ML110140145

Letter to NEI Transmitting Revision to Staff SE on NEI Guidance Report 04-07

ML110140145
Person / Time
Issue date:	10/03/2014
From:	Mcginty T NRC/NRR/DSS
To:	Joseph E Pollock Nuclear Energy Institute
Smith Steve NRR/DSS/SSIB, 301-415-3190
References
GL-04-002, NEI 04-07
Download: ML110140145 (31)

Text

October 3, 2014 Mr. Joseph E. Pollock, Vice President Nuclear Operations Nuclear Energy Institute 1201 F Street, NW, Suite 1100 Washington, DC 20004-1218

Dear Mr. Pollock:

Nuclear Energy Institute (NEI) 04-07, Pressurized Water Reactor Sump Performance Evaluation Methodology, and its respective Nuclear Regulatory Commission (NRC) staff safety evaluation (SE) provides pressurized-water reactor licensees with an acceptable methodology to perform the evaluations requested in Generic Letter 2004-02, Potential Impact of Debris Blockage on Emergency Recirculation During Design Basis Accidents at Pressurized-Water Reactors. In its SE, the NRC staff concluded that the guidance report, as approved, provides an acceptable overall guidance methodology for plant-specific evaluations of emergency core cooling system and core spray system sump performance following postulated design basis accidents. These findings were transmitted to NEI by letter dated December 6, 2004 (Agencywide Documents and Access Management System (ADAMS) Accession No. ML043280631).

In the course of the NRC staff review of historical industry debris generation testing intended to justify zone-of-influence (ZOI) values smaller than those accepted in Table 3-2 of the SE, the staff determined that some ZOI values in the SE should be updated.

This letter transmits a revision to the staff SE that affects previously accepted SE ZOI values for Sure-Hold banded insulations and adds a separate ZOI for Transco reflective-metallic insulation (RMI) to address the potential impacts of aluminum RMI on chemical effects. An additional revision is being made to Table 3-3 of the SE to specify a debris size distribution for jacketed NukonTM with "Sure-Hold" bands that is consistent with the analysis in Appendix II to the SE. Revised SE pages are enclosed (Enclosure 1). Please use these pages to replace pages 30 and 38 from the December 6, 2004 SE. The staffs basis for these revisions and a discussion of how licensees might be affected by the proposed revision was provided in a letter to you dated July 29, 2010 (Accession No. ML100900172). This letter noted that stakeholder comments were being solicited on the draft SE revision and the draft guidance for installing "Sure-Hold" bands that were enclosed therewith. NEI provided comments to the staff via letter to Mr. William Ruland dated September 27, 2010 (Accession No. ML102710400). An attachment to the letter (Accession No. ML102710401) contained the specific comments identified by NEI.

On October 21, 2010, the staff held a public meeting to discuss the proposed revision to the SE.

NEI staff and members of the public attended the meeting and discussed the proposed revision with the NRC staff. The discussions that took place during this meeting are documented in the meeting minutes (Accession No. ML103081038). Based on the feedback to the NRC staff

during the meeting, some changes to the draft revision have been incorporated into the final revision. The most significant change involves the size distribution of debris that may be created during a loss of coolant accident from piping insulated using jacketed Nukon insulation with Sure-Hold bands.

Based on industry feedback, the staff has also incorporated changes to the draft guidance for use of Sure-Hold bands (Enclosure 2) so that the guidance matches the configuration tested when the insulation system destruction pressure was determined. The stakeholder comments, along with the responses by the NRC staff that were discussed during the meeting, are documented in Enclosure 3 to this letter.

Sincerely,

/RA Robert Taylor for/

Timothy McGinty, Director Division of Safety Systems Office of Nuclear Reactor Regulation

Enclosures:

1. Revised SE Pages

2. NRC Staff Guidance for Installation of Sure-Hold Bands

3. Discussion of Stakeholder Comments cc:

D. Walters, Nuclear Energy Institute M. Richter, Nuclear Energy Institute

ML110140145 NRR-106 OFFICE NRR/DSS/SSIB NRR/DSS/SSIB:BC NRR/DPR/PGCB NRR/DPR/PGCB/LA NRR/DSS/D NAME SSmith VCusumano SSanders CHawes TmcGinty (RTaylor for)

DATE 7/25/14 7/29/14 8/13/14 08/15/14 10/03/14

Revised Pages to NRC Staff SE for NEI 04-07 ENCLOSURE 1

30 Table 3-2 Revised Damage Pressures and Corresponding Volume-Equivalent Spherical ZOI Radii Target Material Destruction Pressure (psig)

ZOI Radius/

Break Diameter Reference for Test Description Protective Coatings (epoxy and epoxy-phenolic paints)

N/A 10.0 See note (1)

Darchem DARMET RMI Transco RMI with stainless steel foils 114 2.0 BWR Utility Resolution Guidance, Vol. 3 [27]

Transco RMI with aluminum foils 40(2)

See note (2)

BWR Utility Resolution Guidance, Vol. 3 [27]

Jacketed Nukon with Sure-Hold bands Mirror RMI with Sure-Hold bands 90(3)

See note (3)

BWR Utility Resolution Guidance, Vol. 3 [27]

K-wool 24 5.4 BWR Utility Resolution Guidance, Vol. 3 [27]

Calcium Silicate (Al. cladding, SS bands) 24 5.45 Ontario Power Generation [64]

Temp-Mat with stainless steel wire retainer 10.2 11.7 BWR Utility Resolution Guidance, Vol. 3 [27]

Unjacketed Nukon, Jacketed Nukon with standard bands Knaupf ET Panel 6

17.0 BWR Utility Resolution Guidance, Vol. 3 [27]

Koolphen-K 3.6 22.9 BWR Utility Resolution Guidance, Vol. 3 [27]

Min-K Mirror RMI with standard bands 2.4 28.6 BWR Utility Resolution Guidance, Vol. 3 [27]

(1) The testing is described in guidance sent to NEI on March 28, 2008 (ML080230462). The testing was re-evaluated by NRC staff in a letter to NEI dated April 6, 2010 (ML100960495) and a 4D ZOI was determined to be adequate for coatings systems covered by the testing. Licensees wishing to apply a 4D ZOI for epoxy coating systems should reference the report and validate that the testing is applicable to the installed systems.

(2) The reference value of destruction pressure provided here is appropriate for determining transportable foils using the baseline size distribution for RMI in Table 3-3. In addition, aluminum RMI should be treated as partially damaged cassettes between 4D (40 psig) and 17D (6 psig). Aluminum internal foils of partially damaged cassettes within this range should be considered for contribution to chemical precipitate generation if the cassettes are evaluated to transport to the sump or are directly exposed to sump fluid.

30a (3) The reference value of destruction pressure provided here is appropriate for target material installed on pipe sizes of 12-inch nominal diameter (or smaller). For target material installed on pipes of diameter D, where D exceeds a 12-inch nominal diameter, the destruction pressure is to be determined from the following equation:

Where:

Pdest (D)

=

Destruction pressure for target material installed on a pipe of diameter D Pdest (12)

=

Destruction pressure for target material installed on a pipe of 12-inch nominal diameter R(12)

=

Outer radius for target material installed on a pipe of 12-inch nominal diameter R(D)

=

Outer radius for target material installed on a pipe of diameter D The ZOI radius corresponding to Pdest (D) can then be determined according to the jet model in the ANSI/ANS 58.2-1988 standard. The purpose of this scaling equation is to account for increased stress on insulation banding as the target pipe size is increased beyond the 12-inch nominal diameter pipe on which banded target materials were installed during the testing that determined the reference destruction pressure value.

Formal debris generation studies have confirmed that insulation products having outer casings, jackets, or other similar mechanical barriers resistant to jet impingement yield smaller quantities of debris than do less robust materials. Various studies have also demonstrated dependence between the orientation of the jacketing seam relative to the jet and the amount of debris generation. This suggests that the integrity of the jacket during impingement is an important feature for minimizing debris generation.

Russell reports, for example, that double jacketing an insulation product with a second overcladding of stainless steel having a rotated, opposing seam was very effective at minimizing the distance from the jet to the onset of damage (OPG, 2001). As mentioned in Appendix I to this SE, any improvement in the mechanical resistance of the insulation product will help to avoid inflated ZOI volumes predicted by the ANSI jet model for very low damage pressures.

In applying the ZOI and destruction pressure values from Table 3-2, care should be taken to ensure that plant insulation installations to which these values are applied are at least as robust as the configuration that was tested. This guidance applies not only to the physical properties of the insulation or other target material (e.g., base material, jacketing, bands, latches, etc.), but also to the particular means of installation and attachment of jacketed insulations (e.g., ensuring band and jacket stresses for the installed configuration are bounded by testing, ensuring that more-limiting failure modes would not occur due to configuration or orientation differences, etc.). Scaling from the test condition to the plant condition should be applied where appropriate and justified. Specifically, as noted in Table 3-2 for certain types of insulation where excess band or latch stress was observed to be the failure mode, the reference destruction pressure should be scaled according to pipe diameter for insulation installed on larger plant pipe sizes. A reference that describes the testing that supports each of the values in Table 3-2 is provided in the final column of the table.

As noted above, the ANSI/ANS jet model has been proposed in the GR and found acceptable by the staff for the purpose of estimating potential damage volumes associated with empirically measured damage pressures. Various attributes and interpretations of the ANSI jet model are presented in Appendix I to this SE. Among those observations is the explanation of potentially exaggerated conservatism for very

)

(

)

12

(

)

12

(

)

(

D R

R P

D P

dest dest x

=

38 Table 3-3 NEI Recommended Debris Size Distributions Material Percentage Small Fines Percentage Large Pieces Fibrous Materials in a ZOI NUKON' Fiber Blankets 60 40 Transco' Fiber Blankets 60 40 Knaupf 60 40 Temp-Mat 60 40 K-Wool 60 40 Min-K 100 0

Generic Low-Density Fiberglass 100 0

Generic High-Density Fiberglass 100 0

Generic Mineral Wool 100 0

Jacketed NUKONTM with Sure-Hold Bands 100(1) 0 Reflective Metallic Insulation in a ZOI All Types 75 25 Other Material in a ZOI Calcium Silicate 100 0

Microtherm 100 0

Koolphen 100 0

Fire Barrier 100 0

Lead Wool 100 0

Coatings 100 0

Material Outside a ZOI Covered Undamaged Insulation 0

0 Fire Barrier (Covered) 0 0

Fire Barrier (Uncovered) 100 0

Lead Wool (Covered) 0 0

Unjacketed Insulation 100 0

Qualified Coatings 0

0 Unqualified Coatings 100 0

(1) Within a 90 psig isobar debris is considered to be 100% fines. If target diameter scaling is applied, as destruction pressure decreases to 20 psig, the amount of fines may be assumed to decrease linearly from 100% to 25% and the amount of smalls to increase from 0% to 75%. For pressures less than 20 psi the baseline 60/40 small fine/large piece size distribution is acceptable.

38a Staff Evaluation of GR Section 3.4.3.3: The baseline recommendations can be grouped as follows:

Materials for which adequate debris generation data exists to evaluate the debris size distribution, i.e., NUKON' fiberglass and DPSC Mirror' RMI insulations.

Materials deemed to have a size distribution no finer than the materials for which debris generation data is available.

ENCLOSURE 2 NRC STAFF GUIDANCE FOR INSTALLATION OF SURE-HOLD BANDS Purpose The intent of this document is to provide the U. S. Nuclear Regulatory Commission (NRC) staff guidance regarding the installation of "Sure-Hold" bands so that licensees installing these bands to secure plant insulation will achieve a robust configuration commensurate with the applicable destruction pressures provided in the staffs safety evaluation (SE) on the Nuclear Energy Institute (NEI) 04-07 guidance report.

The guidance herein describes methods acceptable to the staff for installing "Sure-Hold" bands.

These methods are ultimately based on insights derived from destruction testing of insulation secured with these bands. Alternate installation methods may also be found acceptable by the staff provided that they are adequately justified. For example, adequate justification for alternate methods could include analysis and insights using existing data obtained from previous destruction testing of insulation secured with "Sure-Hold" bands or the performance of additional destruction testing of "Sure-Hold" banded insulation.

=

Background===

A number of licensees have expressed interest in using "Sure-Hold" bands to secure problematic insulations currently installed in their containments as an alternative to insulation replacement. Although replacement or removal of the materials from containment is the most reliable way to prevent problematic insulation debris from being generated during a postulated loss-of-coolant accident (LOCA), testing has demonstrated that the installation of "Sure-Hold" bands to secure certain jacketed insulation materials would result in a configuration with significantly improved resistance to damage from LOCA-jet impingement. Depending on plant-specific strainer designs, debris types and quantities, as well as other factors, some licensees may consider installation of "Sure-Hold" bands to be an effective means to reduce quantities of post-LOCA debris that would potentially be generated and subsequently be available for transport and accumulation on the recirculation sump strainer.

"Sure-Hold" Band Design As used in the NRC staffs SE on NEI 04-07, the term "Sure-Hold" band refers to a specific band and latched fastener design that was tested for resistance to jet impingement by the Boiling Water Reactors Owners Group (BWROG) as part of its air jet impact test program. This testing is described further in Volume 3 of the BWROGs Utility Resolution Guidance (URG) document (ADAMS Accession No. ML092530505), which also includes a description of the "Sure-Hold" banding system that was used. Key attributes of the "Sure-Hold" banding system are summarized below:

Table 1: Attributes of the Sure-Hold Banding System Tested by the BWROG Band Material Stainless Steel Band Width 2 inch Band Thickness 0.062 inch (16 gauge)

Latch-and-Strike Fastener CamLoc Series 18L with Modified Strike The modifications performed to the latch-and-strike fasteners that were used for the successful "Sure-Hold" band tests are described in Volume 3 of the URG. A photograph showing the modified latch-and-strike design is also provided therein. Essentially, weld beads were placed at each end of the strikes J hook to prevent the latch from sliding off of the strike. As discussed below, air jet testing performed by the BWROG clearly demonstrated the importance of this modification with respect to the damage resistance of the target material.

Debris Generation Tests Involving "Sure-Hold" Bands In all, Volume 3 of the BWROG Utility Resolution Guidance describes six air jet impact tests that involved "Sure-Hold" bands, four of which used Nukon fiberglass insulation, and two of which used Mirror reflective-metallic insulation. The staff reviewed these test results in the process of developing guidance for the installation of "Sure-Hold" bands. A summary of applicable test parameters is tabulated on the following page in Tables 2 and 3, while some important observations regarding these tests are stated below:

The effectiveness of the modification to prevent release of the latch-and-strike fasteners used to secure the "Sure-Hold" bands was demonstrated by the testing.

Some degree of material destruction occurred in all of the tests involving jacketed Nukon secured with "Sure-Hold" bands.

Considering tests of Nukon involving the modified "Sure-Hold" bands, the maximum damage observed, which was less than 10 percent small fines, appeared to be the result of jacket separation at axial seams and insulation extrusion through these seams, rather than the result of "Sure-Hold" band failures. This is apparent from post-test photographs in Volume 3 of the BWROGs Utility Resolution Guidance.

The tests involving jacketed Nukon secured with "Sure-Hold" bands used longer targets (i.e., approximately 8 ft) than were typically used during the BWROG air jet impact test program (e.g., 2-3 ft). Therefore, a significant length of the target was not exposed to the full jet centerline pressure. As a result, the staff expects that the degree of damage indicated for the Sure-Hold band tests involving Nukon may have been nonconservative.

Although the entire length of the jacketed Nukon targets secured with "Sure-Hold" bands would not have been exposed to the full jet centerline pressure, the staff expects that the "Sure-Hold" bands nearest the target centerline would have been exposed to representative stresses. These bands did not fail. Therefore, the staff believes the banding and spacing is adequate.

The tests involving jacketed Nukon secured with "Sure-Hold" bands did not comprehensively evaluate limiting orientations for the jacket seam(s) or band latching, and further did not expose the band latching directly to the jet.

Although not fully relevant in that a different insulating material was involved, the two tests of Mirror reflective-metallic insulation secured with "Sure-Hold" bands provide a limited indication of the effectiveness of the Sure-Hold banding system for jacketing seam and band latching orientations that are directly exposed to the jet.

1 Jet impacts target at 3-oclock.

2 The tests summarized in Table 2 were performed with 3 axial lengths of jacketing, for a total target length of approximately 8 ft.

3 Bands were not uniformly spaced and indicated value represents an approximate average.

1 Jet impacts target at 3-oclock.

2 The tests summarized in Table 3 were performed with a single length of jacketing, for a total target length of approximately 3 ft.

Table 2: "Sure-Hold" Band Tests Involving Nukon Fiberglass Test L/D (Linear)

Target Pressure (psig)

Jacket Orientation of Jacket Seam(s)1 Jacket Axial Overlap2 (in)

Modified/

Unmodified Fasteners?

Approx.

Band Spacing3 (in)

Orientation of Band Latching1 Degree of Damage (Small Fines)

Notes 31-1 7

160 2-piece 12-o'clock

/ 6-'clock 2

Unmodified 10 9-o'clock

~ 22%

7 of 9 (unmodified)

"Sure-Hold" bands removed 31-2 11 110 1-piece 9-o'clock 2

Modified 10 9-o'clock

~ 3%

All 9 "Sure-Hold" bands remained in place 31-3 5

190 1-piece 9-o'clock 2

Modified 10 9-o'clock

~ 5%

1 of 9 "Sure-Hold" bands removed 31-4 5.3 190 2-piece 12-o'clock

/ 6-'clock 2

Modified 10 9-o'clock

~10%

1 of 9 "Sure-Hold" bands removed Table 3: "Sure-Hold" Band Tests Involving Mirror Reflective Metallic Insulation Test L/D (Linear)

Target Pressure (psig)

Jacket Orientation of Jacket Seam(s) 1 Jacket Axial Overlap2 (in)

Modified/

Unmodified Fasteners?

Approx.

Band Spacing (in)

Orientation of Band Latching1 Degree of Damage Notes 29-1 20 20 2-piece 3-o'clock

/ 9-'clock N/A Modified 14 3-o'clock None All 3 "Sure-Hold" bands remained in place 29-2 8.5 105 2-piece 3-o'clock

/ 9-'clock N/A Modified 14 3-o'clock None All 3 "Sure-Hold" bands remained in place Guidance for Installation of "Sure-Hold" Bands for Nukon Fiberglass In applying the destruction pressure values accepted in the NRC staffs SE (Table 3-3) on NEI 04-07 for jacketed Nukon secured with "Sure-Hold" bands, the staff considers it appropriate for licensees to ensure that their installed configurations are consistent with the guidelines provided below. These guidelines were developed in consideration of the test observations described above. Dimension provided below for overlaps and spacing between bands are representative of the tests performed of Sure-Hold bands for Nukon fiberglass.

"Sure-Hold" bands installed by licensees should be of a design that is equivalent to or more robust than the bands used for the BWROG air jet impact tests as described herein and in Volume 3 of the BWROGs Utility Resolution Guidance.

Fasteners used to secure "Sure-Hold" bands should be of a design that is equivalent to or more robust than the modified latch-and-strike design used for the BWROG air jet impact tests as described herein and in Volume 3 of the BWROGs Utility Resolution Guidance.

The center-to-center spacing for "Sure-Hold" bands should not exceed 10 inches.

An axial overlap of 2 inches should be maintained between adjacent segments of jacketing. A "Sure-Hold" band should be centered on the overlap between adjacent segments of jacketing.

An overlap of 1 inch should be maintained along longitudinal seams of the jacketing.

Adjustable latch-and-strike fasteners should be installed with sufficient tension to ensure a secure insulation.

"Sure-Hold" bands, fasteners, and associated components should be installed in a manner that is consistent with their design specifications. For example, some fastener designs may not be compatible with small-diameter piping.

"Sure-Hold" bands, fasteners, and associated components were designed and tested to support installation on piping insulation. The staff would not consider it acceptable for licensees to apply destruction pressure values approved for "Sure-Hold" bands on piping to insulation installed on non-piping components such as steam generators, pressurizers, reactor coolant pumps, etc.

The staff would not consider it acceptable for licensees to apply the approved destruction pressure values associated with "Sure-Hold" bands at locations on piping where local conditions would prevent installation of "Sure-Hold" bands consistent with these guidelines or manufacturers design specifications. Such conditions may exist due to valves, piping tees, interferences, etc.

In other respects not specifically addressed in this guidance (e.g., jacketing material, jacketing thickness, cloth cover, standard jacketing latches, etc.), licensees installations involving "Sure-Hold" bands should be at least as robust as the conditions tested.

Licensees seeking to install "Sure-Hold" bands to secure insulations that are physically different than Nukon fiberglass and Mirror reflective metallic insulation should perform material-specific testing to determine an appropriate destruction pressure. Testing would not be necessary in cases where an acceptable evaluation determines that a given material is sufficiently similar to either Nukon fiberglass or Mirror reflective metallic insulation to support application of the same or a similar (e.g., including margin to account for uncertainties associated with physical differences) destruction pressure.

ENCLOSURE 3 DISCUSSION OF STAKEHOLDER COMMENTS FROM OCTOBER 21, 2010, PUBLIC MEETING NEI Comment

1. The scaling evoked by the U. S. Nuclear Regulatory Commission (NRC) staff was developed by the Boiling Water Reactors Owners Group (BWROG) Utility Resolution Guidance (URG) to compensate for the smaller diameter test article used in the air jet tests when applied to larger diameter piping. The scaling is based on the observed failures that were associated with the banding/latching of the jacketing/covers and the assertion that the average target pressure is a better figure of merit than the jet centerline pressure. As noted in the URG safety evaluation report (SER) Appendix B, the simplified URG scaling algorithm is acceptable and is within +/- 25% of the results of a more rigorous algorithm such as the one derived in the URG SER. The following observations are noted:

a. In the last paragraph of page B-5 states that the scaling derivation is appropriate for smaller and larger pipes than tested. Therefore it is appropriate to derive an increased damage pressures for smaller diameter pipes than the tested pipes.

b. The technical basis for deriving the scaling algorithm is based on comparison of the cross-sectional areas of the tested targets and the plant targets. There are no limitations noted in either the SER or the URG concerning the application of the algorithm for non-piping structures and components. Therefore it is appropriate to derive scaled damage pressures for components such as the steam generator, pressurizer, and reactor coolant pump.

c. The scaling algorithm modifies the damage pressure to account for differences between the cross-sectional areas of the plant target and the tested target. In neither the URG or the SER to the URG are there specified limitations associated with application of the scaled damage pressure to derive the radius of the ZOI for a specific material/target, i.e. there is no requirement for further considerations of jet direction issues, jacketing material and thickness, jacketing seam orientations, jacketing material overlap, banding material and thickness, types of latches, etc.

NRC Staff Response:

Although scaling of the jet nozzle to the target diameter is a consideration, the target diameter scaling equation would seem applicable, regardless of whether one uses a target area average pressure or a jet centerline pressure. This is a more fundamental scaling concern that is a direct consequence of scaling down the cross-sectional area of the target without scaling the banding and jacketing dimensions in a similar manner.

a. Conceptually, scaling destruction pressure to smaller target pipe diameters may be reasonable in some cases. However, important assumptions are required that may be difficult or impossible to validate without actually performing tests. In particular, the target diameter scaling equation implies an invariant failure mode. However, at pressures higher than those under which the material was actually tested, more limiting failure modes that do not necessarily scale by target diameter can occur that may or may not have been observed during testing at lower pressures. The potential failure mode of extrusion, which was observed during testing of Nukon secured with "Sure-Hold" bands, is one example. Use of the scaling equation for "Sure-Hold" banded insulations on smaller piping could also lead to significant increases in the assumed destruction pressure, to the point where material located within a jet core would be assumed to survive the interaction with the loss-of-coolant accident (LOCA) blowdown jet and pressure waves. The NRC staff would not consider this modeling assumption realistic, since pressures within the jet core can fully recover to the initial stagnation pressure, which would very likely cause significant damage even to robust insulations.

b. The staff considers the target diameter scaling equation to be appropriate for scaling test results to plant insulation installations that are consistent with the material configuration tested. The staff understands that insulation installed on non-piping components and structures is secured in a different manner than piping insulation. In addition, the destruction pressure or failure mode may be affected by the geometry of the component.

For these reasons it would generally not be appropriate to scale a test performed on piping to non-piping components. The staffs guidance for installing "Sure-Hold" bands, further expresses the view that "Sure-Hold" bands should be installed in accordance with their design specifications and that it would not be considered acceptable for licensees to apply test results for "Sure-Hold" bands on piping to "Sure-Hold" bands installed on non-piping components and structures, with or without a scaling equation.

c. The staff has a standard expectation that good engineering practice would have licensees apply test data only insofar as it is representative of the plant condition. Based on its review of licensees responses to Generic Letter (GL) 2004-02, the staff has identified a number of instances where test data was inappropriately applied to plant material in a less robust configuration. To reduce the potential for future misunderstandings, the revision to the safety evaluation (SE) in Enclosure 2 includes clarification of this standard expectation, and references have been added to cite the source of the test data for each of the zone of influence (ZOI) values listed in the SE.

NEI Comment

2. The NRC staff stated that the scaling equation identified in the BWROG URG was inadvertently excluded from the SER. On Page 2 of the NRC letter, the staff stated that this scaling equation does not need to be applied to all types of insulation that were specified in the URG. Specifically for Transco stainless steel RMI or Darchem DARMET RMI, the staff determined that the existing 2D ZOI size is acceptable because RMI is not a significant contributor to strainer head loss. RMI, however, can be a significant contributor to water holdup within containment. Crumpled RMI debris can be retained by grating and meshed doors (such as personnel control doors) that are in the water flow pathway to the sump. The water flows in these constrictions can be significant and cause a decrease in the water level downstream of the RMI holdup. Additionally, crumpled RMI debris can cause partial blockage of drains causing increased water holdup and a reduced water level in the post-LOCA pool. Lower pool water level leads to higher velocities, and therefore higher propensity for debris transport to the sump. Also, the water level could be sufficiently reduced to allow air ingestion and vortexing to occur at the sump strainer.

Note that using the scaling equation could result in significantly more RMI debris than has currently been analyzed under the existing methodology. For example, if the destruction pressure is scaled for the diameter of the lower and upper portions of a steam generator, the ZOI size would increase from 2D to 13D on the lower portion of the steam generator (11.3 ft diameter) and 18D on the upper portion of the steam generator (14.6 ft diameter). As shown in the following figure [not reproduced here], applying this equation for RMI would result in the entire steam generator being enveloped within the ZOI. This would increase the quantity of RMI debris from approximately 2,500 ft2 (38 ft3) to over 60,000 ft2 (900 ft3). Nine Hundred cubic feet of crumpled RMI foils could cause a significant impact on ECCS performance as noted above. Selective dismissal of the application of scaling to certain types of RMI was not recommended by the URG and the associated SER for BWRs.

Selective dismissal of the application of the scaling to certain types of RMI as recommended by the NRC staff in the draft revision to the NEI 04-07 SER is not appropriate, and would be in direct contradiction to the previous NRC position.

NRC Staff Response:

The staffs basis for not applying the target diameter scaling equation to Transco stainless steel RMI and Darchem DARMET RMI is discussed in the July 29, 2010, letter transmitting the draft safety evaluation revision. Essentially, these materials were determined not to result in a significant contribution to strainer head loss. The supporting basis for the concern that RMI debris can be a significant contributor to water holdup in containment is not clear to the staff. RMI debris typically requires high velocities to transport. In reviews where blockage of mesh doors and gates was a potential consideration, RMI was typically not determined to be a credible blockage concern due to the inability to climb up on a vertical surface and block off the entire flow area of gratings and mesh doors. The head loss induced by RMI debris is also typically relatively small, and in some cases it has been observed to reduce the total head loss when mixed with fiber and particulate debris.

However, RMI has been considered for its potential to block floor drains and other restrictive upstream flowpaths in a deterministic manner that typically does not depend on the overall quantity of RMI generated in the containment.

Another important consideration is that the characterization for RMI in the NEI 04-07 guidance is conservative rather than realistic. As noted in response to comment 1.b above, the staff would not consider it appropriate to use the target diameter scaling equation to scale tests for insulation installed on piping to non-piping components. However, when considering the potential for large scale failures of RMI as discussed in the comment, the staff would expect that a significant fraction of the RMI would fail in the form of larger or intact pieces that would be captured on gratings or would sink in the containment pool.

It is also worth noting that there are a number of plants that are insulated with RMI for which a 28.6D ZOI was approved that have analyzed quantities of RMI that are larger than discussed in this response to the satisfaction of the NRC staff. The staff further noted that realistically accounting for large quantities of RMI in the containment pool may actually provide the benefit of presenting obstacles that could reduce the transport of problematic debris to the strainers. The staff also noted that the URG concern with RMI was strictly based on strainer blockage; the staff understands that BWR strainer approach velocities are typically somewhat faster than PWRs, and further, that for BWRs; RMI was typically not a large head loss impact.

In the case of Darmet, essentially no damage was seen in any of the tests, which means the current ZOI which assumes 75% destruction into small foils is already very conservative.

The staff has concluded this insulation system is sufficiently robust that if the true destruction threshold was determined and then scaled up using the scaling equation, that the current ZOI and characterization assumptions would likely still be bounding for this material. The staff is not aware of any PWRs with Darchem Darmet RMI.

NEI Comment

3. The draft revision to the ZOI guidance in the SER dismisses the application of the scaling algorithm to Cal-Sil because the staff determined that the existing ZOI value is sufficient without additional scaling because the ZOI value is based on a different test using a two-phase jet for which the damage mechanism observed during testing was related to tearing of the jacketing rather than failure of the jacketing bands. The scaling algorithm is based on scaling the jet centerline pressure by the ratio of the cross-sectional area of the test target and the plant target to develop the same force on the jacketing, i.e., the larger the cross section the lower the pressure required to generate the same force on the jacketing. The insulation is considered to fail when either the force on the jacketing exceeds the band strength and the band fails, or if the tearing strength is low, when the jacketing tears. Therefore the insulation failure mechanism should not impact the applicability of the scaling algorithm.

NRC Staff Response:

A number of tests were performed by OPG with calcium silicate. These tests showed that the orientation of the longitudinal seam was the dominant variable in determining the distance where the onset of damage occurred. This implies that the onset of damage is largely dependent on the local pressure acting on the seam as opposed to the target cross sectional area. The staff does not expect the local pressure at the longitudinal seam to be strongly affected by the target diameter. On the other hand, the scaling equation is based on the concept of preserving the total force on the target and hence the stresses in the banding and latching. As a result, the NRC staff did not consider it necessary to apply the target diameter scaling equation for this debris type.

NEI Comment

4. As noted by the staff in the proposed revisions to the SER ZOIs, the (Cal-Sil)

ZOI (damage pressure) value is based on a different test using a two-phase jet.

The tests referred by the staff are the OPG two phase tests. The OPG Cal-Sil test series were performed with a 2 target pipe diameter in contrast to the BWROG air jet tests performed with a 12 target diameter pipe. Scaling of the Cal-Sil destruction pressure should therefore be based on the 2 target diameter from which the damage pressure was derived.

NRC Staff Response:

The staff does not consider scaling of the OPG test results by target diameter to be necessary based on the response to Comment 3.

NEI Comment

5. In the basis for the proposed revisions to the SER ZOIs, the staff noted that the (Cal-Sil) damage mechanism observed during testing was related to tearing of the jacketing rather than failure of the jacketing bands. In page 29 of the SER to NEI-04-07, the staff concluded that comparison with OPG data on greater than 40 percent reduction in damage pressure for calcium silicate insulation the NRC staff position is that damage pressures for all material types characterized with air jet testing should be reduced by 40 percent to account for potentially enhanced debris generation in a two-phase PWR jet. It appears that the reason for the reduction in damage pressure was the different failure mechanism of the OPG Cal-Sil, i.e. the jacketing in the OPG test, as opposed to the banding failure observed in other testing and not in the potentially enhanced debris generation in a two phase PWR jet. The postulated potentially enhanced debris generation in a two phase PWR jet noted in the SER to NEI 04-07 is also currently not supported by the NRC as noted in the April 10, 2008 letter from John Grobe to Richard Anderson entitled POTENTIAL ISSUES RELATED TO EMERGENCY CORE COOLING SYSTEMS (ECCS) STRAINER PERFORMANCE AT BOILING WATER REACTORS. This NRC letter notes that Large-scale jet impact testing, such as that conducted by Swedish utilities at the Siemens - KWU facilities in Karlstein, Germany in 1994-95, has clearly demonstrated that saturated water jets are far less destructive than steam jets.

This suggests that single-phase air jet tests would be conservative relative to two-phase saturated water tests (e.g., similar to steam).

The NRC should consider taking this opportunity in revising the SER ZOIs to account for target scaling to also revise upwards all the damage pressures listed in the proposed revision to the SER ZOIs to account for the different failure mechanism in the OPG Cal-Sil that formed a large portion of the basis for reduction in the damage pressures and take in to consideration the staffs recent statement that suggests that single-phase air jet tests would be conservative relative to two-phase saturated water tests.

NRC Staff Response:

As noted above, the OPG testing of cal-sil showed that seam angle is an important factor in determining the distance at which debris was generated. There were some differences in the seam angles and other target parameters in the OPG tests and the BWROG air jet tests to which they were compared. However, OPG tested a number of different seam angles, and even in tests where less exposed seam angles were examined, the destruction pressures at which damage occurred (50-60 psid) remained well below the thresholds from the air jet testing (150 psid). For this reason, the staff does not consider it credible that the difference in observed destruction behavior was simply the result of a different failure mode.

To definitively address this question, it is likely that testing of the same configuration with different jet media would be appropriate. As noted in the SE on NEI 04-07, to some extent the applicability or degree of applicability of the 40% factor may depend on the target configuration (e.g. whether the target is jacketed), as well as the resulting failure mode. In lieu of testing, a comprehensive look at two-phase destruction results versus air jet test results would provide the best chance at demonstrating that the 40% factor is unnecessary.

The NRC staff has reviewed the basis for the 40% factor applied to destruction pressure and determined that the destruction pressures in the SE on NEI 04-07 are appropriate based on the test results. The BWROG provided information for the NRC staff to consider in this determination. The NRC staff has evaluated the differences between destruction pressures accepted for PWR and BWR debris generation analyses. The staff plans to finalize its position on the need for BWRs to adjust destruction pressures for ZOI determination as part of its ongoing work with the BWROG on sump related issues.

With respect to the German testing referenced in the NEI comment, the staff intent was to convey that the isobars from a single phase jet may be propagated further than those from a two-phase jet. Specifying whether a single phase or two-phase jet, at the same destruction pressure, will cause more damage was not the intent of the statement. Additionally, the next sentence in the staffs letter indicates that there are uncertainties associated with conclusions about the destructiveness of a two-phase jet relative to a single phase jet.

NEI Comment

6. The draft revision to the SER has additional selective application of the scaling algorithm to other materials (e.g. Koolphen-K) in contradiction to the recommendations of the URG and its associated SER. The NRC should either determine that the scaling algorithm is appropriate and apply it to all jacketed/banded insulation systems or determine that the scaling algorithm is flawed and should not be used. To selectively impose the scaling algorithm on certain types of insulation systems while dismissing it for others in not technically justifiable.

NRC Staff Response The primary objective of the NRC staff is to ensure that the technical positions provide reasonable assurance that licensees analyses are prototypical or conservative. Although consistency is also a desirable objective, the staff is not willing to pursue it where it results in burdensome positions that are in excess of what is needed to assure prototypicality or adequate conservatism. As explained in the cover letter to the draft revision, one of the fundamental aspects of the current ZOI model is that, for several reasons (model assumptions, reflections and deflections) it tends to incorporate conservatism with increasing significance as the ZOI size increases. In addition, large ZOIs are likely to destroy much of the insulation in the break compartment. However, these conservatisms do not necessarily exist at smaller ZOIs. For this reason, the staff considers it technically justified to apply the target diameter scaling equation to certain materials where it is necessary to achieve a prototypical or conservative analysis, and not to apply it in cases where it is not needed for such assurance.

NEI Comment

7. What is the basis for concluding that a 4D ZOI is appropriate for Transco RMI with Aluminum foils? To be consistent with the revised guidance for Nukon with Sure Hold Bands, shouldnt this ZOI size be based on the target diameter using the URG scaling equation rather than a specific ZOI size?

NRC Staff Response Air jet tests described in Volume 3 of the URG showed that the release of foils from Transco RMI cassettes is not very sensitive to the impingement pressure of the jet. This may be because the jet pressure needed to knock the Transco cartridges off of the pipe is much less than the jet pressure needed to liberate RMI foils from the cartridges. During the air jet tests, Transco RMI was tested at many distances from the nozzle and was observed to have been dislodged with limited damage all the way out to a distance that corresponds to approximately 17D for PWR primary system conditions. However, the amount of damage, regardless of the test distance for Transco RMI, was always less than 0.5% small foils.

While it is true that industry proposed a 75% small foil assumption within the ZOI for all types of RMI, and the staff accepted the assumption in the SE, it is not a prototypical assumption for Transco RMI. However, assuming a uniform distribution of material throughout a given ZOI, 0.5% for a 17D ZOI yields about four times more small foils than 75%

for 2D ZOI (the current RMI ZOI in the SE for Transco). This was discussed and acknowledged in Appendix II to the SE. The staff at the time allowed this potential non-conservatism because RMI is not a significant contributor to strainer head loss. The staff has concluded that this remains true for Transco RMI with stainless steel foils, however, it is not true for Transco RMI with aluminum foils due to the chemical effects potential of aluminum. The 4D ZOI assuming a 75% small foil generation as currently stated in the SE for all RMI is intended to result in the same quantity of small foils as would be obtained using 0.5% over a 17D ZOI.

NEI Comment

8. If it is necessary to consider the partial destruction of Aluminum RMI beyond the RMI ZOI due to the potential contribution to chemical effects, shouldnt it also be necessary to consider partial destruction of fiberglass blankets and other insulation materials outside of the ZOI since additional debris in the containment pool will also affect the chemical debris load? Selective application of the concept of partially destroyed insulation systems is not technically justifiable and could lead to non-conservative conclusions.

NRC Staff Response The staffs position relative to Transco aluminum RMI is specifically based on the outcome of tests with this material that demonstrated that, even at relatively high pressures (150-190 psid), cassettes were not substantially opened and foils were not substantially released. However, even at low pressures (e.g., < 10 psid), partially damaged cassettes were removed from the target pipe. Therefore, for Transco aluminum RMI, there is a potential for a significant quantity of aluminum inside partially damaged cassettes to be submerged in a containment pool and subject to dissolution that was not accounted for in the NEI 04-07 guidance and SE. The staffs review of destruction testing for other materials, such as fiberglass, did not show a large pressure range for which material was dislodged but not considered damaged because the cassette was not opened. In fact, destruction testing showed that the ZOI for jacketed fiberglass to be nearly the same as for unjacketed fiberglass. Furthermore, licensees chemical effects evaluations have shown that aluminum can be a dominant contributor to chemical precipitate formation. Therefore, the staff considered this treatment of aluminum RMI and other materials to be appropriate in this regard. As discussed in response to a previous comment, the staff further notes that there are differences in the level of conservatism associated with the ANSI model associated with larger ZOIs, such as for fiberglass, as compared to smaller ZOIs, such as that associated with aluminum RMI.

NEI Comment

9. Pages 3 and 4 of the NRC letter states that the staff considers 4D to be an acceptable ZOI for epoxy coatings with reference to acceptable test reports. In the revised Table 3-2, however, the SER simply provides the 4D ZOI as an acceptable ZOI without any requirement for licensees to reference the test report directly. Please clarify whether it is acceptable for licensees to reference the SER for the appropriate ZOI size for epoxy coatings and the various insulation types, or whether the NRC expects licensees to reference the original test documents.

NRC Staff Response The NRC expects licensees to reference acceptable test reports in order to ensure that their coatings are compatible with the specific types of coatings that were tested by the PWROG, if they intend to credit the 4D ZOI for epoxy coatings. Table 3-2 has been revised to note that a 10D ZOI is appropriate for epoxy coatings with a note that a 4D ZOI may be used by licensees that have validated their coating systems to be covered by the test report.

NEI Comment

10. Page 4 of the NRC letter states that Nukon with Sure Hold Bands should be considered to be 100% small fines. For licensees performing refined analyses, does the staff expect plants to assume that the small fines will be treated as fine debris, or is it appropriate to use a refined size distribution consisting of both fines and small pieces similar to what many licensees have done with other fiberglass debris?

NRC Staff Response The NRC staff expects licensees to have an adequate technical basis for the debris size distributions used in their analyses or to use conservative values. The staff is not aware of appropriate debris size distribution data for Nukon exposed to jet pressures that would exist within the ZOI for "Sure-Hold" banded Nukon. Specifically, the targets in the "Sure-Hold" banded air jet tests were 8 ft in length, and because they were placed quite close to the nozzle, only a limited area near the centerline of the target was exposed to the full jet pressure. In addition, these tests only exposed the insulation to the threshold destruction pressure, not the substantially larger pressures that would exist within the ZOI. Other tests typically used for determining the debris size distribution of Nukon rely on data for targets exposed to pressures (e.g., 20 psid) much lower than those which would occur within the ZOI for "Sure-Hold" bands (e.g., from 90 psid up to the initial system pressure of 2250 psid).

For this reason, based on the currently existent data, the staff recommends that licensees conservatively assume that Nukon secured with "Sure-Hold" bands within the ZOI would be destroyed as 100% fines. Based on the data available in Appendix II (Figure II-2) of the staff SE on NEI 04-07, the staff believes that a reasonable assumption for debris sizing would assume 100% fine debris for all material within the 2.4D (90 psi) ZOI. Between 90 psi and 20, as the ZOI increases due to scaling, the staff accepts that the quantity of fines will decrease. At 20 psi, the staff accepts that the debris will be 100% small fines with a 25%

fine and 75% small distribution. If the pressure is extrapolated to below 20 psi, the baseline characterization of 60 percent small fines and 40% large pieces may be applied. This is applied to Sure-Hold band installations only because of the relatively low pressure compared to the protection afforded by the insulation system. Licensees that decide to use this size characterization scheme may choose to discuss their methodology with the staff prior to finalizing their calculations. Staff recommendations are not regulatory requirements, and less conservative size distributions may be found acceptable by the staff on a plant-specific basis provided that an adequate technical and/or holistic supporting basis is presented.

NEI Comment

11. Draft SER Table 3-2, Footnote 3 concerning Sure-Hold bands states that the equation should be used for scaling down the destruction pressure for pipes with a larger diameter. As noted previously, in the URG, the equation was provided for use in scaling the destruction pressures down for larger diameter pipes and up for smaller diameter pipes. If this scaling equation is valid, shouldnt it be equally appropriate to scale up the destruction pressure for Nukon with Sure Hold Bands on smaller diameter pipes?

NRC Staff Response Please see staff response to comment 1a.

NEI Comment

12. The equation in Footnote 3 mentioned above, is defined in both the URG and the SER to NEI 04-07 as:

Pdest (D) = Pdest (12) x R (12) / R (D)

Where: Pdest (D) = Destruction pressure for target material installed on a pipe of diameter D Pdest (12) = Destruction pressure for target material installed on a pipe of 12-inch nominal diameter Note that this equation includes the insulation thickness in addition to the pipe diameter. However, in the SER to the URG, the equation was interpreted to be:

Pdest (D) = Pdest (12) x 12 / D This form of the equation does not take into consideration the insulation thickness, and results in a larger scaling factor. Please clarify which form of the equation should be used, and whether the original NRC acceptance of the equation (in the SER to the URG) would have changed if the equation had been interpreted as originally written.

NRC Staff Response The staffs revision to the SE on NEI 04-07 states that the radii intended to be used in the equation are the outer radii of the target material. Thus, the insulation thickness is included.

The staff recognized that a similar equation included in the BWROG URG did not include the insulation thickness. As noted in the comment, larger scaling factors will be achieved when the insulation thickness is neglected, and thus the version of the target diameter scaling equation included in the URG is more conservative than the version in the present SE revision. Appendix B to the staffs SE on the URG provides a brief derivation of the target diameter scaling equation used in the URG, and considers the URG formula an acceptable simplification of the equation included in the present revision to the SE on NEI 04-07. Thus, the staff would consider application of either version of the scaling equation (i.e., either including or not including the insulation thickness) to be acceptable. Although it represents a small change in methodology, the staff did not impose the simplified conservative version of the formula in the enclosed revision to the SE on NEI 04-07 to avoid unnecessary burden.

NEI Comment

13. On Page 30a of the draft SER revision, the staff states:

In applying the ZOI and destruction pressure values from Table 3-2, care should be taken to ensure that plant insulation installations to which these values are applied are at least as robust as the configuration that was tested. This guidance applies not only to the physical properties of the insulation or other target material (e.g., base material, jacketing, bands, latches, etc.), but also to the particular means of installation and attachment of jacketed insulations (e.g.,

ensuring band and jacket stresses for the installed configuration are bounded by testing, ensuring that more-limiting failure modes would not occur due to configuration or orientation differences, etc.). Scaling from the test condition to the plant condition should be applied where appropriate and justified.

Specifically, as noted in Table 3-2 for certain types of insulation where excess band or latch stress was observed to be the failure mode, the reference destruction pressure should be scaled according to pipe diameter for insulation installed on larger plant pipe sizes. A reference that describes the testing that supports each of the values in Table 3-2 is provided in the final column of the table.

How much rigor is the NRC staff expecting licensees to apply when analyzing whether a specific ZOI size is applicable to their insulation configuration? When performing their debris generation analyses, most licensees simply used the ZOI size corresponding to their insulation type without a detailed review of the band spacing, jacket strength, and other factors in the test configuration compared to the plant configuration. In many cases, these details may not be fully reported in the test reports or readily available for all of the insulation installed at the plants.

Does the NRC expect licensees who have not performed this comparison to reevaluate the ZOI used for each insulation type, even if their analysis has been previously accepted by the NRC?

NRC Staff Response The NRC staff considers the SE revision to be forward looking and does not expect that the additional clarification regarding the use of ZOI test data would require additional effort from licensees that have already completed their evaluations. However, it has always been a standard expectation of the staff that licensees apply test results only insofar as they are representative of plant configurations. Although in some cases information gaps exist, the staff believes that there is typically sufficient information available in test reports to ensure acceptable similarity between test conditions and plant installations. It is also worth noting that, for insulation types with relatively large ZOIs, there is typically not significant sensitivity associated with the specific insulation configuration. For example, the 17D ZOI established for Nukon applies to both jacketed and unjacketed material, and the specifics of the configuration are not consequential. However, for insulations with small ZOIs, significant reliance is placed upon the protective capabilities of jacketing and latching or banding systems. Therefore, it is important to assure that acceptable similarity exists. For example, during GL 2004-02 reviews, the staff has discussed with a number of licensees the configuration of calcium silicate jacketing and banding. Similarly, the staff believes that licensees installing "Sure-Hold" bands should do so in a manner that is consistent with the tested configuration as discussed in Enclosure 2.

NEI Comment

14. On Page 4 of the draft guidance for installing Sure Hold Bands, the staff states:

Sure-Hold bands, fasteners, and associated components were designed and tested to support installation on piping insulation. The staff would not consider it acceptable for licensees to apply destruction pressure values approved for Sure-Hold bands on piping to insulation installed on non-piping components such as steam generators, pressurizers, reactor coolant pumps, etc. For the CEESI air jet destruction testing as well as the OPG two phase jet destruction testing, all insulation materials were tested in a configuration consistent with insulation installed on piping. However, the destruction pressures and corresponding ZOI sizes determined from this testing have been widely used by licensees for insulation destruction on both piping and non-piping components including fiberglass and RMI on steam generators, pressurizers, and reactor coolant pumps. The NRC staff has frequently accepted the use of the pipe based ZOI sizes for large equipment without requiring additional testing or analysis. Since the testing for Nukon with Sure Hold Bands was conducted at CEESI in the same manner as the other Nukon and RMI tests, what is the staffs basis for not considering it acceptable to apply the ZOI for Nukon with Sure Hold Bands to non-piping components? Is the staff reconsidering their acceptance of using other ZOIs for non-piping components?

NRC Staff Response Scaling issues associated with the application of ZOI values determined from tests of insulation installed on piping to insulations installed on large components were not recognized at the time the NEI 04-07 guidance and SE were completed. However, they are important, because for PWRs, a significant contribution of the debris load may come from the insulation on steam generators and other components. It is clear that insulation tests of "Sure-Hold" bands on piping cannot be scaled directly to large components due to differences in configuration (e.g., band thicknesses would be much different, the potential for entire panels to be extruded, difference in attachment, etc.). For this reason, the staff does not consider crediting the SE destruction pressure associated with the installation of "Sure-Hold" bands to be appropriate for large components based on the testing of piping configurations. For other materials listed in Table 3-2 of the SE, the staff believes that the materials with larger ZOIs did not receive significant credit for jacketing and banding; as such, whether the material is installed on piping or large components, the staff believes that their application to larger components would be reasonable. The staff has recognized that conservatisms in the approved ANSI ZOI model for low destruction pressures can account for some potential non-conservatisms or unknowns in the debris generation assumptions. In response to other comments, the staff has already stated its views on stainless steel (relatively non-problematic) and Transco aluminum RMI (potential to remove intact cassettes), and the staff believes that these comments also reasonably apply to RMI installed on steam generators. The staff is not aware of any insulation currently installed on components for which non-conservative results may occur due to the application of destruction test results for piping configurations. The proposed SE revision is intended to be forward-looking, and if any non-conservative configurations were to be identified, the staff would address backfit implications on a plant-specific basis. It is and always has been the staffs position that licensees should consider their plant configurations relative to test conditions to ensure that the test results are applicable prior to application.

NEI Comment

15. The basis for the proposed 4D ZOI for Transco RMI with aluminum foils is not specified.

NRC Staff Response The staff provided a basis for the 4D ZOI for Transco RMI in the cover letter accompanying the draft SE revision. A 4D ZOI with the 75/25 % debris characterization proposed in the NEI 04-07 guidance report results in the destruction of a quantity of RMI foils that is approximately equivalent to the more limited damage actually observed for Transco RMI over a larger range of distances. The potential for additional damage beyond 4D is captured for aluminum RMI through consideration of removal of intact cassettes between 4D and 17D.

While more complicated multi-region ZOI / characterization models could be specified, the staff believes that the model specified in the enclosed SE revision captures the essential behavior and is sufficiently defined as to avoid misinterpretations in its application.

NEI Comment

16. What is the proposed applicability of any revised SE to current applications that have used NEI 04-07? Has an evaluation of backfit implications been performed and will the results of this evaluation be made publicly available?

NRC Staff Response The SE revision is intended to be forward looking. Based on its reviews of licensees responses to GL 2004-02, the staff understands that licensees currently do not have "Sure-Hold" bands installed. The revision concerning aluminum RMI is not being generically backfit as part of the SE revision process. However, the staff is aware of one licensee that has aluminum RMI installed and that a small number of other PWRs may also have installed aluminum RMI. Because of the demonstrated significance of aluminum to chemical effects evaluations, the staff is considering, on a plant-specific basis, whether to pursue this issue for such plants.

NEI Comment

17. Table 3-2, the ZOI entry for Protective Coatings should be listed as "4.0" not "4D",

in order to be consistent with the rest of the table.

NRC Staff Response This error will be corrected in the final revision of the SE. Note that the value in the table will be changed to 10.0 with a note allowing a 4D ZOI for plants referencing appropriate test reports.

NEI Comment

18. Cover letter page 2, last full paragraph states; The NRC staff determined that sufficient conservatism did not exist in the ZOI values for Sure-Hold banded insulations to allow the NRC staff to disregard applying the scaling equation to these insulations. The quantity of damaged fiberglass that could be generated when a scaled up ZOI is considered is potentially greater than fibrous debris quantity computed using the existing ZOI.

For example, after accounting for this scaling, the quantity of small fines generated during the "Sure-Hold"-banded air jet tests of jacketed Nukon would potentially result in more small fine debris than assuming 60 percent small fines per the baseline assumption in NEI 04-07 for the existing ZOI. In addition, given the observations of fibrous insulation testing without Sure-Hold bands as referenced in Appendix II of the SE, the staff determined that failure of the bands at some point as the target moves closer to the jet would likely result in the fibrous insulation being fragmented into predominantly fine debris. Lastly, due to the 8-ft length of the jacketed Nukon targets secured with "Sure-Hold" bands relative to the nozzle diameter of 3 inches, at the distances tested, the outer regions of the target would not have been exposed to the full jet centerline pressure. This may have resulted in an underestimation of the potential for debris generation when using Sure-Hold bands.

Comment: The highlighted sections in the paragraph above indicate supposition.

Is there a technical basis for predicting a higher quantity of fine debris?

NRC Staff Response The technical basis for applying the target diameter scaling equation is discussed further in Appendix B to the staffs SE on the URG. Basically, the intent is to ensure that the larger forces applied to larger targets (due to a larger area exposed to the same pressure) are accounted for, since equivalent pressures on larger targets can translate into larger stresses for banding, latches, and jacketing.

For the characterization, Appendix II to the SE on NEI 04-07 indicates that 100% small fines can be generated at pressures larger than about 17-20 psid. For Nukon secured with "Sure-Hold" bands, the pressures to which the material could be exposed could range from approximately 90 psid to the full initial system pressure. Based on the potential for complete or nearly complete destruction at much lower pressures, the staff considers it appropriate to assume 100% destruction into small fines at these larger pressures associated with "Sure-Hold" band/jacketing system failures. The staff makes decisions based on the best information available. In this case, to assume other than 100% destruction into small fines, in view of the lack of adequate testing or other information, would be inappropriate. The staff has agreed that at lower pressures the debris size distribution will likely be different and the staff has provided additional guidance allowing a variable size distribution depending on the destruction pressure. This position is based on currently available size distribution data.

Although testing at very high destruction pressures is likely to prove challenging, the staff remains open to revising this position based on valid test data that shows different behavior.

In response to the assertion that the language cited in the comment that implies supposition, the staffs choice of language reflects the question of absolute certainty versus expected behavior. For example, there is no direct testing that shows that if the pressure is increased that more fines will be generated from the Sure-Hold banded Nukon, but considering the available data there is an expectation that as pressure increases, the amount of fine debris will increase. In lieu of test data to the contrary, the staff considers it reasonable to expect that this trend would continue. A similar consideration was made regarding the exposure of the 8 foot-long targets with Sure-Hold Bands exposed to the very high but localized pressures from a 3nozzle. The staff does not have direct data showing the reduction in pressure at the edges of the target because measurements were not taken. However, that is the expected effect based on jet behavior.

NEI Comment

19. Cover letter page 4, first full paragraph states:-

The NRC staff further plans to revise Table 3-3 of the SE to specify a debris size distribution for jacketed Nukon with "Sure-Hold" bands that is consistent with the analysis presented in Appendix II to the SE. Appendix II documents that the staffs acceptance of the baseline size distribution of 60% small pieces and 40%

large pieces for Nukon fiber blankets was based on a ZOI size of 17D. As demonstrated in Figure II-2 from this appendix, the degree of insulation fragmentation is not constant with jet pressure, but increases as the jet pressure impacting the target is increased, approaching 100% fragmentation into small fines at impingement pressures of approximately 20 psig. Although jacketing secured with "Sure-Hold" bands would provide the underlying Nukon fiberglass greater protection than was afforded the targets in the tests shown in Figure II-2, much larger impingement pressures ranging from greater than 20 psig potentially up to the reactor coolant system pressure for targets near the break opening would be experienced. Air jet testing of jacketed Nukon secured with "Sure-Hold" bands did not result in the failure of the jacketing system; however, based on the dimensional scaling of the jet nozzle to the target, the target would not have been fully impacted by the peak jet centerline pressure. Furthermore, the jet centerline pressures tested were intended to establish the threshold of destruction and are not sufficient to characterize the resulting debris size distribution within the ZOI. Therefore, the NRC staff is planning to revise Table 3-3 of the SE to indicate that jacketed Nukon with "Sure-Hold" bands destroyed within a ZOI should be considered to be 100% small fines.

Comment: What is the basis or source of the statement highlighted above? The basis for this statement could not be found within the CEESI testing report.

The referenced Figure II-2 describes NUKON, Transco and Knauf. The NUKON product is not 100% destroyed to small fines at 20 psig. The curve shown is a best-fit curve for all manufacturers data. If you are describing using Sure Hold Bands and it can only be used for NUKON, then only the data for NUKON should be considered. Looking at the CEESI test data, the worst case reported for NUKON was 60% fines at 17 psig. In fact, testing at pressures higher than 20 psig yielded less than 10% fines and at 190 psig 25% fines. From reading this it appears other manufacturers data is being used to describe NUKON because it shows worse jet impingement destruction and it subsequently supports elimination of fibrous insulation.

NRC Staff Response Low-density fiberglass data from different manufacturers was plotted together in Figure II-2 of the SE on NEI 04-07 because the staff concluded that the base material was fundamentally similar for all the types of fiberglass listed in the figure. Further, the number of tests for any one type of material was insufficient to make conclusions on its own.

Regarding the testing at higher pressures, a number of the test targets exposed to pressures approaching 20 psid shown in Figure II-2 began to be only partially exposed to the jet as the jet pressure was increased, leading to the appearance of reduced damage.

This effect is discussed in Appendix II to the SE on NEI 04-07, and is exemplified by the difference in destruction seen for 3 nozzles as compared to 4 nozzles. The limited exposure of the target was an even more significant effect for the test at 190 psid, and for this reason, based on the staffs review of this information at the time the NEI 04-07 SE was issued, these tests were not considered appropriate basis for the debris characteristics assumptions.

The staff did not state that "Sure-Hold" bands can only be used for Nukon. The final bullet in the staffs guidance for installation of "Sure-Hold" bands for Nukon fiberglass recognizes that some installations of materials may be sufficiently similar to tested conditions that use of the air jet test data may be appropriate regardless of the manufacturer of the low-density fiberglass base material. However, the staff stated that the configuration of plant material should be sufficiently similar to test conditions with respect to the important parameters and provided some examples of parameters that may be important in certain cases.

NEI Comment

20. Guidance for Installation: Page 4, fourth bullet - An axial overlap of at least 2 inches should be maintained between adjacent segments of jacketing. A "Sure-Hold" band should be centered on the overlap between adjacent segments of jacketing.

Comment - The standard NUKON Jacket that is installed at all of the plants has a one-inch longitudinal overlap and that is what was tested with the "Sure-Hold Bands. What is the basis, or test report, that supports a 2 inch axial offset?

NRC Staff Response Based on the information available regarding the tests of Nukon with "Sure-Hold" bands, the staff understood that a 2-inch overlap was used. There is ambiguity concerning the specific configuration and whether the 2-inch overlap applied to the longitudinal seam, the radial seam between adjacent insulation segments, or both. Following the public meeting a licensee provided information that justified to the staff that a one inch longitudinal overlap was present during testing and that this is the design overlap for the Nukon system. The staff revised the guidance to reflect the one inch overlap.

NEI Comment

21. Guidance for Installation document: Page 4, fifth bullet - A circumferential overlap of at least 2 inches should be maintained along radial seams of the jacketing.

Comment - This is the standard instruction for NUKON; does this mean every joint must be inspected for the overlap?

These two bulleted items may be essentially trying to say the same thing. One statement is regarding the two-inch overlap between adjacent segments covered with a Sure-Hold band the second then stating a circumferential overlap of two inches along radial seams. The two bulleted statements could be read as one and the same regarding the overlap. Is it the NRCs intent to dictate an overlap for the longitudinal seam that runs the jacket length but have failed to adequately describe their meaning?

NRC Staff Response The staff expects that plant installations will be at least as robust as the tested configuration.

Because extrusion of the Nukon blanket from the seams between adjacent segments of insulation was demonstrated to be important in testing, the staff expects that licensees installations of "Sure-Hold" band-secured jackets with Nukon underneath will be at least as robust as the tested configuration with respect to overlap. It is up to licensees to determine whether each seam must be inspected in order to provide assurance that the plant configuration satisfies the test requirements. Check of overlap could be made during installation of the "Sure-Hold" bands, for instance. The staff intended to specify separately an overlap for the longitudinal seam that runs the length of the piece of insulation, as well as the overlapping radial seam between adjacent jacketed segments. The guidance document has been revised to improve clarity and differentiate between the staffs recommendations for the two seams. Additionally, with the recognition that the overlaps are nominal, and the longitudinal overlap is built into the jacket system, the guidance has been revised to remove the words at least from the overlap dimensions.

NEI also made some comments in the letter (ML102710400) that transmitted the specific comments addressed above. Some of these comments, which were not directly addressed in the responses to issues above, are paraphrased and discussed below.

NEI Comment

1. The greater damage observed in OPG tests was due to failure mechanism of jacket tearing and was not proof that a two phase jet is more damaging than an air jet. Therefore the 40%

destruction pressure reduction factor should be removed from ZOI determinations.

NRC Staff Response The NRC staff has reviewed the basis for the 40% reduction in target destruction pressures and determined that the destruction pressures accepted in the NRC staff SE on NEI 04-07 are appropriate based on test results.

NEI Comment

2. The NRC should conduct, with industry participation, a comprehensive and critical review of the existing tests and data used in support of debris generation models. Such a review should be performed in tandem with debris generation testing being performed by PWROG.

NRC Staff Response The NRC staff has reviewed many debris generation tests, as well as tests designed to provide information regarding the behavior of jets, in an attempt to understand how debris generation is influenced by the many variables that can affect the phenomenon. Based on the reviews which note significant differences in test methodologies and results, the staff has been unable to justify reductions in the currently documented ZOIs. The staff would consider participation in an industry sponsored working group designed to evaluate the information available on debris generation in an effort to refine the destruction pressures currently accepted in the SE. The staff notes that due to the proprietary nature of the testing conducted by the PWROG that participation in any working group may be limited and the results may not be appropriate for inclusion in the publically available SE. The staff has reviewed the PWROG test results and the ZOI methodology developed as part of the program, but has not developed an SE based on its findings. The staff will consider individual licensee application of the more recent PWROG testing on a plant specific basis.

NEI Comment

3. Debris generation methodology should be revised to allow a generalized credit for seam orientation since longitudinal seams on horizontal pipes are located only at nominally 0 and 180 degrees to prevent dripping liquid from entering insulation.

NRC Staff Response The staff expects worst case debris generation to be considered by each licensee. Because the orientation of the break with respect to the seam cannot be predicted, any credit would likely be based on statistical or probabilistic evaluations. The consideration for a seam orientation credit is beyond the scope of this revision, but will be considered by the staff as appropriate in the development of any future guidance in this area.

NEI Comment

4. Issuance of an SE revision should be withheld until new NRC staff critical review (including new ZOI testing) is completed.

NRC Staff Response The staff believes that the issuance of the revision in the near term is necessary to provide better certainty for licensees that are considering installation of Sure-Hold Bands as part of their solution to sump clogging issues. The staff has completed a review of more recent proprietary ZOI test results and evaluation methods. However, a safety evaluation on the topical report was not requested or completed. Therefore, the staff will issue this revision at this time and consider additional changes as more information is made available.

Additionally, as noted above, the staff is not clear on its ability to use proprietary testing as a basis for a generic revision to the SE. The staff will accept licensee evaluations that do not follow SE methodology if appropriate justification is provided for the alternate method. The recent test results and methodology may be referenced by licensees, but should be justified as applicable to plant conditions.