium core using standard perturbation methods with Argonne's VARI3D computer code'.

VARI3D uses the DIF3D diffusion theory code to compute real and adjoint flux solutions for use in perturbation theory.

The kinetic parameters were recalculated here because the models that were used to calculate the values in Table 4-3 in the early 1990's are no longer available.

The table below shows that the recomputed values of the delayed neutron fraction and the prompt neutron generation time agree very well with those in Table 4-3.

14

Equilibrium Core Kinetic Parameters 2004 2010 Calculation Report With equilibrium No Xe Xe Delayed Neutron Fraction, f3-eff 0.764 0.755 0.756

(%)

Neutron Generation Time, pts 68.3 69.4 68.6 Delayed Parameters by Family Group 13(1) 2.6580E-04 2.6591E-04 P3(2) 1.3707E-03 1.3713E-03 P3(3) 1.3188E-03 1.3194E-03 P3(4) 2.8985E-03 2.8996E-03 P3(5) 1.1 990E-03 1.1995E-03 P3(6) 5.0074E-04 5.0095E-04 k(1) 1.3337E-02 1.3337E-02

?,(2) 3.2712E-02 3.2712E-02

)(3) 1.2075E-01 1.2075E-01 X(4) 3.0279E-01 3.0279E-01 k(5) 8.4966E-01 8.4966E-01 k(6) 2.8538E+00 2.8538E+00

'A.P. Olson to J. Matos, "Availability of the VAR13D Code on Linux", Argonne National Laboratory Internal Memorandum, February 11, 2004.

4.11 Section 4.5. Describe and justify the calculation methods for the coefficients of reactivity for temperature, void, and power. Discuss any measurements made to confirm the reactivity coefficients. Include estimates of accuracy for the coefficients.

Fourth Response Submitted September 8, 2010 Coefficients of reactivity for temperature, void, and power were recomputed for the equilibrium core for these RAI responses because the models and results in Section 5 were originally computed in the early 1990's and are no longer available.

Neutron cross sections in seven energy groups as functions of moderator temperature, fuel temperature, and coolant void fraction were prepared using the WIM/ANL cross section generation codel. Keff values were computed using the DIF3D diffusion theory code. Coefficients of reactivity were determined from these data.

15

RINSC does not make measurements of these parameters. The accepted values for these coefficients are the values provided by the model.

4.12 Table 4-3 lists coolant reactivity coefficients for the coolant temperature range of 20-40 degrees Celsius (degrees C). TS 3.2.1 allows operation of the reactor with coolant temperatures up to 126 degrees Fahrenheit (degrees F) (52 degrees C). Provide coolant reactivity coefficients over the entire coolant temperature range allowed by the TS.

Fourth Response Submitted September 8, 2010 Separate reactivity coefficients were recalculated for this RAI for increasing coolant temperature only and increasing coolant density only using the methods described in RAI 4.11. The reactivity coefficients are shown below for a coolant temperature range between 20 'C and 100 °C.

Equilibrium Fission Products Change Coolant Temperature Only Change Coolant Density Only Cumulative Cumulative Reactivity Reactivity Change x Water Change x Water Water 103 Temp. Reactivity, 103 Temp. Density Reactivity, Rel. to 20

°C dk/k Rel. to 20 'C 0C mg/ml dk/k °C 20 0.027602 0.00 20 0.99811 0.027249 0.00 30 0.026447 -1.1551 30 0.99564 0.026636 -0.6131 40 0.025298 -2.3044 40 0.99227 0.025881 -1.3678 50 0.024155 -3.4478 50 0.98810 0.024985 -2.2641 60 0.023017 -4.5855 60 0.98323 0.023947 -3.3020 70 0.021885 -5.7174 70 0.97773 0.022767 -4.4815 80 0.020759 -6.8434 80 0.97171 0.021446 -5.8026 90 0.019639 -7.9636 90 0.96525 0.019983 -7.2653 100 0.018524 -9.0781 100 0.95845 0.018379 -8.8696 16

Cumulative Reactivity Change as a Function of Coolant Temperature Only 0.0

-1.0 S-2.0

-3.0

-8.0

-9.0

-10.0 Coolant Temperature, C Cum ulative Reactivity Change Versus Coolant Tem perature Corresponding to Different Coolant Densities 0

.1

.2

-3

-4

.5

.6

.7

-8

-9

-10 Coolant Tem perature, C 4.13 Table 4-3 includes the Doppler coefficient of reactivity over the temperature range of 20-40 degrees Celsius. This temperature range appears to apply to the coolant temperature and not the fuel temperature. Provide the Doppler coefficient of reactivity over the range of fuel temperatures anticipated during all allowed modes of reactor operation and reactor transients.

Fourth Response Submitted September 8, 2010 Reactivity coefficients were recalculated for this RAI for increasing the fuel temperature only using the methods described in RAI 4.11. The reactivity coefficients are shown below for a fuel temperature range between 20 'C and 600 0C.

17

Equilibrium Fission Products Change Fuel Temperature Only Cumulative Reactivity Change x 103 Reactivity, Rel. to 20 Ak/k °C 20 0.027479 0.0 30 0.027280 -0.19854 40 0.027083 -0.39578 50 0.026887 -0.59172 60 0.026693 -0.78635 70 0.026499 -0.97968 80 0.026307 -1.17170 90 0.026117 -1.36243 100 0.025927 -1.55185 150 0.025000 -2.47940 200 0.024105 -3.37437 300 0.022412 -5.06657 400 0.020851 -6.62845 500 0.019419 -8.06001 600 0.018118 -9.36125 Cumulative Reactivity Change as a Function of Fuel Temperature 2

X 0 u-2

-4

-10 Fuel Temperature, C 4.14 Table 4.5 gives a maximum total power peaking factor of 3.06 for grid position D6. Explain how this power peaking factor accounts for localized power peaking that could be caused by a flooded experiment located in or adjacent to the core. (See RAI 14.65)

Tenth Response Submitted July 15, 2011 18

All of the neutronics analysis was done with the assumption that the central flux trap was filled with water. Consequently, power peaking under this condition is already taken into account.

4.15 Section 4.6 references the computer code PLTEMP as the code used to determine many of the thermal-hydraulic characteristics of the reactor core.

Provide a discussion of the use of this code including models of the RINSC core, applicability of the code to the thermal hydraulic conditions in the RINSC core, validation and benchmarking of the code, and code uncertainty. Provide a copy of Reference 4.6.

Fourth Response Submitted September 8, 2010 In Section 4.6 the original PLTEMP code' was used to determine many of the thermal-hydraulic characteristics of the reactor core. This code has been superseded by the PLTEMP/ANL V4.0 code. 2 The analysis of steady-state forced-convection operation has been redone using the newer code.

Although both the old and the newer code have models that can be used to obtain the flow distribution in a reactor core, a more direct, transparent, and tractable approach was taken in the new analysis. A hydraulics model of the RINSC core was developed based on engineering fundamentals. The equations used in the analysis are provided in the memo dated 3 September 2010 from Earl E.

Feldman to James E. Matos entitled "Steady State Thermal Hydraulic Analysis for Forced Convective Flow in the Rhode Island Nuclear Science Center (RINSC) Reactor". The application of the equations are explained in detail. Key intermediate results for a reactor flow rate of 1580 gpm are given in tables so that one can verify that the analysis was performed correctly. The hydraulics analysis yielded the flow rate for each fueled channel in the reactor as a function of total reactor flow rate.

In the next phase of the analysis, the individual channel flow rates obtained in the previous phase were used to perform a thermal analysis of the core. The limiting channel was identified as an internal channel in element D6 and next to the highest power fuel plate. The new PLTEMP code was used to perform thermal analysis of this limiting channel. The highest power fuel plate bounded by two channels, each with half of the flow area of single channel, was modeled.

This model is simple and easy to check. For the channel flow rate corresponding to a reactor flow rate of 1580 gpm the PLTEMP results for the powers at which the onset of nucleate boiling and the onset of flow instability were predicted to occur were verified by a hand calculation. Key intermediate results used in the verification are given in tables so that one can verify that the results are correct.

19

In the new analysis, key results such as coolant flow rates, bulk coolant and clad surface temperatures, the conditions that produce onset of nucleate boiling and the conditions that cause flow instability are either performed by a hand calculation or are verified by one. Thus, PLTEMP code validation, benchmarking, and code uncertainty are less relevant because results that are essential to demonstrate the safety of the RINSC core have been hand calculated or hand checked. Moreover, the PLTEMP/ANL code used in the new analysis is based on an evolutionary sequence of "PLTEMP" codes in use at ANL for the past 26 years. 39 Validation of PLTEMP/ANL has followed standard practice in any code development task, where comparisons are made with other codes, with measurements, and with hand calculations where possible. Many examples of validation are given in the PLTEMP/ANL V4.0 manual 2 , and in References 3-9.

The users guide to the PLTEMP/ANL V4.0 code. 2 is provided.

Also provided is Reference 4-6:

K. Mishima, K. Kanda and T. Shibata, "Thermal-hydraulic Analysis for Core Conversion to Use of Low-Enrichment-uranium Fuels in KUR," KURRI-TR-258 (1984).

References:

1. Kaichiro Mishima, Keiji Kanda, and Toshikazu Shibata, "Thermal-Hydraulic Analysis for Core Conversion to the Use of Low-Enriched Uranium Fuels in the KUR," Proceedings of the 1984 International Meeting on Reduced Enrichmentfor Research and Test Reactors, Argonne National Laboratory, October 15-18, 1984.

2. Ame P. Olson and Kalimullah, "A Users Guide to the PLTEMP/ANL V4.0 Code," Global Threat Reduction Initiative (GTRI) - Conversion Program, Nuclear Engineering Division, Argonne National Laboratory, March 24, 2010.

3. K. Mishima, K. Kanda and T. Shibata, "Thermal-Hydraulic Analysis for Core Conversion to the Use of Low-enrichment Uranium Fuels in the KUR,"

KURRI-TR-258, Research Reactor Institute, Kyoto University, Dec. 7, 1984.

4. K. Mishima, K. Kanda and T. Shibata, "Thermal-Hydraulic Analysis for Core Conversion to the Use of Low-Enriched Uranium Fuels in the KUR,"

ANL/RERTR/TM-6, CONF-8410173, p. 375, 1984.

5. W. L. Woodruff and K. Mishima, "Neutronics and Thermal-Hydraulics Analysis of KUHFR," ANL/RERTR/TM-3, CONF-801144, p. 579, 1980.

6. W. L. Woodruff, "Some Neutronics and Thermal-hydraulics Codes for Reactor Analysis Using Personal Computers," Proc. Int. Mtg. on Reduced Enrichment for Research and Test Reactors, Newport, RI, Sept. 23-27, 1990, CONF-9009108 (ANL/RERTR/TM-18), Argonne National Laboratory (1993).

20

7. W. L. Woodruff, J. R. Deen and C. Papastergiou, "Transient Analyses and Thermal- hydraulic Safety Margins for the Greek Research Reactor (GRR1),"

Proc. Int. Mtg. on Reduced Enrichment for Research and Test Reactors, Williamsburg, VA, Sept. 19-23, 1994, CONF-9409107 (ANL/RERTR/TM-20), Argonne National Laboratory (1997).

8. W. L. Woodruff, "A Kinetics and Thermal-hydraulics Capability for the Analysis of Research Reactors," Nucl. Technol., Volume 64, 196 (1983).

W. L. Woodruff and R. S. Smith, "A Users Guide for the ANL Version of the PARET Code, PARET/ANL (2001 Rev.)," ANL/RERTR/TM-16, Mar. 2001.

Fifth Response Submitted November 26, 2010 In Section 4.6 the original PLTEMP code' was used to determine many of the thermal-hydraulic characteristics of the reactor core. This code has been 2 The analysis of steady-state superseded by the PLTEMP/ANL V4.0 code.

forced-convection operation has been redone using the newer code and is attached as a new Section 4.6 of the SAR, "4.6 Steady-State Thermal-Hydraulic Analysis" (Reference BB).

Although both the old and the newer code have models that can be used to obtain the flow distribution in a reactor core, a more direct, transparent, and tractable approach was taken in the new analysis. A hydraulics model of the RINSC core was developed based on engineering fundamentals. The equations used in the analysis are provided. Their application is explained in detail. Key intermediate results for a reactor flow rate of 1580 gpm are given in tables so that one can verify that the analysis was performed correctly. The hydraulics analysis yielded the flow rate for each fueled channel in the reactor as a function of total reactor flow rate.

In the next phase of the analysis, the individual channel flow rates obtained in the previous phase were used to perform a thermal analysis of the core. The limiting channel was identified as an internal channel in element D6 and next to the highest power fuel plate. The new PLTEMP code was used to perform thermal analysis of this limiting channel. The highest power fuel plate bounded by two channels, each with half of the flow area of single channel, was modeled.

This model is simple and easy to check. For the channel flow rate corresponding to a reactor flow rate of 1580 gpm the PLTEMP results for the powers at which the onset of nucleate boiling and the onset of flow instability were predicted to occur were verified by a hand calculation. Key intermediate results used in the verification are given in tables so that one can verify that the results are correct.

In the new analysis, key results such as coolant flow rates, bulk coolant and clad surface temperatures, the conditions that produce onset of nucleate boiling and the conditions that cause flow instability are either performed by a hand calculation or are verified by one. Thus, PLTEMP code validation, benchmarking, and code uncertainty are less relevant because results that are 21

essential to demonstrate the safety of the RINSC core have been hand calculated or hand checked. Moreover, the PLTEMP/ANL code used in the new analysis is based on an evolutionary sequence of "PLTEMP" codes in use at ANL for the past 26 years.3- 9 Validation of PLTEMP/ANL has followed standard practice in any code development task, where comparisons are made with other codes, with measurements, and with hand calculations where possible. Many examples of validation are given in the PLTEMP/ANL V4.0 manual 2, and in References 3-9.

As a copy of the original PLTEMP code,' reference 4.6 in the 2004 SAR is provided.

Also provided is Reference 4-6:

K. Mishima, K. Kanda and T. Shibata, "Thermal-hydraulic Analysis for Core Conversion to Use of Low-Enrichment-uranium Fuels in KUR," KURRI-TR-258 (1984).

References:

9. Kaichiro Mishima, Keiji Kanda, and Toshikazu Shibata, "Thermal-Hydraulic Analysis for Core Conversion to the Use oaf Low-Enriched Uranium Fuels in the KUR," Proceedings of the 1984 International Meeting on Reduced Enrichment.for Research and Test Reactors, Argonne National Laboratory, October 15-18, 1984.

10. Arne P. Olson and Kalimullah, "A Users Guide to the PLTEMP/ANL V4.0 Code," Global Threat Reduction Initiative (GTRI) - Conversion Program, Nuclear Engineering Division, Argonne National Laboratory, March 24, 2010.

11. K. Mishima, K. Kanda and T. Shibata, "Thermal-Hydraulic Analysis for Core Conversion to the Use of Low-enrichment Uranium Fuels in the KUR,"

KURRI-TR-258, Research Reactor Institute, Kyoto University, Dec. 7, 1984.

12. K. Mishima, K. Kanda and T. Shibata, "Thermal-Hydraulic Analysis for Core Conversion to the Use of Low-Enriched Uranium Fuels in the KUR,"

ANL/RERTR/TM-6, CONF-8410173, p. 375, 1984.

13. W. L. Woodruff and K. Mishima, "Neutronics and Thermal-Hydraulics Analysis of KUHFR," ANL/RERTR/TM-3, CONF-801144, p. 579, 1980.

14. W. L. Woodruff, "Some Neutronics and Thermal-hydraulics Codes for Reactor Analysis Using Personal Computers," Proc. Int. Mtg. on Reduced Enrichment for Research and Test Reactors, Newport, RI, Sept. 23-27, 1990, CONF-9009108 (ANL/RERTR/TM-18), Argonne National Laboratory (1993).

15. W. L. Woodruff, J. R. Deen and C. Papastergiou, "Transient Analyses and Thermal- hydraulic Safety Margins for the Greek Research Reactor (GRRI),"

Proc. Int. Mtg. on Reduced Enrichment for Research and Test Reactors, Williamsburg, VA, Sept. 19-23, 1994, CONF-9409107 (ANL/RERTR/TM-20), Argonne National Laboratory (1997).

22

16. W. L. Woodruff, "A Kinetics and Thermal-hydraulics Capability for the Analysis of Research Reactors," Nucl. Technol., Volume 64, 196 (1983).

9. W. L. Woodruff and R. S. Smith, "A Users Guide for the ANL Version of the PARET Code, PARET/ANL (2001 Rev.)," ANL/RERTR/TM- 16, Mar. 2001.

4.16 Section 4.6. This section makes multiple references to "Reference 4-Y." Provide the correct reference and a copy of the reference document.

First Response Submitted June 10, 2010 Reference 4-Y will be re-numbered in the Final Safety Analysis Report, and will refer to "Report on the Determination of the Hot Spot Factors for the Rhode Island Nuclear Science Center Research Reactor Using LEU Fuel", Eugene Spring, August 24, 1989. A copy of this report is attached.

4.17 Provide the values of coolant temperature and coolant height used in the analysis of Section 4.6.2. Provide justification for the use of these values.

Fourth Response Submitted September 8, 2010 Section 4.6 in the 2004 SAR will be superseded by a new Section 4.6. In the new analysis the values of inlet coolant temperature and coolant height used in the analysis are based on the Technical Specifications Safety Limit values for reactor power (2.4 MW), reactor flow (1580 gpm), outlet temperature (1250 F),

and pool water depth (23.54 feet above the active core). These are slightly more limiting than was necessary since the original Limiting Safety System Setting values would have been sufficient.

The added pressure at the top of the active core due to the weight of the water at a depth of 23.54 feet is 989.8 kg/mi3 x 9.80665 m/s2 x 23.54 ft x 0.0254 x 12 m/ft =

69645 Pa = 0.696 bar. Narragansett, Rhode Island is at 20 feet above sea level, where atmospheric pressure is 1.013 bar. Thus, the absolute pressure at the top of the active core is 1.013 + 0.696 bar = 1.709 bar. A pressure of 1.7 bar was used in the analysis. The enthalpy at the core exit was obtained for this pressure and the 1250 F via the NIST Steam Tables. The power to flow ratio of 2.4 MW / 1580 gpm yielded the enthalpy rise from core inlet to core outlet. The outlet enthalpy minus the enthalpy rise yielded the inlet enthalpy. The inlet enthalpy and 1.7 bar pressure yielded the inlet temperature of 114.50 F. This was rounded up to 115'F and used as the inlet temperature for the new Section 4.6 thermal analyses. The density of water at 115°F and 1.7 bar is 989.8 kg/m3 , which is the value use above to determine the added pressure due to the depth of the water.

Thus, the direct answer to the question is 1150 F and 23.54 feet of water above the active core. The justification is that these values are consistent with the Technical Specifications Safety Limit values, which bound the Limiting Safety System Setting values.

23

Fifth Response Submitted November 26, 2010 Section 4.6 in the 2004 SAR has been superseded by a new Section 4.6 and is attached as "4.6 Steady-State Thermal-Hydraulic Analysis" (Reference BB). In the new analysis the values of inlet coolant temperature and coolant height used in the analysis are based on the Technical Specifications Safety Limit values for reactor power (2.4 MW), reactor flow (1580 gpm), outlet temperature (1250 F),

and pool water depth (23.54 feet above the active core). These are slightly more limiting than was necessary since Limiting Safety System Setting values would have been sufficient.

The added pressure at the top of the active core due to the weight of the water at a depth of 23.54 feet is 989.8 kg/mi3 x 9.80665 m/s2 x 23.54 ft x 0.0254 x 12 m/ft =

69645 Pa = 0.696 bar. Narragansett, Rhode Island is at 20 feet above sea level, where atmospheric pressure is 1.013 bar. Thus, the absolute pressure at the top of the active core is 1.013 + 0.696 bar = 1.709 bar. A pressure of 1.7 bar was used in the analysis. The enthalpy at the core exit was obtained for this pressure and the 125' F via the NIST Steam Tables. The power to flow ratio of 2.4 MW / 1580 gpm yielded the enthalpy rise from core inlet to core outlet. The outlet enthalpy minus the enthalpy rise yielded the inlet enthalpy. The inlet enthalpy and 1.7 bar pressure yielded the inlet temperature of 114.50 F. This was rounded up to 115°F and used as the inlet temperature for the new Section 4.6 thermal analyses. The density of water at 115°F and 1.7 bar is 989.8 kg/m 3, which is the value use above to determine the added pressure due to the depth of the water.

Thus, the direct answer to the question is 1150 F and 23.54 feet of water above the active core. The justification is that these values are consistent with the Technical Specifications Safety Limit values, which bound the Limiting Safety System Setting values.

4.18 Section 4.6.2. Confirm that the units for the values of Tsurface, Tsat, and Tonb found in Table 4-9 and Table 4-10 are degrees C.

Fourth Response Submitted September 8, 2010 It is confirmed that the values of Tsurface, Tsat, and Tonb found in Table 4-9 and Table 4-10 are degrees C.

4.19 Section 4.6.2. Describe the methods used to determine the values of Tsurface and Tonb found in Table 4-9 and Table 4-10. Include all assumptions and correlations used in the calculations and provide justification for their use given the thermal-hydraulic characteristics of the coolant channels.

Fourth Response Submitted September 8, 2010 24

Section 4.6 in the 2004 SAR will be superseded by a new Section 4.6. In the new analysis the values of the clad surface temperature (Tsurface) and the values of the surface temperature that cause onset of nucleate boiling (Tonb) are calculated by the PLTEMP/ANL V4.0 code.' A copy of the manual of the code is provided. The key heat transfer correlations used in the analysis of onset of nucleate boiling are the Bergles and Rohsenow correlation, which is used to determine the amount of superheat to reach onset of nucleate boiling at the clad surface, and the Dittus-Boelter correlation, which is used to determine the Nusselt number. These two correlations and a full description of the models, assumptions, and correlations of the PLTEMP/ANL code can be found in the PLTEMP/ANL V4.0 code manual.

In the analysis for onset of nucleate boiling, the PLTEMP/ANL V4.0 code was used to analyze the limiting channel in the reactor. This analysis was performed once for each of eight reactor flow rates, spanning the flow range from 1000 to 2200 gpm. Since the model represented only the limiting channel, it was easy to hand-check the PLTEMP/ANL V4.0 results. This hand-check was performed for a reactor flow rate of 1580 gpm. At this flow rate the limiting channel has a flow rate of 0.2210 kg/s. For this flow rate PLTEMP/ANL V4.0 found that the channel power is 22.80 kW, which corresponds to a reactor power of 4.72 MW.

The code results also indicate the axial locations where onset of nucleate boiling was first reached. Key values of the verification are provided in the new Section 4.6 in a table, which is repeated below as Table 4.19-1. Additional specifics of the analysis can be found in the memo dated 3 September 2010 from Earl E.

Feldman to James E. Matos entitled "Steady State Thermal Hydraulic Analysis for Forced Convective Flow in the Rhode Island Nuclear Science Center (RINSC) Reactor", including a listing of the PLTEMP/ANL V4.0 input used for this analysis.

25

I Table 4.19 Verification of PLTEMP/ANL Onset of Nucleate Boiling Prediction at the Limiting Axial Location for a Core Flow Rate of 1580 gpm Hand PLTEMP Quantity Calculation Channel Dimensions Thickness, in 0.088 Width, in 2.62 Heated Length, in 23.25 Heat Width, in 2.395 Wetted Perimeter, m 0.1376 Hydraulic Diameter, m 0.004325 Heat Transfer Area (2 Faces), m2 0.07185 Channel Flow Rate, kg/s 0.2210 Channel Power, kW 22.80 Pressure, bar 1.7 Inlet Temperature, C 46.11 Saturation Temperature, C 115.15 Cp @ 55 C, kJ/kg-C 4.1828 At Onset of Nucleate Boiling Location Layer 13 13 Channel Power to Middle of Layer, kW 15.64 Channel Power to Exit of Layer, kW 16.56 Local Peak-to-Average Power 1.2894 Bulk Temperature at Middle of Layer w/o Hot Chan. Fac., C 63.03 63.01 Random Hot Channel Factor on AT bulk 1.24 Bulk Temperature with Hot Channel Factor, C 67.09 67.06 Bulk Temperature at Exit of Layer w/o Hot Chan. Fac., C 68.32 Viscosity, Pa-s 4.3910E-4 Reynolds Number 14634 Thermal Conductivity, W/m-C 0.65817 Cp, kJ/kg-C 4.1867 Prandtl Number 2.793 Nusselt Number without Hot Channel Factors 74.55 Global Film Coefficient Hot Channel Factor 1.2 Film Coefficient with Hot Channel Factor, W/mr-C 9454 Heat Flux without Hot Channel Factors, MW/mI 0.4092 0.4091 Random Hot Channel Factor on AT film 1.28 Film Temperature Rise with Hot Channel Factor, C 55.77 Clad Surface Temperature with All Hot Channel Factors, C 122.5 122.5 Random Hot Channel Factor on Heat Flux 1.23 Heat Flux with Hot Channel Factors, MW/mz 0.4779 ATsaturation based on Bergles and Rohsenow, C 7.41 Surface Temperature For Onset of Nucleate Boiling, C 122.6 122.6 Table 4.19 Verification of PLTEMP/ANL Onset of Nucleate Boiling Prediction at the Limiting Axial Location for a Core Flow Rate of 1580 gpm 26

Reference:

Arne P. Olson and Kalimullah, "A Users Guide to the PLTEMP/ANL V4.0 Code," Global Threat Reduction Initiative (GTRI) - Conversion Program, Nuclear Engineering Division, Argonne National Laboratory, March 24, 2010.

Fifth Response Submitted November 26, 2010 Section 4.6 in the 2004 SAR has been superseded by a new Section 4.6 and is attached as "4.6 Steady-State Thermal-Hydraulic Analysis" (Reference BB). In the new analysis the values of the clad surface temperature (Tsurface) and the values of the surface temperature that cause onset of nucleate boiling (Tonb) are calculated by the PLTEMP/ANL V4.0 code.' A copy of the manual of the code is provided. The key heat transfer correlations used in the analysis of onset of nucleate boiling are the Bergles and Rohenow correlation, which is used to determine the amount of superheat to reach onset of nucleate boiling at the clad surface, and the Dittus-Boelter correlation, which is use to determine the Nusselt number. These two correlations and a full description of the models, assumptions, and correlations of the PLTEMP/ANL code can be found in Reference 1.

In the analysis for onset of nucleate boiling, the PLTEMP/ANL V4.0 code was used to analyze the limiting channel in the reactor. This analysis was performed once for each of eight reactor flow rates, spanning the flow range from 1000 to 2200 gpm. Since the model represented only the limiting channel, it was easy to hand-check the PLTEMP/ANL V4.0 results. This hand-check was performed for a reactor flow rate of 1580 gpm. At this flow rate the limiting channel has a flow rate of 0.2210 kg/s. For this flow rate PLTEMP/ANL V4.0 found that the channel power is 22.80 kW, which corresponds to a reactor power of 4.72 MW.

The code results also indicate the axial locations where onset of nucleate boiling was first reached. Key values of the verification are provided in the new Section 4.6 in a table, which is repeated below as Table 4.19-1. Additional specifics of the analysis can be found in the new Section 4.6, including a listing of the PLTEMP/ANL V4.0 input used for this analysis.

27

Table 4.19 Verification of PLTEMP/ANL Onset of Nucleate Boiling Prediction at the Limiting Axial Location for a Core Flow Rate of 1580 gpm Quantity Hand PLTEMP Calculation Channel Dimensions Thickness, in 0.088 Width, in 2.62 Heated Length, in 23.25 Heat Width, in 2.395 Wetted Perimeter, m 0.1376 Hydraulic Diameter, m 0.004325 Heat Transfer Area (2 Faces), m 2 0.07185 Channel Flow Rate, kg/s 0.2210 Channel Power, kW 22.80 Pressure, bar 1.7 Inlet Temperature, C 46.11 Saturation Temperature, C 115.15 Cp @ 55 C, kJ/kg-C 4.1828 At Onset of Nucleate Boiling Location Layer 13 13 Channel Power to Middle of Layer, kW 15.64 Channel Power to Exit of Layer, kW 16.56 Local Peak-to-Average Power 1.2894 Bulk Temperature at Middle of Layer w/o Hot Chan. 63.03 63.01 Fac., C Random Hot Channel Factor on AT bulk 1.24 Bulk Temperature with Hot Channel Factor, C 67.09 67.06 Bulk Temperature at Exit of Layer w/o Hot Chan. Fac., 68.32 C

Viscosity, Pa-s 4.391OE-4 Reynolds Number 14634 Thermal Conductivity, W/m-C 0.65817 Cp, kJ/kg-C 4.1867 Prandtl Number 2.793 Nusselt Number without Hot Channel Factors 74.55 Global Film Coefficient Hot Channel Factor 1.2 Film Coefficient with Hot Channel Factor, W/m 2-C 9454 Heat Flux without Hot Channel Factors, MW/m 2 0.4092 0.4091 Random Hot Channel Factor on AT film 1.28 Film Temperature Rise with Hot Channel Factor, C 55.77 28

Clad Surface Temperature with All Hot Channel 122.5 122.5 Factors, C Random Hot Channel Factor on Heat Flux 1.23 Heat Flux with Hot Channel Factors, MW/m 2 0.4779 ATsaturation based on Bergles and Rohsenow, C 7.41 1 Surface Temperature For Onset of Nucleate Boiling, C 122.6 122.6

Reference:

Ame P. Olson and Kalimullah, "A Users Guide to the PLTEMP/ANL V4.0 Code," Global Threat Reduction Initiative (GTRI) - Conversion Program, Nuclear Engineering Division, Argonne National Laboratory, March 24, 2010.

4.20 Section 4.6.2. Provide the uncertainties for the limiting safety system setting (LSSS) values for coolant height and overpower trip. Provide justification for all uncertainty values associated with the LSSS for coolant height, overpower trip, coolant temperature, and coolant flow. (See RAI 14.44)

Second Response Submitted August 6, 2010 Pool Level:

A low pool level is determined by the change in state of a float switch. The height of the switch is set so that there is no uncertainty that the switch will change state before the height of the water level above the core is less than 23.7 ft. However, in order to be conservative, an error of 0.5 inches is assumed.

Temperature Measurement:

Inlet, outlet, and bulk pool temperatures at RINSC are measured with an RTD sensor. The technical manual for the meters associated with these sensors indicates that they have an accuracy of + or - 0.5 C. See the reference entitled "Flow and Temperature Meter Specifications".

The normal operating temperature range of the primary coolant is between 90 F and 110 F. Consequently, an error of 0.5 C is:

5/9(90 F - 32) = 32 C 1.5 C/32 C=0.015=1.5%

29

Flow Measurement:

Coolant flow rate is measured by looking at the pressure differential across an orifice plate. This differential is transmitted to a meter as a voltage signal.

The errors associated with this measurement are:

+ or - 2% of the upper range for the orifice plate (See the reference entitled "Flow Measurement Uncertainty")

+ or - 0.25% accuracy for the transmitter (See the reference entitled "Flow Measurement Uncertainty")

+ or - 0.05% accuracy for the meter voltage input (See the reference entitled "Flow and Temperature Meter Specifications")

Power Level:

From the "Report on the Determination of the Hot Spot Factors for the Rhode Island Nuclear Science Center Research Reactor Using LEU Fuel",

Eugene Spring, August 24, 1989 which is used for hot spot analysis, the error associated with power level is 10 %.

421 Section 4.6.2 states that with a flow rate of 1,950 gallons per minute (gpm), the incipient boiling temperature (defined in the SAR as Tonb) occurs at about 2.6 MW.

From Table 4-9, it appears that the incipient temperature occurs somewhere between 1,715 gpm and 1,800 gpm. Clarify this apparent discrepancy.

Fourth Response Submitted September 8, 2010 Section 4.6 in the 2004 SAR will be superseded by a new Section 4.6. In the new analysis Table 4-9 is longer relevant. A more detailed explanation, if one is needed, follows.

Table 4.9 represents a double search. First a range of solutions for various core pressure drops are searched and interpolated to find the one that provides the correct total core flow rate. Then at the desired core pressure drop and flow rate, solutions for a range of assumed reactor power levels are searched and interpolated to find the one that provides the clad surface temperature corresponding to the onset of nucleate boiling. In the new analysis, on the other hand, for each specific reactor flow rate, the flow rate of each fueled channel is determined via a custom-developed hydraulics model. (This model and its governing equation are described in the memo dated 3 September 2010 from Earl E. Feldman to James E. Matos entitled "Steady State Thermal Hydraulic Analysis for Forced Convective Flow in the Rhode Island Nuclear Science Center (RJNSC) Reactor" in considerable detail. Key intermediate results for a reactor flow rate of 1580 gpm are given in tables so that one can verify that the 30

analysis was performed correctly.) The new analysis uses the PLTEMP/ANL V4.0 code' to find the power at which onset of nucleate boiling occurs. This code has an internal search method, which determines the power level at which onset of nucleate boiling is first achieved and provides the power level value as a code output. Thus, in the new analysis the double search is not used and Table 4-9, or another similar table, is not needed.

In the new analysis, the onset of nucleate boiling for a reactor flow rates of 1715, 1800, and 1,950 gpm are predicted to occur at a reactor powers of 5.1, 5.3, and 5.7 MW, respectively. The increased allowed power is, in part, attributable to 1) a reduction in the hot channel factors due to more use of statistical, rather than multiplicative, methods of combining uncertainty factors and 2) improvements in the determination of the reactor power distribution.

Reference:

Arne P. Olson and Kalimullah, "A Users Guide to the PLTEMP/ANL V4.0 Code," Global Threat Reduction Initiative (GTRI) - Conversion Program, Nuclear Engineering Division, Argonne National Laboratory, March 24, 2010.

Fifth Response Submitted November 26, 2010 Section 4.6 in the 2004 SAR has been superseded by a new Section 4.6 and is attached as "4.6 Steady-State Thermal-Hydraulic Analysis" Reference BB. In the new analysis Table 4-9 is longer relevant. A more detailed explanation, if one is needed, follows.

Table 4.9 represents a double search. First a range of solutions for various core pressure drops are searched and interpolated to find the one that provides the correct total core flow rate. Then at the desired core pressure drop and flow rate, solutions for a range of assumed reactor power levels are searched and interpolated to find the one that provides the clad surface temperature corresponding to the onset of nucleate boiling. In the new analysis, on the other hand, for each specific reactor flow rate, the flow rate of each fueled channel is determined via a custom-developed hydraulics model. (This model and its governing equation are described in the new Section 4.6 in considerable detail.

Key intermediate results for a reactor flow rate of 1580 gpm are given in tables so that one can verify that the analysis was performed correctly.) The new Section 4.6 analysis uses the PLTEMP/ANL V4.0 code' to find the power at which onset of nucleate boiling occurs. This code has an internal search method, which determines the power level at which onset of nucleate boiling is first achieved and provides the power level value as a code output. Thus, in the new analysis the double search is not used and Table 4-9, or another similar table, is not needed.

31

In the new analysis, the onset of nucleate boiling for a reactor flow rates of 1715, 1800, and 1,950 gpm are predicted to occur at a reactor powers of 5.1, 5.3, and 5.7 MW, respectively. The increased allowed power is, in part, attributable to 1) a reduction in the hot channel factors due to more use of statistical, rather than multiplicative, methods of combining uncertainty factors and 2) improvements in the determination of the reactor power distribution.

Reference:

Arne P. Olson and Kalimullah, "A Users Guide to the PLTEMP/ANL V4.0 Code," Global Threat Reduction Initiative (GTRI) - Conversion Program, Nuclear Engineering Division, Argonne National Laboratory, March 24, 2010.

4.22 Section 4.6.2 states that the reduced flow trip setting is 1,700 gpm. The requirements of TS 2.2.1 and TS 3.2.1 allow the reduced flow trip to be set at 1,600 gpm. Clarify this apparent discrepancy.

Fourth Response Submitted September 8, 2010 Section 4.6 in the 2004 SAR will be superseded by a new Section 4.6. See the response to RAI 14.36 and 14.32.

4.23 Note number 2 to Table 4-12 states that the calculations are based on a reactor inlet temperature of 42.3 degrees C. Explain the reason this temperature is used in the analysis given that it is less conservative than the coolant temperatures allowed by the proposed TS.

Fourth Response Submitted September 8, 2010 It is true that an inlet temperature of 42.30 C (1080 F) it is less conservative than the coolant temperatures allowed by the proposed TS. Section 4.6 in the 2004 SAR will be superseded by a new Section 4.6 developed for these RAI responses. In the new analysis described in the memo dated 3 September 2010 from Earl E. Feldman to James E. Matos entitled "Steady State Thermal Hydraulic Analysis for Forced Convective Flow in the Rhode Island Nuclear Science Center (RINSC) Reactor", the inlet temperature is assumed to be 115'F (46.11' C). As explained in the answer to Question 4.17, this is based on the Technical Specifications Safety Limit values for reactor power (2.4 MW), reactor flow (1580 gpm), outlet temperature (1250 F), and pool water depth (23.54 feet above the active core).

Fifth Response Submitted November 26, 2010 It is true that an inlet temperature of 42.30 C (1080 F) it is less conservative than the coolant temperatures allowed by the proposed TS. Section 4.6 in the 2004 SAR has been superseded by a new Section 4.6 developed for these RAI responses and is attached as "4.6 Steady-State Thermal-Hydraulic Analysis" 32

Reference BB. In the new analysis the inlet temperature is assumed to be 11 5'F (46.11 C). As explained in the answer to Question 4.17, this is based on the Technical Specifications Safety Limit values for reactor power (2.4 MW), reactor flow (1580 gpm), outlet temperature (1250 F), and pool water depth (23.54 feet above the active core).

4.24 The rising power transient analysis of Section 4.6.4 of the SAR shows that the reactor power would reach 2.78 MW. Explain how the analysis supports the safety limit of 2.4 MW given in TS 2.1.1.1. (See RAI 14.32)

Fourth Response Submitted September 8, 2010 The rising power transient analysis of Section 4.6.4 of the SAR is misplaced and will be moved to the section on reactivity insertion accident analyses in Chapter 13.

The analysis in the SAR was redone using the PARET/ANL code for a slow insertion of reactivity from high power. Since the power level is 2.0 +/- 0.2 MW, including 10% uncertainty in the power level measurement, an initial power of 1.8 MW was selected to achieve the maximum rise in power after the trip on power is initiated at 2.3 MW. The description of the transient is given below.

The reactor is initially operating at 1.8 MW, 123 F coolant inlet temperature, and 1740 gpm. There is a water head of 23' 9.1" above the top of the fuel meat, which provides a pressure of 1.715E+5 Pa. Then a long, slow ramp reactivity insertion begins at a ramp rate of 0.02 % Ak/k / s, continuing for 100 s. Power rises slowly. The power trip at 2.3 MW is actuated at 6.774 s. Since no actual negative reactivity from the control system occurs for 100 ms after the trip, the reactor power continues to rise from the trip level of 2.3 MW to a maximum of 2.313 MW at 6.874 s. The reactor power drops rapidly to shutdown conditions.

Peak temperatures for fuel meat centerline, and clad surface are: 76.7 C and 75.9 C. The peak coolant temperature of 59.6 C is reached at 6.90 s. These peak fuel and clad surface temperatures are far below the maximum temperature of 530 'C for LEU silicide fuel that the NRC finds acceptable as fuel and clad temperature limits not to be exceeded under any conditions of operation (See NUREG-1537, Part I, Appendix 14.1 and NUREG-1313). The peak coolant temperature is well below the saturation temperature of 115.4 C.

4.25 Section 4.6.4. The assumptions used in the rising power transient calculation are not consistent with the requirements of the TS. The analysis assumes a minimum reactor period of slightly greater than 7 seconds, while the TS allow a minimum reactor period of 4 seconds. Also, the surface temperature value of 122.93 degrees C appears to be based on a coolant flow rate of 1,715 gpm, which is greater than the TS requirement of 1,600 gpm. Provide a revised 33

calculation that supports the requirements in the TS. Include justification of all assumptions, including the assumed coolant temperature and coolant height.

Fourth Response Submitted September 8, 2010 See the response to RAI 4.24.

4.26 Section 4.6.4 gives a surface temperature of 122.93 degrees C for normal flow at 2.6 MW. Table 4-9 indicates that this temperature corresponds to 1,715 gpm at 2.6 MW. Page 4-3 indicates that nominal flow is 1,950 gpm. Clarify this apparent discrepancy.

Fifth Response Submitted November 26, 2010 See the response to RAI 4.24. A new analysis has been completed for which this discrepancy no longer exits.

4.27 Section 4.6.4 states, "for a hot channel analysis, the ONB region would not present a problem for the LEU fuel." Provide justification for this conclusion.

Fifth Response Submitted November 26, 2010 See the response to RAI 4.24.

4.28 Provide the values of coolant temperature and coolant height used in the analysis of Section 4.6.5. Provide justification for the use of these values.

Fifth Response Submitted November 26, 2010 Section 4.6.5 of the 2004 RINSC Reactor SAR provides an analysis of the thermal behavior of the LEU core during steady-state operation in the natural-convection mode. This analysis has been completely redone and is replaced by section 4.7 of Reference AA.

Reference AA refers to the completely redone analysis of the thermal behavior of the LEU core during steady-state operation in the forced-convection mode, which is provided in sections 4.6.1 through 4.6.12 of the attached Reference BB and is a replacement for sections 4.6.1, 4.6.2, 4.6.3, and 4.8 of the 2004 RINSC Reactor SAR.

In the Reference AA analysis of natural convection, the inlet (and pool) coolant temperature is 1300 F and the coolant height is 23 feet 6.5 inches (23.54 feet) above the active core. These are the Safety Limit values and are more restrictive than the Limiting Trip values of 1280 F and 23 feet 9.1 inches, respectively. (See the table at the end of the response to RAI 14.52.)

34

AA. Argonne National Laboratory intra-laboratory memo, Earl E. Feldman and M. Kalimullah to James E. Matos, "Steady State Thermal-Hydraulic Analysis for Natural-Convective Flow in the Rhode Island Nuclear Science Center (RINSC) Reactor," November 8, 2010.

BB. Argonne National Laboratory intra-laboratory memo, Earl E. Feldman to James E. Matos, "Steady State Thermal-Hydraulic Analysis for Forced-Convective Flow in the Rhode Island Nuclear Science Center (RINSC)

Reactor," September 3, 2010.

4.29 Table 4-17. Explain the meaning of the negative value for Margin to Incipient Boiling at 209.1 kW. (See RAI 14.35)

Fifth Response Submitted November 26, 2010 Table 4-17 is part of section 4.6.5 of the 2004 RINSC Reactor SAR, which provides an analysis of the thermal behavior of the LEU core during steady-state operation in the natural-convection mode. This analysis has been completely redone and is replaced by section 4.7 of Reference AA.

As stated in the response to RAI 4.28, Reference AA refers to the completely redone analysis of the thermal behavior of the LEU core in the forced-convection mode, which is provided in Reference BB.

The meaning of the negative value of Margin to Incipient Boiling at 209.1 kW in the 2004 RINSC SAR implies that the incipient boiling occurs before 209.1 kW is reached. However, in the analysis of section 4.7 of Reference AA, incipient boiling, which is referred to as "onset of nucleate boiling", is predicted to occur at 369 kW with all uncertainties included. The Limiting Trip value of power is 125 kW and the Safety Limit power is 200 kW. (See the table at the end of the response to RAI 14.52.) Thus, the Reference AA analysis shows a large margin to incipient boiling.

AA. Argonne National Laboratory intra-laboratory memo, Earl E. Feldman and M. Kalimullah to James E. Matos, "Steady State Thermal-Hydraulic Analysis for Natural-Convective Flow in the Rhode Island Nuclear Science Center (RINSC) Reactor," November 8, 2010.

BB. Argonne National Laboratory intra-laboratory memo, Earl E. Feldman to James E. Matos, "Steady State Thermal-Hydraulic Analysis for Forced-Convective Flow in the Rhode Island Nuclear Science Center (RINSC)

Reactor," September 3, 2010.

35

4.30 Section 4.7. Clarify whether the correct reference to the figure showing the expanded core configuration is Figure 4-1 or Figure 4-2.

Second Response Submitted August 6, 2010 Section 4.7 will be modified to say:

A modification to the standard 14 element core shown in Figure 4-1 was analyzed and approved. The objective was to increase the neutron flux in the thermal column experimental facility. Several options were considered, including shifting the entire fuel matrix toward the thermal column.

However, the RERTR group at Argonne National Laboratory suggested that an increase in the thermal column neutron flux could be obtained by expanding the core into the 17 element core shown in Figure 4-2. This core is achieved by replacing three graphite reflectors on the thermal column side of the core with fuel.

431 Section 4.8 - Clarify which figure is meant by "Figure 4" in the text.

Second Response Submitted August 6, 2010 The first Paragraph (lines 25-29) will be modified to say:

Two fresh core models were chosen for the thermal hydraulic calculations that would produce the highest possible fuel cladding temperatures under normal reactor operation at 2 MW reactor power. Figure 4-1 show the standard 14 element core, while Figure 4-2 shows the expanded 17 element core. Both cores were analyzed.

The reference to Figure 4-5 in the third paragraph, line 29 will be changed to "Figure 4-3".

4.32 Section 4.8. Provide a discussion of the correlation and/or calculations used to develop the Departure from Nucleate Boiling (DNB) and Departure from Nucleate Boiling Ratio (DNBR). Include all assumptions made in the analysis and justification for those assumptions. Clarify whether the term "Margin to Departure from Nucleate Boiling" in Table 4-19 is synonymous with the term "DNBR."

Fifth Response Submitted November 26, 2010 Sections 4.6.1, 4.6.2, 4.6.3, and 4.8 of the 2004 RINSC Reactor SAR provide the analysis of the thermal behavior of the LEU core during steady-state operation in the forced-convection mode. This analysis has been completely redone in sections 4.6.1 through 4.6.12 of the attached Reference BB and is a replacement for sections 4.6.1, 4.6.2, 4.6.3, and 4.8 of the 2004 RINSC Reactor SAR 36

All assumptions made in the analysis and justification for those assumptions are provided in detail in Reference BB, which includes the hydraulic modeling needed to obtain the individual channel flow rates, the determination of the hot channel factors, a description of the PLTEMP/ANL V4.0 code, which was used in the analysis, and a listing of the code input used in the analysis. In Reference BB section 4.6.10, "Results of Steady-State Thermal Analysis," and section 4.6.11, "Discussion" describe the determination of the power at which departure from nucleate boiling (DNB) is predicted to occur and include a discussion of the correlations and the calculations that were used.

The new analysis does not use either the term "departure from nucleate boiling ratio" or the term "DNBR". Instead, the new analysis indicates the power in kW or MW at which departure from nucleate boiling is predicted to occur.

BB. Argonne National Laboratory intra-laboratory memo, Earl E. Feldman to James E. Matos, "Steady State Thermal-Hydraulic Analysis for Forced-Convective Flow in the Rhode Island Nuclear Science Center (RINSC)

Reactor," September 3, 2010.

5.1 Section 5.5. The makeup water system operates to automatically add water to the reactor pool upon a low level indication. Excessive operation of the system could either indicate a leak in the reactor coolant, a malfunction of the pool level indicator, or a malfunction of the makeup water controls. This could result in overfilling of the pool. Describe the controls to detect abnormalities or leaks in the makeup water system.

First Response Submitted June 10, 2010 Make-up water into the reactor primary system is metered. As part of securing the facility at the end of each weekday, the make-up water meter reading is recorded and checked to verify that abnormal amounts of water have not been added. The Make-Up water System solenoid valve opens and closes to control the fill. A float switch is used to sense pool level. When the float switch senses that the pool level has dropped one inch, the solenoid valve is opened to allow flow into the pool. If the float switch senses that the pool level has dropped two inches, a reactor scram is initiated, and an alarm is sent to a contracted alarm company that monitors the facility security, fire, and other parameter alarms 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> per day, seven days per week. In addition to the normal float switch position that closes the make-up supply valve, a second overflow switch position exists as a secondary overflow protection.

6.1 Describe the operating parameters and design features of the confinement, including: free volume, number and type of penetrations, etc.

Fourth Response Submitted September 8, 2010 37

The volume of the confinement building is approximately 203,695 cubic feet.

The volume of the pool structure and water is approximately 21,740 cubic feet, leaving 181,955 cubic feet of open space. The control room takes up about 3,612 cubic feet of this space, with the air changed regularly with the confinement air for temperature and humidity control.

During normal operation air enters the confinement building through the Confinement Air Intake opening. This air will disperse throughout the confinement building until it is collected at the Normal Ventilation Intake, where air is drawn in by way of the Confinement Exhaust Blower. A sample of this air is collected through a copper tube and delivered to a Continuous Air Monitor in the basement, known as the Stack Monitor. This monitor samples the air for gaseous and particulate contamination then returns the sample to the exhaust line. The air is then sent up the stack, where it is diluted with air from outside the EPZ pulled in by the Dilution Blower, then released to the atmosphere at 115 feet above ground. Air from the various irradiation and experiment facilities are collected and combined with the confinement exhaust before the Confinement Exhaust Blower.

At five locations throughout the facility there are red evacuation buttons. Three are within the confinement building: one to the right of the portal entrance, one to the right of the roll-up door, and one in the control room. In addition to an audible alarm heard throughout the building, pressing any of these buttons will cause the confinement air handling systems to go into emergency mode. When this is initiated power is cut to the Confinement Exhaust Blower, the Off Gas Blower, and the Rabbit Blower. As the flow rate drops pneumatic switches trigger the pneumatic Exhaust and Intake Dampers to close. As this occurs, the Emergency Blower is energized, pulling air through the Emergency Ventilation Intake. This air is passed through a series of filters, including a charcoal filter, removing the majority of Iodine that may be present. This air is then directed to the Emergency Ventilation Exhaust where the air is diluted with air from outside the EPZ that is pulled in by the Dilution Blower. This air then travels up the stack and is released 115 feet above ground level. Because the normal sampling location for stack exhaust activity no longer represents the air being released to the environment, sampling is instead taken from approximately half way up the stack. This is achieved by way of a motorized three-way valve located at the Stack Monitor that is activated during emergency mode.

All other major penetrations through the confinement remain closed during operation, with the exception of the Lab Wing Portal, which utilizes a secondary door to maintain confinement integrity. During operation, the confinement building maintains at least a one-half inch of negative pressure in relation to the outside environment. Any other penetrations not mentioned, such as for coolant piping or wiring, are sealed or provide negligible effects on potential radioactive 38

release or unauthorized access. A summary of penetrations are provided in the following table:

Summary of Confinement Penetrations 7.1 Provide a listing of the interlocks of the reactor protection system.

Fourth Response Submitted September 8, 2010 Table 3.1 will be revised to be:

39

Protection Cooling Channels Function Set Point Mode Required Over Power Both 2 Scram by Power Level Less than or 105% of Equal to Licensed Power Low Pool Level Both 1 Scram by Pool Level Less than or 23 ft 9.6 Drop Equal to in Primary Coolant Inlet Forced I Alarm by Inlet Temp Less than or 111 F Temperature Equal to Primary Coolant Outlet Forced I Alarm by Outlet Temp Less than or 117 F Temperature Equal to Forced I Scram by Outlet Temp Less than or 120 F Equal to Pool Temperature Both I Scram by Pool Temp Less than or 125 F Equal to Primary Coolant Flow Primary Flow Less than or 1800 Rate Both I Scram by Rate Equal to gpm Rate of Change of Less than or 4 Power Both I Scram by Period Equal to seconds Seismic Disturbance Both I Scram if Seismic Disturbance Detected Bridge Low Power Position Forced 1 Scram if Bridge Not Seated at HP End Bridge Movement Both 1 Scram if Bridge Movement Detected Coolant Gates Open Forced 1 Scram if Inlet Gate Open Forced I Scram if Outlet Gate Open Detector HV Less than or Detector HV Failure Both 1 Scram if Decrease Equal to 50 V Detector HV Less than or Both 1 Scram if Decrease Equal to 50 V Detector HV Less than or Both 1 Scram if Decrease Equal to 50 V No Flow Thermal Column Forced 1 Scram by No Flow Detected Manual Scram Both 1 Scram by Button Depressed Both 1 Scram by Button Depressed No Automatic Servo Control Interlock Both 1 Servo if Regulating Blade not Full Out No Automatic 30 Both I Servo if Period Less than seconds Shim Safety No SS Withdrawal Both 1 Withdrawal if Count Rate Less than 3 cps No SS Both I Withdrawal if Test/Select SW not Off Rod Control Loss of Less than or 10 Communication Both I Scram if Communication Equal to seconds 40

7.2 Section 7.2.5. Provide more detail regarding the interconnections of the neutron flux monitoring system, including equipment lists and performance specifications to clarify its operation.

First Response Submitted June 10, 2010 Neutron flux is monitored by three independent instruments:

Neutron Flux Monitor Wide Range Monitor #1 Wide Range Monitor #2 The Neutron Flux Monitor consists of a fission chamber located in grid position B-9. This instrument contains the Start-Up Channel, a Wide Range Channel, and a Linear Power Channel. The Start-Up and Wide Range Channels each have a Period Channel associated with them.

Each of the Wide Range Monitor Instruments has a Linear Power Channel. Each monitor uses a compensated ion chamber to detect neutrons. The detector for WRM #1 is located in grid position A-I, and the detector for WRM #2 is located in grid position G-9.

7.3 Section 7.2.12 discusses the relay scram circuit. Provide more detail regarding the design and operation of the bridge misalignment and bridge movement safety channels required by TS 3.2.1, Table 3.1. Do these channels have set points?

What is the minimum motion detectable by the bridge movement safety channel? What is the alignment tolerance associated with the bridge misalignment safety channel? Explain any interlocks that prevent reactor startup in the forced convection mode if the bridge is misaligned. (See RAI 14.71)

First Response Submitted June 10, 2010 The bridge misalignment scram consists of a position switch mounted at the high power end of the reactor pool. When the bridge is not fully seated at the high power end of the pool, the switch is in a state that initiates the "Bridge Lo Pwr Pos" scram. This scram is only active when the reactor is being operated in the forced convection cooling mode. For natural convection cooled operation at 100 kW or less, this scram is bypassed.

The bridge movement scram consists of a limit switch mounted on the gear that turns the wheels of the bridge. The switch is in the non-scram state when the switch actuator is positioned so that it is resting on the top of a gear tooth. If the bridge moves and the switch actuator rests in a valley between two gear teeth, the switch changes state to initiate a scram. The gear teeth are spaced 1/2 inches apart, so movement of less than 1/4 inches will cause a scram.

41

7.4 Section 7.2.15. Provide more detail regarding the fixed radiation monitoring instrumentation. Include a list of the positions and types of fixed monitors and indicate which have local readouts and/or alarms.

Fourth Response Submitted September 8, 2010 Available Radiation Instrumentation:

Instrument Position Detector Readout Alarm Setpoint Type Stack Gaseous Stack Gamma Local and Local and 2.5 X Normal Monitor Sensitive Control Room Control Room Stack Particulate Stack Gamma Local and Local and 2 X Normal Monitor Sensitive Control Room Control Room Main Floor Reactor Main Floor Gamma Local Only Local Only 2 X Normal Particulate Monitor Sensitive Reactor Bridge Area Reactor Bridge Area Gamma Local and Local and 2 X Normal Monitor Sensitive Control Room Control Room Fuel Safe Area Fuel Safe Area Gamma Local and Local and Higher of 2 Monitor Sensitive Control Room Control Room X Normal or 5 mR/hr Thermal Column Thermal Column Neutron Local and Local and Higher of 2 Area Monitor Area Sensitive Control Room Control Room X Normal or 2 mRihr Heat Exchanger Area Heat Exchanger Gamma Local and Local and 2 X Normal Monitor Area Sensitive Control Room Control Room Primary Clean-Up Primary Clean-Up Gamma Local and Local and 2 X Normal Demineralizer Area Demineralizer Area Sensitive Control Room Control Room Monitor Required Radiation Instrumentation:

1. Area Radiation Monitors:

A. A minimum of one radiation monitor that is capable of warning personnel of high radiation levels shall be at the experimental level. The Thermal Column Area Monitor, or equivalent may serve in this capacity.

B. A minimum of one radiation monitor that is capable of warning personnel of high radiation levels shall be over the pool. The Reactor Bridge Area Monitor, or equivalent may serve in this capacity.

C. Alarm set points may be adjusted higher with the approval of the Director or Assistant Director.

42

D. If either of these detectors fail during operation, the staff shall have one hour to either repair the detector, or find an acceptable replacement without having to shut the reactor down.

2. Air Monitors:

A. A minimum of one radiation monitor that is capable of warning personnel of high radiation levels in the confinement gaseous effluent shall be operating.

The Stack Gaseous Monitor, or equivalent may serve in this capacity.

B. A minimum of one radiation monitor that is capable of warning personnel of high radiation levels in the confinement particulate effluent shall be operating.

The Stack Particulate Monitor may serve in this capacity.

C. Alarm set points may be adjusted higher with the approval of the Director or Assistant Director.

D. If either of these detectors fail during operation, the staff shall have six hours to either repair the detector, or find an acceptable replacement without having to shut the reactor down.

9.1 Section 9.1.2 provides no detail regarding the design specifications of the normal and emergency ventilation system other than general arrangement. TS 3.7.2 credits the ventilation system with a dilution of waste streams by a factor of 4 x 104. Provide sufficient details regarding both the normal and emergency ventilation system flows to confirm the appropriateness of the dilution factor.

Fifth Response Submitted November 26, 2010 Chapter 13 will be re-written based on the basis document entitled "Fuel Damage Radiological Assessment". In this analysis, no credit is taken for dilution air.

Consequently, there is no longer a need to justify a dilution factor of 4 X 104.

9.2 Section 9.2 uses inconsistent units when discussing criticality protection for fuel in storage. Confirm that KIff is less than 0.8 for fuel in storage.

First Response Submitted June 10, 2010 P.8-5 Line 27, Line 42, and Line 43 should not refer to Ak/k. We are saying that keff is less than 0.8, which is a pure number, not a reactivity.

43

9.3 Figure 9-2 displays the fuel element cut-off saw. This saw is not described in the SAR. Provide a discussion of its use, when it is used, and the design features and controls in place to prevent cutting into the fissile material and control of cutting debris.

Second Response Submitted August 6, 2010 References to this saw will be removed from the SAR. This saw is not related to the safety margin associated with the operation of the reactor.

9.4 Provide a legible copy of Figure 9-7.

Seventh Response Submitted December 14, 2010 Figure 9-7 was a schematic of the Bay Campus Water System. The original drawing is somewhat illegible. Consequently this figure will be removed from the SAR. We will reference the original drawing as needed.

9.5 The references listed for this chapter lack dates and detail. Provide a more formal reference list for Chapter 9 that includes this information.

Ninth Response Submitted February 24, 2011 Most of the references are sufficient as listed. The SAR will not refer to specific revisions of the RINSC Emergency Plan, RINSC Operating Procedures, and RINSC Security Plan because these documents are revised and updated on a regular basis. The reference entitled "TRTR-5 Fabrication Requirements" will be removed. The reference to the BMI Cask Letter (Certificate) of Compliance will be removed because that shipping cask is no longer in service. References entitled "RINSC Quality Assurance Program", and "NRC Approval Letter" have been removed because they are not referenced in the text. The new list of references will include:

9-1 RINSC Emergency Plan 9-2 RINSC Operating Procedures 9-3 Removed - Need to remove this reference from SAR Page 9-5 Line 38 9-4 RINSC Security Plan 9-5 Removed - Need to remove this reference from SAR Page 9-7 Line 3 9-6 Removed - Not used 9-7 Removed - Not used 9-8 RINSC Safety Manual 9-9 IAEA-TEC-643, April 1992, Appendix N-3.1 "Nuclear Criticality Assessment of LEU & HEU Fuel Storage", Argonne National Laboratory The references will be re-numbered so that they appear in a numerical order that is consistent with the order in which they are referenced in the text.

44

10.1 Section 10.2.1 does not describe the design features of the beam port covers or administrative controls regarding beam port use that support the assumptions made in the loss-of-coolant-accident (LOCA) presented in Section 13.2.3 of the SAR. Provide a description of the design features and administrative controls that is consistent with the assumptions made in the LOCA analysis.

Fifth Response Submitted November 26, 2010 Section 13.2.3, Loss of Coolant Accident (LOCA), has been completely replaced with the analyses provided in the re-written section of RINSC SAR Chapter 13.2.3 "Loss of Coolant Accident".

If one of the beam ports is severed, all six beam ports are flooded because of the common interconnected drain lines. The revised LOCA analysis assumes that water can flow from each of the six beam ports onto the reactor floor.

Administrative controls are needed to guarantee that the flow resistance in each beam port is equal to or more restrictive than a round half-inch diameter, sharp-edged orifice at the exterior (beyond the reactor shielding) end of the beam port.

This could be accomplished by having a cover on the exterior of each beam port that seals the beam port, except for an optional hole with a diameter of up to one half inch.

10.2 Section 10.2.2 discusses administrative controls in place to limit draining of the reactor pool via the through-port. The text states that the through-port should not be opened for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> following reactor shutdown. The LOCA analysis presented in Section 13.2.3 of the SAR does not analyze pool drainage through the open through-port, nor does it provide a comparison of pool drain time for a closed and open through-port. Provide justification for the statement that opening the through-port 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after reactor shutdown is conservative.

Tenth Response Submitted July 15, 2011 The LOCA Analysis presented in section 13.2.3 of the SAR has been revised and is entitled "Section 13.2.3 Loss of Coolant Accident (LOCA) of the Safety Analysis Report of the Rhode Island Nuclear Science Center Reactor, Submitted May 3, 2004". In this analysis, the LOCA model is one in which an eight inch beam port extension is sheared off, and water drains through six sharp edged round half inch diameter round holes in the pool wall. It was conservatively assumed that since the drain lines of all of the beam ports are tied together by a common drain line, it would be possible for the drain line to back up and allow the un-damaged beam ports to fill with water, in which case each beam port would act as a drain path to confinement. Administratively, the area of each beam port that is open to confinement has been limited to a one half inch diameter hole. Consequently, the drain model considered a system which has six, half inch diameter holes, which corresponds to one for each beam port. The through port was not considered.

45

In response to this RAI question, the drain model for this analysis has been refined to include the through port experimental facility, and to provide a more realistic and less conservative representation of the experimental facility drain piping system. This analysis is entitled "LOCA Analysis Addendum. In this analysis, it is shown that each experimental facility drain, including the through port has a one half inch diameter orifice plate welded into the drain line. In each case, this empties into a one inch diameter drain line. All of these drain lines empty into a common two inch drain line that is at an elevation below all of the experimental facilities. The common drain line empties into a five inch line, which is reduced back into a two inch line that opens to atmosphere.

Experiment Drain System Since the drain line diameter gets progressively larger, and it opens to atmosphere, it is not possible for the drain line to get backed up. This fact, coupled with the fact that the elevation of the common drain line is below the lowest elevation of any of the experimental facilities means that for the Design Basis Accident in which one beam port is sheared off, there are only two drain paths for the coolant water:

1. Damaged Port Drain

2. Area of Port Open to Confinement 46

Shearing off a through port is not considered to be credible because there is virtually no access to the through port from the top of the pool. As shown in the photograph below, this port runs across the back wall of the pool, underneath the thermal column extension:

Consequently, the drain model in the Addendum has been changed from a tank in which six, one half inch diameter holes are at the elevation of the bottom of an eight inch beam port, to a tank in which there are two, one half inch diameter drain holes at the centerline elevation of the common two inch drain line. Under this scenario, the amount of time that it takes for the reactor pool level to drop from the pool level scram set point, to the top of the grid box is 19.34 hours3.935185e-4 days <br />0.00944 hours <br />5.621693e-5 weeks <br />1.2937e-5 months <br />.

The analysis done in "Section 13.2.3 Loss of Coolant Accident (LOCA) of the Safety Analysis Report of the Rhode Island Nuclear Science Center Reactor, Submitted May 3, 2004" shows that if the pool level does not drop below the elevation of the bottom of the eight inch beam ports before the decay power fraction is 0.827% after infinite reactor operation, then the fuel cladding will not be damaged. For the drain model in the addendum, it is impossible for the pool level to drop below this point due to shearing an experimental port because the through port is inaccessible, and the eight inch ports have the lowest drain level.

47

The revised analysis from 2004 includes a reference that provides data for the amount of time that it takes for decay power to reach various power fractions under various operating histories (Stillman). The analysis in the addendum indicates that the amount of time that it takes for the power fraction to reach 0.827% after infinite reactor operation is 162332 seconds (4.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />).

All of this means that it takes 4.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> for the decay power to reach the point at which the fuel cladding will not become damaged, given that the pool level is not below the elevation of the bottom of the eight inch beam ports. If a Design Basis Accident occurs, it will take 19.34 hours3.935185e-4 days <br />0.00944 hours <br />5.621693e-5 weeks <br />1.2937e-5 months <br /> for the pool level to drain to the top of the grid box, which is well above the elevation of the bottom of the eight inch beam ports. As a result, the power fraction will decay to a harmless level by the time that the pool level reaches the top of the core box. As part of the addendum analysis, the maximum allowable drain area between an experimental port and confinement was determined to be 1.48 in2 .

The through port is at a lower elevation than the beam ports. As stated earlier, a catastrophic failure of the through port is not considered to be credible. This is why the Maximum Credible Accident for the facility has always been the catastrophic failure of a beam port rather than the through port. Therefore, if the through port were to develop a leak, the pool drain time would be no greater than the time calculated in the LOCA Addendum analysis, provided that the area open between confinement and the through port is no greater than 1.48 in2.

Section 2.3.3 of the Safeguards Report for the Rhode Island Open Pool Reactor (4 April 1962) indicates that the original criteria for using the through port was that both ends of the port would have gate valves that could be closed in the event of a leak. Like the beam ports, the through port also has a one inch diameter drain line that can be used to isolate the port from the experimental drain system. Consequently, provided that the gate valves are in place, it is possible to close off all of the potential pool water drain pathways associated with this experimental facility.

Since the drain time estimation in the LOCA Addendum is on the order of hours for the pool level to reach the top of the core box, there is sufficient time to perform mitigating actions. In the event that it were deemed to be worthwhile to reopen the through port after it has been isolated due to a leak, it would be possible to move the core to the opposite end of the pool, and isolate that section of the pool from the end with the through port so that the high power section of the pool with the ports could be drained independently form the low power section of the pool where the core is positioned. This action makes the twelve hour delay time prior to opening the port, irrelevant.

48

10.3 Section 10.2.4 discusses the thermal column experiment facility. The discussion states that cooling air is required to remove heat generated in the thermal column graphite in order to prevent the graphite from overheating.

However, there is no analysis of the flow rate necessary to adequately cool the graphite, and TS 3.2.1 does not provide a set point for the safety channel associated with the thermal column. Provide an analysis of air cooling of the graphite that includes the minimum flow rate necessary to cool the graphite.

Explain the basis for the graphite temperature limit of 107 degrees C. (See RAI 14.74)

Second Response Submitted August 6, 2010 The temperature limit of 107 'C is cited in the original reactor operating manual

[Operation and Maintenance Manual, One-Megawatt Open Pool Reactor for Rhode Island Atomic Energy Commission, Providence, R.I., General Electric Document GEI-77793, October 1962] but no basis is given. Since the ignition temperature of graphite is well above this temperature (ranging from approximately 400 'C upwards, depending on the specific type and form of the graphite) it is reasonable to assume the limit was placed on the graphite temperature to preclude any unexpected releases of the stored lattice energy (i.e.,

Wigner energy) induced by neutron irradiation.

Numerous references address the release of stored energy in graphite including:

1. Radiation Defects in Graphite, R. H. Telling, University of Sussex, December 18, 2003;

2. Evaluation of Graphite Safety Issues for the British Production Piles at Windscale: Graphite Sampling in Preparation for the Dismantling of Pile I and the Further Safe Storage of Pile 2, B. J. Marsden et al., AEA Technology plc;

3. Nuclear Engineering Handbook, Harold Etherington, ed., McGraw-Hill Book Company, 1958.

As neutron-irradiated graphite is annealed (heated above irradiation temperatures) little or no stored energy is released below approximately 100-125

'C. A significant release peak occurs at approximately 200 'C, so limiting the RINSC thermal column graphite to temperatures below 107 'C provides a margin of nearly a factor of two to this energy release temperature. It is not credible that the entire thermal column contains lattice defects. The threshold energy for the lowest form of induced defect is approximately 1 eV (see Telling paper). After passing through the first several inches of graphite, the neutron flux in the thermal column is, as the name indicates, thermalized to an energy spectrum with a peak near 0.025 eV, which is below the threshold for inducing lattice defects.

The figure below shows the energy release rate as a function of temperature (from Ref. 2 above).

49

2 00

, ".." ....

. /

ft" +*,+.  : * * . +i**.*...*... - 3310 9s +

1.500 - -3171

++, +:

.,: : . . 331$713

.000 0ý0 --. .. ... -i7 -*

- +

10 003040 W

-0.500 Temperature (C Figure 8. Typical curves for rate of release of stored energy for a number of blocks in Pile 2 channel 33/57 TR. [Evaluation of Graphite Safety Issues for the British Production Piles at Windscale: Graphite Sampling in Preparation for the Dismantling of Pile I and the Further Safe Storage of Pile 2, B. J. Marsden et al.,

AEA Technology plc.]

The only graphite that is exposed to neutrons of energies of I eV or greater is the graphite in the first several inches toward the core. This section of the graphite protrudes into the reactor pool via the thermal column extension. Consequently, the cooling for this is provided by the reactor pooi rather than by airflow through the thermal column. The purpose of the airflow is to prevent Ar-4 1 from being released into the reactor room. As a result, no airflow rate specification is needed.

10.4 Section 10.2.8.2 references Section 10.8.3 which does not appear in the SAW.

Provide the correct reference or provide the missing Section 10.8.3.

First Response Submitted June 10, 2010 The reference should be 10.2.7.

11.1 The bases for TS 4.1 state that "Shim safety blade inspections are the single, largest source of radiation exposure to facility personnel." However, the safety blades are not listed explicitly in Chapter I11 of the SAR as one of the facility radiation sources. Verify that this statement is accurate and describe any additional radiological controls that are used for safety blade inspections that are not included in Chapter 11.

First Response Submitted June 10, 2010 50

P. 14-32 Line 43 Revise the statement that "Shim safety blade inspections are the single largest source of radiation exposure to the facility personnel" to say:

"Shim safety blade inspections have the potential to be the single largest source of radiation exposure to the facility personnel".

In keeping with ALARA principles, the shim safety blades are inspected in place. When a blade is inspected, it is raised to it's full out position. A visual inspection is made to the extent possible. Annual measurements are made of the blade drive times and drop times. These measurements provide assurance that there is no significant swelling.

We do not take credit for using a camera system because we do not want to be committed to having a functioning radiation resistant camera system available.

11.2 Section 11.1.1.1. The text references calculations of airborne activity that are described in Appendix A. This document was not provided with the license renewal application. Provide a copy of the referenced Appendix A.

First Response Submitted June 10, 2010 Please see attached Appendix A.

11.3 Section 11.1.4. of the SAR states that the Radiation Safety Office conducts routine radiation and contamination surveys. Discuss the bases of the methods and procedures used for conducting routine radiation and contamination surveys.

First Response Submitted June 10, 2010 A survey is an evaluation of the radiation hazards associated with the presence of radioactive materials and/or radiation sources under a given set of circumstances.

The Radiation Safety Office conducts routine radiation and contamination surveys described in standard procedures to evaluate basic radiological conditions at the RINSC. Surveys require the use of a calibrated survey meter with an appropriate detector as well as a wipe test for removable contamination.

Wipe tests may be counted using a liquid scintillation counter, proportional counter, gamma counter or other suitable radiation instrument depending on the isotopes thought to be present. An appropriately calibrated survey meter may also be used to count wipe tests.

Routine survey frequencies are determined by an evaluation of the radiological hazards likely to be present in the area, its frequency of routine entry or use, and ALARA considerations for the surveyor. The routine survey frequencies for all areas of the RINSC are reviewed and approved by the NRSC. Each survey consists of measurements of fixed and loose contamination and radiation levels.

In accordance with Technical Specification requirements:

51

" Weekly surveys are completed at intervals not exceeding ten days.

" Monthly surveys are completed at intervals not exceeding six weeks.

" Quarterly surveys are completed at intervals not exceeding four months.

" Annual surveys are completed at intervals not exceeding fifteen months.

11.4 Section 11.1.5. Describe the provisions for the use of extremity monitoring and the conditions under which extremity monitoring is used at the RINSC.

First Response Submitted June 10, 2010 The Radiation Safety Office issues personnel monitoring devices to individuals working in controlled areas of the facility, employs bioassay techniques and keeps records of doses received by all individuals for whom monitoring is required. Since the primary airborne contaminant is Argon-41 and it is an immersion hazard, personnel monitoring devices are sufficient to monitor exposure. Respirators are not used for routine entry into controlled areas since airborne levels of other contaminants during routine operations do not exceed 10% of derived air concentrations in 10 CFR Part 20.

The use of radiation monitoring devices for external dose is required for adults who are likely to receive an annual dose in excess of any of the following (each evaluated separately):

• 0.5 rem (5 mSv) deep-dose equivalent 0 1.5 rems (15 mSv) eye dose equivalent

• 5 rems (50 mSv) shallow-dose equivalent to the skin

• 5 rems (50 mSv) shallow-dose equivalent to any extremity The use of radiation monitoring devices for external dose is required for minors who are likely to receive an annual dose in excess of any of the following (each evaluated separately):

• 0.05 rem (0.5 mSv) deep-dose equivalent

• 0.15 rem (1.5 mSv) eye-dose equivalent

• 0.5 rem (5 mSv) shallow-dose equivalent to the skin

• 0.5 reins (5 mSv) shallow-dose equivalent to any extremity The use of radiation monitoring devices for external dose is required for declared pregnant women who are likely to receive an annual dose from occupational exposure in excess of 0.05 rem (0.5 mSv) deep-dose equivalent, although the dose limit applies to the entire gestation period.

The use of radiation monitoring devices for external dose is required for individuals entering a high or a very high radiation area.

52

The monitoring records include (whenever applicable):

(a) The deep dose equivalent to the whole body, lens dose equivalent, shallow dose equivalent to the skin, and shallow dose equivalent to the extremities; (b) The estimated intake of radionuclides; (c) The committed effective dose equivalent (CEDE) assigned to the intake of radionuclides and the information used to assess the CEDE; (d) The total effective dose equivalent; and (e) The total of the deep dose equivalent and the committed dose to the organ receiving the highest total dose.

11.5 Section 11.1.5. Describe the provisions for internal monitoring at the RINSC.

Include any provisions for use of radiological respirators at the RINSC.

First Response Submitted June 10, 2010 Bioassay is the determination of kinds, quantities or concentrations, and, in some cases, the locations of radioactive material in the human body. Bioassays may be conducted by direct measurement (in vivo counting) or by analysis and evaluations of materials excreted or removed from the human body.

The primary methods of bioassay used are the liquid scintillation counting of urine samples for a wide variety of radioisotopes and in vivo counting for gamma-emitting radioisotopes.

The primary purpose of the bioassay is the determination of the committed dose equivalent (CDE) and the committed effective dose equivalent (CEDE). CDE is the dose equivalent to organs or tissues that will be received from an intake of radioactive material by an individual during the 50-year period following the intake. The CEDE is the dose equivalent to the whole body from the internal uptake of radioisotopes.

Bioassays also act as an independent check on the adequacy of working habits and engineered safety features. When we are determining whether a potential intake should be evaluated, we consider the following circumstances:

" The presence of unusually high levels of facial and/or nasal contamination

• Entry into airborne radioactivity areas without appropriate exposure controls

" Operational events with a reasonable likelihood that a worker was exposed to unknown quantities of airborne radioactive material

" Known or suspected incidents of a worker ingesting radioactive material

• Incidents that result in contamination of wounds or other skin absorption

" Evidence of damage to or failure of a respiratory protective device 53