U-600295, Forwards Response to 850913 Request for Addl Info Re NRC Concurrence to Terminate Overinsp Program for Electrical Hangers,Hvac Duct & Duct Supports & Cable Tray,Conduit & Cable Terminations

Forwards Response to 850913 Request for Addl Info Re NRC Concurrence to Terminate Overinsp Program for Electrical Hangers,Hvac Duct & Duct Supports & Cable Tray,Conduit & Cable Terminations

ML20133N069
Person / Time
Site:	Clinton
Issue date:	10/11/1985
From:	Hall D ILLINOIS POWER CO.
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
U-600295, NUDOCS 8510280376
Download: ML20133N069 (28)

Text

,

U-600295 L37-8 5 (10 - 11)-L

_ _ , _ 1A.120 ILLINDIS POLVER 00MPANY i .ml i[] cusm. rmn srecs e o sa m cuuos uusnis con Docket No. 50-461 October 11, 1985 i

Mr. James G. Keppler Regional Administrator < (ff[..

' A Region III . _j U.S. Nuclear Regulatory Commission L7i 799 Roosevelt Road L_. __ _. , __ j Glen Ellyn, Illinois 60137 ' " Ulu Subj ect: Illinois Power Company Overinspection Program Termination Request

Reference:

Let ter , J. G. Keppler to Illinois Power Company, (Attention: Mr. W. C. Gerstner) of September 13, 1985

Dear Mr. Keppler:

The referenced letter requested additional information in support of the on-going NRC review of Illinois Power Company (IP) requests for NRC concurrence to terminate the Overinspection Program for electrical hangers, heating, ventilating and air conditioning (HVAC) duct and duct supports, and cable tray, conduit, cable and cable terminations. The information requested by the NRC is described on page 2, Items 1 through 4, of the referenced letter. IP responses to Items 1 through 4 are con-tained in Attachments 1 through 4, respectively, to this letter.

I trust that the information provided herein, together with previous information submittals, provides a sufficient informa-tion base for the NRC to arrive at a prompt conclusion to this issue.

n e: rours,

. . Ia 1 Vice President J EK/j sp Attachments cc: Director, Office of I&E, USNRC, Washington, D.C. 205:5 B. L. Siegel, NRC Clinton Licensing Project Manager NRC Resident Office Illinois Department of Nuclear Safety Mark Jason, Assis tant Attorney General, Sta te of Illinois 8510280376 851011 PDR ADOCK 05000461 OCT 151985 j;f t) .,,

ATTACHMENT 1 NRC Request Number 1 To enable Region III to determine if any new trends or types of discrepancies have been discovered, provide 1985 data to update Tables D-2, Number of Attributes Inspected by Type of Major Commodity, and D-3, Nonconforming Attributes by Cammodity, of your "Results of Quality Programs for Con-struction of Clinton Power Station" dated April 1985.

IP Response te NRC Request Number 1 The information requested is contained in Table 1-1 and 1-2 which follow. These tables contain both 1984 and 1985 data.

The 1985 data is presented alone and as cumulative data which includes 1984. In updating this data to include 1985, IP has not identified new types of discrepancies. The 1985 data c?oes show a trend toward continued improvement in results. This is demonstrated by review of information in this Attachment and in Attachments 2 and 3.

Following Table 1-2 is a discussion of the engineering evaluation of the nonconformances listed in Table 1-2 by type of commodity.

\

TABLE 1-1 NUMBER OF ATTRIBUTED INSPECTED BY TYPE OF MAJOR COlBIODITY Number of Inspected Items Cumulative January 1, 1985 Cumulative Through Through Through Commodity December 31, 1984 August 9, 1985 August 9, 1985 FV OI FV 01 FV OI Electrical Hangers 381,062 97,411 274,643 86,497 655,705 183,908 HVAC Duct 187,955 11,402 122,542 302,065 310,497 313,467 HVAC Hangers 109,117 12,266 39,813 211,968 148,930 224,234 Cable Trays 87,361 8,970 31,395 44,540 118,756 53,510 Conduit 22,716 1,056 26,328 3,806 49,044 4,862 Cable 22,020 8,059 19,205 2,599 41,225 10,658 Cable Termination 116,901 48,213 241,589 48,702 358,490 96,915 Total 927,132 187,377 755,515 700,177 1,682,647 887,554

TABLE l-2 h0NCONFORMING ATTRIBUTES BY COMMODITY

"' Cumulative Through January 1.1985 Through Cumulative Through December 31 1984 August 9, 1985 August 9 1985 Number og Number of Number of Number of Number of Number of Nonconforming Safety-Significant Noncontorming Sa fe t y-Signii t t an t Nonconforming Safety-Significant Commodity Att ribute Att ribut e s Nmconformances Att ribut e s Noncontorn.ances Att ributes Nonconformances Conduit Weld size 1.392 0 356 0 1.748 0 Supports l'nde rcut 1.165 0 178 0 1.34 3 0 overlap 434 0 31 0 465 0 Convexity 7 0 0 0 7 0 Concavity 22 0 86 0 108 0 1.ack of Fusion 313 0 44 0 357 0 Porosity 8 0 4 0 12 0 Slag 797 0 200 0 997 0 Crack 26 0 20 0 46 0 Reinforcement 3 0 0 0 3 0 Wrong hardware 313 0 74 0 387 0

• Hardware Missing 96 0 23 0 119 0 Hardware Loose 94 0 46 0 140 0 Incoirplete 8 0 1 0 9 0 Orientation / Configuration 166 0 51 0 237 0 Tolerance 907 0 224 0 1.131 0 Cle.iranc e/ Int e r f erenc e 32 0 3 0 35 0 kuu t i ny. 0 0 1 O l 0 Wrong Welds ibb o 33 0 321 0 Caps 215 0 62 0 297 0 Thread inp6es=nt 23 0 10 0 33 0 Arc Strike 2.754 0 25 0 2.779 0 C rioding 203 0 42 0 245 0 Den t / Bent / Warped 180 0 21 0 201 0 Couge/ Scratch / Cut 195 0 15 0 210 0 Bolt / Nut Broken 7 0 1 0 8 0 Costing Missing 112 0 0 0 112 0 Def ective Mat erial 48 0 0 0 46 0 Dirt / Debris  !! 0 0 0 11 0 Protection 1 0 0 0 1 0 kust 37 0 0 0 37 0

ii I

TABLE l-2 NONCONFORMING AT111buTES BY COMMODITY CL.uulative Through Janua ry 1, 1965 Through Cumulative Through Decembe r 31, 1964 August 9, l' 35 August 9, 1985 Number et Number of Nurber of Number of Number of Number of Nonconforming bafety-Significant honconforming Safety-Significant Nonconforming Safety-Significant Commodity Att ribute Att ribu t e s Nan conf ormance s Att ribut e s hanconformances At t ributes Nonconformances Conduit Holes 122 0 8 0 130 0 Supports Caps (damage) 36 0 10 0 46 0 (continued) ID Missing / Incorrect /

Danaged 307 0 12 0 319 0 Drawing Incorrect 213 0 74 0 287 0 1raceability 37 0 6 0 43 0 Inspection Error 64 0 5 0 69 0 Tot al 10,656 0 1,68 6 0 12,342 0 t

s~

• Cable Tray hangers Weld size 276 0 82 0 358 0 I nde rcut 41 0 55 0 96 0 (Nerlap 28 0 48 0 76 0 Concavity 1 0 2 0 3 0 1.ack of Fusion 37 0 24 0 61 0 F eren i t y 7 0 > 0 12 o Slag 29 0 12 0 41 0 Crack 3 f' 4 0 7 0 Reinforcement lo 0 0 0 19 0 Wrong Hardware 34 0 0 0 34 0 Hardware Ptissing 124 0 26 0 150 0 Hardware Loose 32 0 17 0 49 0 Inc omplete 1 0 0 0 1 0 Orientation / Configuration 22 0 9 0 31 0 Tolerance 79 0 37 0 116 0 Clearance / interference 2 0 1 0 3 0 Wrong Welde 108 0 10 0 lib 0 Gaps 9 0 5 0 14 0 1

TABLE l-2 NONCONFORMING ATTklBUTES BY COMMODITY Cumulative lhrough January 1,1985 Through Cumulative Through Decembe r 31, 1984 August 9, 1985 Aogust 9, 1985 Number of Number of Number of Number of Number of Number of Nenconforming Sa t et y-S ignif ican t Nonconforming Safety-Signiticant Nonconforming Safety-Significant Commedity Att ribut e Att ributes Nm cont c reances Att ribut es Non conf ormanc e s Attributes Nonconformances Cable Tray Arc Strike 37 0 25 0 62 0 Hangers Crinding 31 0 4 0 35 0 (continued) Dent / Ben t / Warped 7 0 1 0 8 0 Couge/ Scratch / Cut 30 0 12 0 42 0 Dirt / Debris 1 0 5 0 6 0 Holes 0 0 2 0 2 0 ID Missing / Incorrect /

D.maaged 4 0 3 0 7 0 Drawing Incorrect 12 0 11 0 23 0 Traceability 0 0 5 0 5 0 Inspection Error 5 0 7 0 12 0 Tot al 979 0 412 0 1,391 0 16 AC Duc t s and l Supports Veld stic 305 0 e8 0 3a3 0 l iti o rudercut 55 0 161 0 (Nerlap 103 L 1 0 10a u j Concavity 2 0 0 0 J o l Laca ut Fus te n 15.' O 25 0 l'/ / 0 Porosity 44 0 14 0 56 0 l Slag 3 0 0 0 3 0 1

C rac k 19 0 10 0 29 0 Wrong Hardware 83 0 86 0 169 0 Hardware Missing 176 0 84 0 260 0 Hardware 1,oose 220 0 34 0 254 0 Incomplete 47 0 47 0 94 0 Orient at ion /Cor f igu a a t ion 407 0 38 0 445 0 Tolerance 568 0 162 0 730 0 Clear.n(c/Interferente 9 0 10 0 19 u l

l ,

= ^ >*mw w ~ ~ " ' ww " -- ^ : s ~ ~ *~'-*M " w m m , _ _ _ e .. . ~~ , . . - .- s - -~- ~ ~ H'- - - ~ ^ ~

TABLE l-2 NONCONFORMING AURIBUTES BY COMMODITY cumulative Through N Decembe r 31, 1984 January 1. 1985 Through August 9. 1985 Cumulative Through August 9, 1985 Number of Number of Number of Number of Number of Number of Nonconforming Sa fety-S ignificant Nonconforming Safety-Significant Nonconforming Safety-Significant Cen:nooit y At t ribut e Att ributes Nericonformences Att ribute s Nan conf ormanc e s Attributes Nonconformances CVAC Ducts Routing 0 0 1 0 1 0 and Supports Wrong Welds 9 0 7 0 16 0 (continued) Cars 89 0 25 0 114 0 Thread Engagement 5 0 15 0 20 0 Arc Strike 89 0 9 0 98 0 Crinding 81 0 6 0 87 0 Dent / Bent / Warped 32 0 19 0 51 0

(,ouge /Scr at ch/ Cut 140 0 22 0 162 0 Bolt / Nut Broken 9 0 0 0 9 0 e Coating Missing 0 0 1 0 1 0 i Defective Material 20 0 0 0 20 0

@- Holes 84 0 27 0 111 0 Caps 34 0 6 0 40 0 D Missing / Incorrect /

Dasared 17 0 7 0 24 0 Drawing Incorrect 36 0 6 0 42 0 Traceat:111ty 3 0 0 0 3 0 Inspection Error 3 0, 1 0 4 0 Tetal 2.915 0 806 0 3.721 0 Cable Tray Weld size 0 0 2 0 2 0 krong Bazerare 2 0 0 0 2 0 Hardware Missing 26 0 26 0 54 0 Eardware lease 3 0 1 0 4 0 crientation/ Configuration 1 0 3 0 4 0 Tolerance 21 0 0 M Clearance / Interference 1 0 0 1

0 22 1

0 0

Eauting 0 0 1 0 1 0 tent / Bent / Warped 0 0 1 0 1 0

.___m., _

_ . _ . _ _ _ _ _ _ _ . _ . _ _ _ _ .___4 ._

t l

i 4

l 1

1 +

i I

TABLE l-2 "

I NONCONFORMING ATTk1BLTIES BY COMMODITY i

' ' ~ '

Cua.ulative Through January 1.1985 Through Cumulative Throuah I Decerber 31. 1984 August 9, 1985 August 9, 1985

! Number of humbe r of Number of humber of humN r of NucScr of hencenferming Satery-Significant konsouterming satsty-Significant hoacenforming Safety-Significant 4

Commodity Attribute Att ribut e s Eat conf ormances Att ributes Nanconformancer Attributes Nonconformances

.i i Cable Tray Bolt /Eut Broken 0 0 1 0 1 0 (ccetinued) Dirt / Debris 47 0 12 0 59 0 i Protectton 0 0 12 0 12 0 l Inoles 0 0 5 0 5 0 '

ID Missing / Incorrect /

Dmaged 8 0 1 0 9 0 Drawing incorrect 1 0 2 0 3 0 Traceability 0 0 1 0 1 0 Inspectico Error 1 0 0 0 1 0, Total 113 0 69 0 182 0

'q i Conduit Hardware Missing 4 0 3 0 7 0

! Fardware Loose 8 0 3k 0 46 u incomplete 1 0 0 0 I t trientation/Configuratten th O O O 18 0 Clearance /Inte terence I o 0 0 1 t i tending Radies e 0 0 0 4 0 i thread Engagement i te 0 0 t c d

le t / Bent /W r pe d 2 O I O 3 L touge/ Scratch /Lut  ! O O O 1 0 j t=fective Material 1 C 1 0 2 0 ID Missing / Incorrect /

J Damaged 16 0 5 0 21 0 Drawing incorrect 1 0 0 0 1 0 Inspection Error 1 0 0 0 1 0 Teta1 59 0 4B O 107 0

-i f

I 1

i l

}

I l

,l j TABLE 1-2 NONCONFORMlhG ATTRIBUTES BY COMMODITY 4

i j Cumulative Through Jan uary 1, 19b5 Through Cumulative Through i

December 31, 1984 August 9, 1985 Aueust 9, 1985 50eber of Number of Number of Number of humber of Number of l boncontorstng Safety-Significant koncontorr.ing hafety-Significant Nonconforming Safety-Significant Cmmodit y Att ribute Att ribut es Em conf ormanc e s Att ribut e s Nan conf ormance s g tributes Nonconforsancen Cable wrong Hardware 25 0 2 t> 0 51 0 liardware Missing 114 0 140 0 254 0 Hardware Loose 2 0 23 0 25 0 Incomplete 17 0 41 0 58 0 ,

Orientation / Configuration 12 0 26 0 38 0 Tolerance 17 0 11 0 28 0 Clearance / Interference 3 0 18 0 21 0 Routing 22 0 68 0 90 0 Bend Radius 25 0 e2 0 107 0 1 Termina' ion Error 0 0 11 0 11 0 3

&* Dent / Ben t / Warped 6- 0 2 0 8 0 couge /scr at ch/ Cut 141 0 34 0 175 0 Detective Material 1 0 0 0 1 0 Dirt /De b ris 0 0 2 0 2 0 Protection 0 0 7 0 7 0 1 ID Missing / Incorrect /

D.ade'd I2I U 4I O If'3 fl

, Drawing incrissct la 0 18 t- 24 0

) T axeabilit y 0 0 1 0 i fj Inspection trror 14 0 6 o 70 0 Tot al 533 0 556 0 1.089 0 1

5 Cable i Terminations Wrong Harduare 65 0 16 0 81 0 i Hardware Missing 20 0 21 0 41 0 Hardware Louse 43 0 40 0 83 0 Incomplet e 4. 0 16 0 60 0 d

Orientation / Configuration 15 0 21 0 36 0

Telerance 93 0 47 0 140 0 i

l I

I

- _ _ . --- --- . _ . - . . - - _ _ - _ . . . - . . . . . _ . - . _ - _ - .- -_ -- _ - . ._ = . - - .-- ..

TABLE l-2 NOhCONFORMING ATTRIBL"It.S BY COMMODITY Cumulative Through Janua ry 1, 1985 Thruugh Cumulative Through December 31, 1984 August 9, 1985 August 9, 1995 Nua.ber of humbe r o f Number of humber et Nun:ber of Numoer of honc on t c-rsin g Saf e t y-S ignif icar.t h0nconforming Satety-Signit hant hanconforming Safety-Significant 1

Cas. mod it y Attribute Att ributes Em conformances Att ribut es Nan coa f ormanc es _ Attributes hanconformances

! Cable Clearance / interference 3 0 0 0 3 0 3

Terminations send Radius 43 0 15 0 58 0 (conticued) Terutnation Error 106 0 67 0 173 0 Dent / Bent / Warped 16 0 13 0 31 0 couge/ Scratch / Cut 5 0 14 0 19 0 i Bolt /but Missing 1 0 0 0 1 0

! Defective hterial 5 0 0 0 5 0 ID Missing / Incorrect /

Dm aged 11 0 25 0 36 0

. Drawing Incorrect 4 0 0 0 4 0

? Traceability 1 0 0 0 1 0 Inspection Error 7 0 8 0 15 o a

Tot al 464 0 303 0 787 0 I

i 4

I s

i

_ - - . _ . - . _ _ _ - . _ _ . . _ . - . _ . - . - _ - - . . . - - - = . .-- -

ENGINEERING EVALUATION OF NONCONFORf tANCES

, LISTED IN TABLE l-2 BY TYPE OF COMMODITY l

1 l The results of the Overinspection Program were evaluated for i electrical hangers, HVAC duct and supports, conduit, cable tray, t

cable and cable terminations. The number of nonconforming attributes for each type of commodity is presented in Table 1-2, 4

together with an indication of the number of nonconformances that have been determined to be safety-significant. As is evident j from this table, there were no safety-significant nonconformancee i for any of the commodities. For each commodity, a discussion is

provided which explains the basic reasons why the nonconformances t l are insignificant and do not encroach on existing design margins.

Attachment 3 to this letter discusses the results of the engi-neering evaluation of the safety significance of identified l

nonconformances and provides further details on why they are not safety significant. ,

I 1 -

l Conduit Supports l

l The Overinspection Program identified 12,342 nonconformin attributes on conduit supports. As is discussed below, a 1 nonconforming attributes were judged to be insignificant and none were safety-significant.

A] proximately one-third of the nonconformances were very minor, .

{ which included documentation nonconformances (such as unsigned i j travelers), minor damage (such as gouges), and installation crrors (such as tolerance nonconformances, wrong welds, and wrong i hardware). These types of nonconformances do net significantly reduce the strength of the conduit support since the attributes 1 in question have little or no bearing on support strength.

Approximately two-thirds of the nonconformances (7,865 in total) l pertained to arc strikeu and welds between the conduit support (

and the supporting steel. Of these, 2,786 nonconformances were arc strikes and weld profile violations that generally are i

cosmetic and do not reduce the strength of the weld. Most of the

} remaining welding nonconformances involved weld size, undercut,

, slag, and other discrepancies which may contribute to a minor j

strength reduction of the weld but did not affect the capability i of the weld to satisfy design loading conditions. None of the

', nonconformances were determined to be safety-significant, Cable Tray Hangers 4

The Overins ?cetion Program identified 1,391 nonconforming attri-butes on caale tray hangers. In general, these nonconformances j vere similar to those identified on conduit supports. Approxi-

mately 52% of the nonconformances were very minor, having littic or no relationship to the strength of the hangers. The remaining

5 -

l j nonconformances all pertained to discrepancies in the welds

! attaching hangers to supporting steel. All of these welding i i nonconformances were either cosmetic or did not reduce the '

strength of the weld below that required to satisfy design i loading conditions. Consequently, none of the nonconformances were determined to be safety-significant.

I i

l HVAC Ducts and Duct Supports

The Overinspection Program identified 3,721 nonconforming attri-butes in HVAC. These nonconforming attributes affected both HVAC ductwork and the duct supports and primarily involved installa-i tion and welding nonconformances. As is discussed below, none of these nonconformances were safety-significant.

Over half of the nonconforming attributes identified were minor  ;

tolerance violations, incorrect orientation or configuration, i installation gaps in the ductwork, and loose, missing, or wrong i hardware. The nonconforming conditions involving tolerance, j orientation, and configuration were evaluated against the design i drawings and were determined to have no impact on operational i performance of the duct system or the support system. Installa-1 tion gaps occurred when ductwork sheet metal would deflect and not lay flat against a support or when a small portion of the gasket used in companion angle mating surfaces was missing or damaged. These gaps were evaluated and found not to affect the integrity of the companion anF les or duct support system.

Other  !

hardware nonconformances involved duct companion angles for which (

i design criteria had been revised subsequent to the performance of

! the Overinspection Program for the items. When compared against

! these revised criteria, the items were acceptable, i

i Nonconforming attributes involving loose and missing hardware primarily involved nuts and bolts used to connect duct companion angles. These were not tightened adequately, were improperly installed, or were missing from the connection. Wrong hardware

! was reported primarily for incorrectly-sized access door assem-l blies in the ductwork. None of these nonconforming attributes i

were determined to affect adversely the integrity of the connec-i tions or the function of the duct systems, f Ap roximately one-fourth of the nonconforming attributes involved we ding and consisted of minor nonconformances in weld size and "

! various weld defects, resulting in reduced capacity of the weld i connection. Each nonconforming condition was evaluated to

} determine if the connection, excluding the defective weld, would ,

i cause code allowable stresses to be exceeded. No nonconforming  :

! conditions were found to result in excessive stresses in the '

j ductwork or support system.

I i

r

Physical damage to HVAC ductwork and supports consisted mostly of scratches, gouges, arc strikes, and grinding marks. A number of pin holes in ductwork were the result of abandoned insulation pins. All cases of physical damage were evaluated to determine if the integrity of the duct system was violated or if a reduc-tion in strength of a duct support occurred as a result of base metal damage. None of the nonconformances was determined to affect the design function of the duct system.

Cable Tray One hundred eighty-two (182) nonconforming attributes were identified on cabic trays by the Overinspection Program. As is discussed below, none of these were safety-significant.

Seventeen (17) nonconformances involved the lack of proper support for cables which drop out of the bottom of ladder-type trays. This sup port was only required to prevent the weight of the cables from bending the supporting rung of the cable tray.

These nonconformances were determined not to affect the function of the cables or the cable trays' ability to function as de-signed.

Fifty-nine (59) nonconformances involved dirt and debris in cabic trays. This condition, though not desirable, was evaluated as having no impact on the system design capabilities.

The majority of the remaining nonconformances involved missing or incorrect tray identification tags or cable wear stripa, an inspection error dealing with cabic routing, or tolerance viola-tions. All of these were shown not to affect the tray design capabilities or the subsequent installation of cables in the trays because (1) other documentation properly identified the trays, (2) the wear strips provide an extra measure of cable protection, but are not necessary in and of themselves, (3) the cabic routing was verified to be correct, and (4) the tolerancea were not exceeded by a significant margin and the affected items were capable of performing their design functions.

Con duit One hundred seven (107) nonconforming attributes were identified on conduits by the Overinspection Program. As is discussed below, none of the nonconformances were safety-significant.

Twenty-three (23) of the nonconforming attributes involved documentation. Twenty-one (21) nonconformances were missing or damaged conduit identification labels, liowever, in each case, other documentation, such as innta11ation drawings, was available to verify the identity of the conduits. The remaining two documentation nonconformances included an incorrect conduit elevation on the installation drawing and the installation of an ,

l l

j i

i

. i i

i alternate bonding jumper on a flex conduit. A review of the l 1 installation drawings showed that the conduit was properly [

I 1

installed and that the drawing should be revised accordingly.

Also, the alternative conduit bonding jumper installation is now ,

! an accepted practice and installation documentation has been

revised accordingly.

}

i Seventy-eight (78) of the nonconforming attributes involved

! conduit installation discrepancies, such as installation outside i of specified tolerances , loose or missing hardware, or bend l 1

radius violations of flex conduit. The tolerance discrepancies

} were evaluated and found to satisfy the: functional requirements  ;

! of the design as installed. Loose and missing hardware discrep-l i ancies formerly involved conduit fittings used on conduit teruin- i

ations at electrical boxes. The loose fittings were evaluated i and in each case found not to affect the proper operation of the l circuit since the conduit is adequately supported by the elec- .

trical box and conduit supports. The missing fittings involved  !

j conduit bushings which are used for additional cable protection i

{ but are not necessary to the function or protection of the

cables. These cases were evaluated and found to have no safety i

! significance. The minimum bending radius discrepar.cica of the l flex conduit were evaluated and found not to affect the integrity of the cable within and hence were considered not to be safety-significant.  !

The remaining six nonconformances involved minor damage, predomi-

{ nantly dents or nicks, on flex conduit or rigid steel conduit.

l The dents were very small and did not affect the integrity of the ,

conduit.

i l Cable '

r The Overinspection Program identified 1,089 nonconforming attri-butes affecting cabic. As is described below, none of these nonconformances were found to be safety-significant. ,

Two hundred thirteen (213) of the nonconforming attributes

[

) involved documentation nonconformances, such as the absence of or '

1 damage to a cable's identifying tag. Since other documentation j- was availabic to verify that the correct types of cabic were properly installed, this type of nonconformance did not affect l the cable's capabilities.

! Three hundred twelve (312) nonconforming attributes were attrib-uted to missing hardware, including missing edge guards for  !

! cables, missing cable supports, and missing ground wires. Edge guards provide an extra margin of cable protection, but each multiconductor cabic and each conductor's insulation was provided with a protective jacket. Thus, the conductor insulation, as per  ;

design, was adequately protected even though edge guards were missing. With respect to the missing cable supports, the cable ,

j manufacturer's cable support requirements are less restrictive

! (

f .

{  ! i l

t i

l ,

I than project requirements. When missing supports for cables were l cvaluated against the cable manufacturer's criteria, it was

determined that the cables were adequately supported. Instances

{ of missing ground connections on cable tray sections and on a '

! flexible conduit connection were noted. These ground connections j are installed to provide a primary ground path for fault currents ,

I and also to provide for personnel safety. Alternating ground paths were determined to provide adequate grounding and this condition was found not to be safety-significant.

]

l Twenty-five (25) small cuts were reported in the insulation of  :

l the conductors near the termination points. These cuts, which ,

did not remove any insulation, were on conductors whose insula- l tion is rated 600 volts. The applied voltage was only 125V DC or  !

120V AC, which is far lower than the rated 600 volts. Also, the i

cuts in the insulation were all located inside a junction box,  !

j termination cabinet, or other such controlled environment which

! would prevent further damage. The locations of all cuts were

evaluated and found not to provide a grounding fault path. Thus, i

none of these nonconformances affected the design function of the I cables.

, One hundred fifty-nine (159) gouges and dents were reported in i r the outer jacket of cables: however, 100 of these were accounted l for by one low voltage instrumentation cable and the gouges werc

! determined to be cosmetic damage to the jacket. The conductor insulation was not damaged on any of these cables therefore, these nonconformances would not affect the design function of any of these cables.

3 The remaining nonconformances included incorrect, improperly installed, or loose hardware, which were primarily cable grips, 1

and cable routing errors (in the routing itself or in bend

{ radius). Each at these was evaluated and determined to have no '

! safety significance.

l '

l Cable Terminations i

! Seven hundred eighty-seven (787) nonconforming attributes in

! cable terminations were identified by the Overinspection Program.

l As is discussed below, none of these nonconformances was found to j have safety significance.

l One of the more frequently occurring nonconforming attributes (11%) involved lugs that were not tightly connected to the

! terminal block (inspector could move the lug from side to side). i j However, all lugs used were of the ring-tongue type, which were j specifically used to prevent the lug from coming off of the ,

l terminal block even if the hold-down screw is not properly I l

tightened. Thus, electrical continuity was ensured. l l Other nonconformances (30%) involved conductor termination

crrors. IP will be conducting preoperational s l which will include operation of all interlocks,ystem testing alarmn, lights, i

and relays. This will ensure construction completeness and i proper system operation and will identify nonconformances involv-ing electrical termination errors. Consequently, these noncon-formances were not safety-significant because they would have been identified and corrected even if the Overinspection Program had not been performed.

Discrepancies concerning bending radii of the control conductors near the termination point comprised 7% of the nonconformances.

The specifications for bending radius at CPS are more conserva- i tive than required by the manufacturer. The actual bending I radius was compared to the manufacturers' requirements and most were found to slightly exceed those requirements. These were further reviewed and found to be installed in a physically protected and controlled environment. Consec uently, even if the insulation at the bend were to have developec a crack, the connection would not have been grounded.

Bent lugs comprise 4% of the nonconformances in cable termina-tions. Information furnished by the suppliers of these lugs showed that these bends did not adversely affect the integrity or function of the lugs. Consequently, none of the bent lugs were safety-significant.

Approximately 18% of the nonconformances involved wire crimps in which the wire did not extend out the far side of the lug barrel.

Ilowever, in each case, it was determined that the wire was firmly secured in the lug barrel, and, therefore, the termination was able to perform its design function.

An additional 7% of the nonconformances involved documentation errors such as missing wire code labels. These labels are only for later convenience in maintenance and checkout. They are not needed for correctly terminating the wire.

About 10% of the discrepancies involved the use of wrong hard-ware. The majority were mechanical compression connections which were used instead of ring-tongue lugs. Ilowever, use of the compression connections was permitted by the equipment supplier.

Thus, these terminations were capabic of performing their in-tended design function.

Approximately 12% of the nonconformances pertained to spare conductors, trimming and cosmetic scratches to the lugs, untidy bundling of conductors in panels, and missing wear strips. Spare conductors, trimming of lugs to fit the terminal block, and cosmetic scratches would not prevent the terminations from meeting their design requirements. Neat bundling of conductors is not necessary, but is desirable to facilitate Inter checkout and maintenance. Also, wear strips are an extra measure of protection for the control wire, aut are not necessary because the wires are provided with a tough outer jacket which prevents damage to the underlying insulation. Consequently, none of those nonconformances were safety-significant.

1 i

! t l

None of the remaining 1% miscellaneous nonconformances were j determined to be safety-significant.

l I

j NONCONFORMANCE REPORTS CONTAINING [

! INSUFFICIENT INFORMATION l i

Two NCRs prepared during the Overinspection Program did not i contain sufficiently detailed information, and the nonconforming i j item was reworked prior to commencement of the S&L evaluation. ,

Therefore, it was not possible for S&L to evaluate the impact of i those nonconformances on the integrity of the affected item.

1 However, as is demonstrated below, S&L was able to determine that '

the nonconformances were not significant to safety.

One nonconformance involved a cut on a jumper wire inside the t valve limit switch enclosure for valve 1E12-F027B. The location l of the jumper wire and the cut on the wire was not documented in  !

l enough detail to evaluate its significance. However, in the (

! event that the jumper would fail, the valve indicating lights on  ;

j the Main Control Board would go out. Also, an alarm in the Main Control Room would be annunciated, indicating a loss of valve  ;

control power, t i

l Valve 1E12-F027B is the outboard containment isolation valve on '

l* the B-loop of the residual heat removal low pressure coolant injection line. This valve is open in all modes (normal and accident) of plant operatirn and can only be closed by indepen-  !

j dent operator action from :he Main Control Room. Valve  ;

1E12-F027B is provided to comply with 10CFR50, General Design  !

,! Criterion 56. There are no design basis events that require

! closure of this valve. Therefore, this would not result in a j safety-significant condition, f The second nonconformance also involved a cut on a jumper wire f f inside the valve limit switch enclosure for valve 1FC037. The 1 1 location of the jumaer wire and the cut on the wire was not  !

documented in enouga detail to evaluate its significance.

Therefore, it is not possible to evaluate whetaer this condition i might have been safety-significant. However, in the event that i the jumper fails, the valve would fail in its current position i l and the indicating lights on the Main Control Board would go out. I i Also, an alarm in the Main Control Room would be annunciated, I indicating a loss of valve control power. Valve 1FC037 is the  ;

i inboard containment isolation valve on the reactor vessel pool I j flooding line. This valve is normally open and closes on a  !

4 containment isolation signal. Automatic closure of outboard i containment isolation valve 1FC036 will provide for containment isolation. Therefore, no safety significant conditions would exist. i k l 1

i i

l j l i  !

- _,-%-m-www-y,--n,,--.. ..-.----, wee._ ,,rew.v-e,-,,.~m .--mmm.--- www. e .

NrTACllMENT 2 NRC Request Number 2 i To enable Region III to deteroine if the sample inspected is sufficiently large to be representative of the entire population, provide actual (in place of estimated) data for the percent of commodities inspected and the percent defec-tive similar to the data provided in the attachment to reference 5.

IP Response to NRC Request Number 2 The information requested is contained in Tabic 2-1 which follows. The actual percentages inspected and defective have not changed significantly from the estimates provided in IP's letter, U-60 0244 of Sep tember 5,1985. In general, those minor changes which have occurred have resulted in improved results. Tabic 2-1 demonstrates that the amount of hardware reinspected under the Overinspection Program is sufficiently large to be representative of the entire population.

TABLE 2-1 STATUS OF IP OVERINSPECTION PROGRAM August 9, 1985 i

Percent of Total Plant Percent Defective Inspected (By Att ribute)

Field Fie ld I

Verification Overinspection Verification Overinspection l

Electrical Hangers 18.6% 5.2% 2.0% 0.5%

HVAC Duct 29.6% 49.7% 0.8% <0.1%

HVAC Supports 36.2% 44.6% 0.7% 0.1%

Cable Tray 62.9% 17.0% 0.2% (0.1%

Conduit 30.8% 1.5% 0.2% (O.1%

Cable 61.2% 15.8% 2.5% 0.4%

Cable Terminations 41.1% 13.2% 0.2% (0.1%

i i

j ATTACHMENT 3 l

, NRC Request Number 3 i

l To enable Region III to assess the significance of the

nonconformances identified in 1985, provide the results of j safety significance evaluations for those nonconformances '

identified after December 31, 1984, and categorized as B (B1 j and B2) or C as in references 1 through 3. You need not j provide these results for nonconformances which would not j have been identified as discrepant under the new Visual Weld Acceptance Criteria (VWAC) issued by the Nuclear Construc-tion Issues Group, which is now governing document for j visual weld acceptance criteria at Clinton Power Station.

IP Response to NRC Request Number 3 j

The objective of the engineering evaluations performed on the nonconformances was to determine the potential signifi-cance to plar safety had the nonconforming condition (s) been undetected by the Overinspection Program. Thesc  ;

engineering evaluations demonstrated that the identified i nonconformances would not have impaired the ability of the  :

i components to perform their safety related design function.

l The design margins of each component, considering the reported nonconformances, were determined to be within the specified design limits.

I The engineering evaluations considered the potential effect j that all identified nonconforming attributes may have had on i the components. This evaluation addressed both singular and

cumulative effects (attributes which may not act indepen-l dently), as appropriate for the nature of the reported j nonconforming attributes and the affc. ted components. For example, if two or more nonconforming attributes affected a  :

, load bearing weld, the net effect of all nonconforming 3

attributes in reducing the weld capacity was considered.

t The results of the engineering evaluations on a component .

1 basis were divided into three basic categories as described l below. These categories have been developed in order to l

quantify the significance of the nonconformances with

! respect to the design or design margin. Additionally, in  !

! order to further define, qualitatively, the significance of the nonconformances for electrical hangers, category B 3 nonconformances, as defined below, were further subdivided ,

l into B1 and B2 categories based on a 10% threshold for j reduction of capacity or component design margin, i

4

l Category A - The nonconforming attribute (s) reported on l the components are acceptable because they do not affect the structural integrity or functional capability of the component.

l These items are not significant with respect to the plant design and, therefore, have no effect on the plant safety.

Category B - The nonconforming attribute (s) reported on the components resulted in an acceptable ,

reduction in the functional capability or '

structural integrity of the com?onent.

Electrical hangers have been subdivided into Categories B1 and B2 to further quantify the significance of the nonconforming attri-bute(s) as follows:

Category B1 - The reported nonconforming attribute (s) resulted in a potential reduction in capacity or component design margin of less than 10%.

Category B2 - The reported nonconforming attribute (s) resulted in a potential reduction in the capacity or component design margin equal to or grcater than 10%. ,

Category C - The nonconforming attribute (s) resulted in a i reduction in functional capability or struc- i tural integrity beyond that allowed by the plant design basis. There are no components in this category.

Table 3-1 summarizes the results of the engineering evalu- I ations in terms of the three basic categories and, for electrical hangers, shows the distribution between sub-category B1 and B2. Following Table 3-1 is a description of the results of the engineering evaluation which demonstrates why the nonconforming conditions are not safety significant.

Additionally, since of the identified noncon-formances relate to the majority welding and /or the strength of com-ponents, there are two general factors which should be understood since they significantly affect the relationship <

between nonconforming conditions and the safety of com-ponents. These factors are the margin which is built into the design and the inspection criteria used to determine i that an anomaly in a weld represents a nonconforming condi- ,

tion. I Design margin primarily affects the strength of connections and the strength of components and is defined as the differ-ence between the stress allowed by a particular engineering l

j design code and the actual stress the connection or compo-

! nent will be subject to in service. The design of all l electrical supports, HVAC supports, and HVAC duct maintain j the presence of design margins by ensuring that actual '

stress is less than code allowable stress. In the area of i hanger design, the design margins are high due to the i generic design approach employed by Sargent & Lundy. For l t

e xcmple , connections are designed in groups rather than  ;

l individually. The most highly stressed connection is ~

designed to be within code allowable stresses therefore,

all other connections within the group, which are not as highly stressed, have even greater design margins.  ;

Additionally, code writers include in the design process

! additional margin in the form of the difference between code i

allowable stresses and the stress that would cause failure '

of a component. This means that a component designed to  !

j withstand code allowable stress is actually capable of L y carrying more than the maximum design load and not fail. (

, t i For electrical sup i j to code allowable, ports, is a the average factor of 2.9. design Formargin, standardcompared HVAC  :

i hangers and ducts, the design margins, compared to code '

! a llowable , is approximately a factor of 3.0. When one also ,

! considers the design margin maintained by code writers, the f j total factors of safety for these components will be higher.

In addition to the margin contained in the design, hardware 1

fabrication and installation often adds significantly to the

! theoretical capacities of components, particularly for  ;

l welded connections. l For example, as part of a separate test program, IP had samples of welded connections for HVAC duct and hangers j removed from the plant and physically tested to destruction.

2 The welds tested were selected by random item identification and then inspection to obtain worst case samples. All of ,

the test samples contained defects, many contained gross defects.

The test specimens failed at up to 47 times the design load.

l In some cases, two of the three welds on hanger connection i j were removed prior to tecting and the joint still sustained i

1 loading above that required by the design.  ;

' Thus, when the margins in the engineering codes, the design  !

and the fabrication / installation are considered, it is not i surprising that the installed hardware well exceeds the  !

j s trength required for safe operation.  !

' In reviewing results of an inspection program such as the IP Overinspection Program, consideration must be given to the 1 i

criteria used by inspection personnel for determining I

whether the hardware is acceptable. At CPS, the visual

! i

! 1 1 i 1

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

inspection of structural welds under the overinspection l Program was conducted following the criteria contained in '

AWS Dl.l. Those results have been reported by IP.

The Nuclear Construction Issues Group, a joint effort by several participating utilities, prepared a document en-titled " Visual Weld Acceptance Criteria for Structural Welding at Nuclear Power Plants" (VWAC) which proposed criteria for visual weld inspections generally less rigid than AWS Dl.l. This document was endorsed by the NRC and is now being implemented at CPS. In order to determine the impact that VWAC would have on Overinspection Results, IP requested S&L to select a sample of 100 nonconformance reports which identified welding defects and determine the effect that evaluating the welds using the VWAC would have had. S&L's review concluded that of the 240 nonconforming attributes reported, 84, or over 30%, would have been acceptable if VWAC rather than AWS Dl.1 had been the govern-ing criteria at the time of the inspection. While this is not a rigorous statistical sampic, it does provide some perspective on the conservatism inherent in IP's Overinspec-tion Program results. It further supports IP's conclusions that there is high confidence in the quality of CPS con-struction, and that the nonconforming conditions identified by the Overinspection Program are not safety significant.

_ . . . _ _ __ __ _ _ _ .__._. ___.__ _.____.__ _ ______.-_._ _ __._ - __ _ _ _ _ _ _ _ _ _ _ . _ _ - _ _ _ . . ~ _ . _ _ _ . _ - - . - _ _ _ - _ _ . - _ _ _

i 1

) TABLE 3-1 SIGNIFICANCE OF IDENTIFIED NONCONFORMING f CONDITIONS EXPRESSED IN TERMS OF CAPACITY i

I OF DESIGN MARCIN REDUCTION:

1 Cumulative Data as of August 9, 1985

{

e j

^

Items With One or More Nonconforming Attributes i

i

! Sub Sub Category A Category B Category Bl* Category B2* Category C Total l Cccmodity (No Impact) (<10%) (210%)

1 e

, N

u i Electrical Hangers 1397 (44%) 1799 (56%) 1147 (36%) 652 (20%) 0 (0%) 3196 (100%)

1 I HVAC Duct 445 (40%) 680 (60%) NA NA 0 (0%) 1125 (100%)

I HVAC Supports 261 (%%) 220 (46 %) NA NA 0 (0%) 481 (100%)

j Cable Tray SS (55%) 71 (45%) NA NA 0 (0%) 159 (100%)

i

Condui t 70 (72%) 27 (28%) NA NA 0 (0%) 97 (100%)

i

! Cable 331 (45%) 406 (55%) NA NA O (0%) 737 (100%)

I Cable Terminations 424 (59%) 296 (41%) NA NA 0 (0%) 720 (100%)

• These are Subtotals of Total Category B Classification 4

l i -- . __ __ .-. - _ _ - - - - . ._

._ __ ~ ___ _ . . _ _ _ _ _ -- _ . _ _ __ _ ___ _ _ _ _ _ _ _ _ _

w #

j -

l 1 ,

l ENGINEERING EVALUATION OF CATEGORY B NONCONFORltANCES i

1 ELECTRICAL HANGERS l

l Nonconformances classified as Category B involved those noncon-j formances which required the comparison of the discrepancy to the weld capacity or component design margins. In many cases de- i tailed engineering analysis and calculational comparison to the 1 original design were required.

4 Discrepancies that resulted in potential weld capacity or com-ponent design margin reductions of less than 10% are classified as Category B1 and those with capacity / margin reductions equal to l

or greater than 10% as Category B2. The most prominent examples

involved welding discrepancies such as undersized welds, lack of

) fusion, overlap, slag inclusion and undercut. These nonconfor-t j mances were determined to reduce the strength of the connection, 1 out thev did not affect the capability of the weld to satisfy j design loading conditions, therefore, a Category B classification i resulted.

1

! Installation and damage nonconformances also often resulted in a 1

Category B classification. Installation discrepancies included

, 'coleranca and wrong hardware nonconformances which resulted in an increase in stress on a rtructural component. Damage discrepan-cies included dents, gouges, and cuts which resulted in a reduc- -

l tion of the component e capacity. However, in all cases of j_ instrilation snd damage discrepancies the structural component was able to satisfy the design basis requirements.

4 i As noted in the above table, six hundred fifty-two (652) electri-cal hangers Sad nonconforming attributes identified that poten-j tially could have reduced their capacity or design margin by j equal to or greater than 10%. The majority of these nonconform-l ind attributes were welding discrepancies. Welding nonconfor-

mances resulting in' Category B2 classifications on conduit i

supports and cabic tray supports primarily involved undersized welda , ' overlap, undercut , and sing inclusion. These nonconform-ing welds were evaluated by comparing the actual weld stresses j (excluding affected areas) to code allowable stresses. .

j Most'of the remaining nonconformances that resulted in a B2 clarsification were caused by improper installation or damage.

These conditions required an analysis which compared the stresses in the as-built component to code allowable stresses. In no case 4

were the actual stresses in the as-built component greater than '

l code allowabic stresses.

1 Y ,

I

HVAC l Nonconformances classified as Category B involved those noncon-  !

formances which required the comparison of the discrepancy to the weld capacity or component design margin. Engineering analysis was not required due to the revision of the design criteria subsequent to the performance of the overinspections. When compared to these revised design criteria, the items were accep-table.

The most prominent example of Category B nonconformances involved welding. Weld related deficiencies of this type included such nonconformances as weld size, lack of fusion, undercut and ,

overlap. In most cases, the defective weld could be ignored and the remaining welds would be adequate to maintain the system integrity. No nonconforming conditions were found to result in excessive stresses in the ductwork or support system.

Installation nonconformances were comprised of hardware that was either loose, missing, or the wrong size. These attributes involved primarily nuts and bolts used to connect duct companion angles which were not tightened adequately, were installed crooked, or were missing from the connection. Wrong hardware was reported primarily for incorrectly sized duct access door assem- )

b lies .

Physical damage to HVAC ductwork and supports consisted mostly of scratches, gouges, arc strikes and grinding marks and pin holes.

All cases of physical damage were evaluated to detercine if the integrity of the duct system was violated or if a rr. duction in strength of a duct support occurred as a result of case metal damage.

I CONDUIT, CABLE TRAY, CABLE AND CABLE TERMINATIONS Nonconformances classified as Category B involved those nonconfor-mances which required the comparison of the discrepancy to the component design margins. In some cases, detailed engineering analysis and calculational comparison to the original design were required.

Typical nonconformances identified by the Overinspection Program that resulted in a Category B classification are missing hard-ware, control conductor bending radius and cuts on conductor insulation.

Missing hardware nonconformances were comprised mostly of missing cable supports and missing edge guards for cables. With respect to the missing cable supports, the cable manufacturer's cable support requirements are less restrictive than project require-ments. When missing su the cable manufacturer'pports for cables s criteria, it waswere evaluated determined against that the cables were adequately supported. Edge guards provide an extra margin of cable protection, but each multiconductor cable and each conductor's insulation was provided with a protective j acket . Thus, the conductor insulation, as per design, was ade-quately protected even though some edge guards were missing.

Bending radius nonconformances of control conductors occurred near the conductor termination point. The specifications for bending radius at CPS are more conservative than required by the manufacturer. The actual bending radius was compared to the manufacturer's requirements. Those found exceeding these re-quirements were further reviewed and found to be installed in a physically protected and controlled environment. Con s equently ,

even if the insulation at the bend were to have developed a cra ck, the connection would not have been grounded.

Cuts on conductor insulation nonconformances occurred near the termination points. These cuts, which did not remove any insula-tion, were on conductors whose insulation is rated 600 volts (the applied voltage was only 125V AC, which is far lower than the rated 600 volts). Also, the cuts in the insulation were all locked inside a junction box, termination cabinet, or other such controlled environment which would prevent further damage. The locations of these cuts were evaluated and found not to provide a i grounding fault path. Thus, none of these nonconformances affected the design function of the cables.

Two nonconformances involving cuts did not contain sufficiently detailed information for purposes of this evaluation, and the nonconforming items were reworked prior to commencement of the S&L evaluation. Therefore, it was not possible for S&L to evaluate the impact of those nonconformances on the integrity of the affected items. However, as discussed in Attachment 1 to this letter, S&L was able to determine that the nonconformances were not safety significant.

F ATTACHMENT 4 NRC Request Number 4 To enable Region III to determine that a sufficient sample of both conduit and cable tray supports have been overin-spected, provide an approximate percentage of cable tr .. ,

supports and conduit supports in the total population s.*

overinspected electrical raceway hangers.

IP Response to NRC Request Number 4 There are 39,356 safety related electrical han s b' itt .

plant. Of these, 31,405 (80%) are conduit han, '+ i 7,951 (20%) are cable tray hangers. BA Field Verittuation has reinspected 7,305 electrical hangers. Of these, approx-icately 5,990 (82%) are conduit hangers, and the remaining 1,315 (18%) are cable tray hangers. IP Overinspection has reinspected 2,029 electrical hangers. Of these, approx-imately 1,664 (82%) are conduit hangers, and the remaining 365 (18%) are cable tray hangers. This demonstrates that the percent reinspected for these two items reflects their respective plant populations and that a sufficient sample of both conduit hangers and cable tray hangers have been reinspected for IP's conclusions to be meaningful for both, i

I l - . ._ _ . _ _ _ -