ML20084S916
| ML20084S916 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 01/31/1977 |
| From: | Johari Moore SOUTHERN CALIFORNIA EDISON CO. |
| To: | Engelken R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V) |
| References | |
| 10CFR50.55E, NUDOCS 8306200076 | |
| Download: ML20084S916 (40) | |
Text
{{#Wiki_filter:w i OO 00, e i i, v Southern Califomia Edison Company g~ R O. eOx 800 2244 WALNUT GROVE AVENUC' JACK S. MOORE ROSCMCAD, CALtFORNIA St77C, TELEPe<oset ~....., January 31, 1977 ~ Mr. R. H. Engelken, Director Office of Inspection and Enforcement U. S. Nuclear Regulatory Commission Region V Suite 202, Walnut Creek Plaza 1990 North California Boulevard Walnut, Creek, California 94596
Subject:
Docket Nos. 50-361 and 50-362 San Onofre Nuclear Generating Station, Units 2 and 3
Dear Mr. Engelken:
Our letter of November 19, 1976, submitted an interim report concerning field welding and inspection of devices commercially identified as Nelson Studs.
- Enclosed, in accordance with 10CFR50.55(e), are twenty-five (25) copies of the final report regarding this deficiency, en-titled, " Nelson Stud Problem, Final Report, San Onofre Nuclear Generating Station, Units 2 and 3."
If you have any questions regarding this report, we would be pleased to discuss this matter with you at your convenience. Very truly yours, s .l;'.',?~/I.b.ie J't ! i y' b't.). /n i / /, s ? >. fr '~h Enclosures Dr. Ernst Volgenau (NRC, Director I&E)V cc: I l rj bla o l 8306200076 770131 l PDR ADDCM 05000361 S PDR Ab I -M+Ab m g$
0Q 00 Attachment NELSON STUD PROBLEM, FINAL REPORT SAN ONOFRE NUCLEAR GENERATING STATION. UNITS 2 AND 3 .? PURPOSE r The purpose of this report is to provide the informa-tion pursuant to 10CFR50.55(e)(3) subsequent to notification of a reportabic deficiency as provided by the Southern California Edison Company (SCE) to the Nuclear Regulatory Commission (NRC), Region V office on October 21, 1976. The subject deficiency is associated with the installation of structural steel attachments known as Nelson Studs in reinforced concrete structures. SCE has identified that inspections of the stud attach-ments, in the form of selective bend tests, have not been docu-mented or controlled in accordance with approved procedures and specified requirements. This omission constitutes a significant breakdown in that portion of our quality assurance program related to the attachment of these studs. This final report is intended to document the results of the investigations conducted concerning the installation of the attachments, and to establish their ade-quacy and the final disposition of this problem. BACKGROUND Before the placement of concrete for the construction of floors in specific locations, appurtenances commercially identi. ficd as Nelson Studs are secured to structuraH steel members, principally by resistance welding. (i.e., An electrical current is passed through the Nelson Stud and the structural steel member, causing localized molting and fusion at the interface.) The studs are subsequently embedded in concrete and serre primarily to provide a shear connection between the steel rnd the concrete. The integrity of the weld between the Nelson Stud and the structural steel is visually inspected and tested by selectively bending the studs after completion of the welding process. (The l bending of the stud does not impair the functioning of the stud I as a shear anchor.) Bend tests are required en the first two l studs welded to each structural steel member zud additionally to ~ l 1% of those welded thereafter to that member, j m
1.n e. oO OO =- Attachment - Page 2 Nelson Stud Problem, Final Report San Onofre Nuclear' Generating Station, Units 2 and 3 .l DISCUSSION 1. Description of the Deficiency Attached as Appendix A hereto is a self-contained report which addresses the specific deficient conditions relative to installation of the Nelson Studs. The following discussion addresses aspects of the deficiency related to the SCE quality assurance program in addition to those discussed in Appendix A. In reviewing the circumstances surrounding the reportable deficiency, we have identified the following: a. On October 8,1976, it was noted.by Edison that Nelson Stud welding done at the jobsite was not receiving the in-process bend testing previously described. i b. Subsequent investigation disclosed that Bechtel inspec-l tion procedures did not adequately provide for the i documentation of Nelson Stud testing and that work procedures, which required that testing be performed, were not being followed. At the time the deficiency described above was discovered, c. Nelson Studs had been embedded in concrete in the follow-ing safety-related buildings: Unit 2 containment, Unit 2 containment penetration building, radwaste building, and the control building. 2. Analysis of Safety Implications As noted above, the Nelson Studs provide a shear anchor to permit composite concrete / steel action by the floors. A reduction in the ultimate integrity of the. shear anchor. has no effect where the margin in. shear strength remains adequate for the design loading. Where it has been determined that additional shear anchors should Ens included to assure that an adequate design margin is maintained, alternative corrective actions have been developed. These are discussed in Appendix A. I em en m-7 ,m,y,p.--- -g -a,w9-
3 .l oo 0 0 .~;. Attachment - Page 3 'f" Nelson Stud Problem, Final Report 5' San Onofre Nuclear Generating Station,. Units 2 and 3 CORRECTIVE ACTION Appendix A includes a detailed description of corrective action taken with respect to the Nelson Stud problem. This corrective action is briefly summarized as follows: a. A Stop Work Notice was issued to preclude further welding or embedment of Nelson Studs until proper work controls and documentation requirements were established. b. 100% of all exposed Nelson Studs were tested to the same criteria as the 1% sample bend test specified for in-process inspection. c. Failure data were tabulated from the 100% bend test and utilized in establishing a conservative projection of embedded Nelson Stud integrity. d. A comprehensive program of. procedure development and upgrading, personnel retraining and process surveillance was implemented to allow Nelson Stud welding to be resumed. e. Personnel training sessions have been conducted in order to familiarize supervision, field engineers and quality control engineers with Nelson Stud welding procedures. f. Audits of welding and the related inspection were performed. g. Additional quality assurance, quality control and field welding engineers have been assigned to the San Onofre Units 2 and 3 jobsite. r o -.,-~-n.-- _ ~
~ 7. J. oo oo y.- '^ APPENDIX A . *\\ s* i NELSON STUD PROBLEM i l FINAL REPORT l l SAN ONOFRE NUCLEAR GENERATING STATION, UNITS 2 AND 3 l t t h January 31, 1977 l s a l I l* 4 k -h s y ...a..
OO OO! CONTENTS s 9r Section_ P_ age ^ l.0 PURPOSE 1 2.0 BAdKGROUND 1
3.0 DESCRIPTION
OF DEFICIENCIES 2 4.0 ANALYSIS OF SAFETY IMPLICATIONS 2
5.0 TECHNICAL EVALUATION
OF THE DEFICIENCY 5 6.0 CORRECTIVE ACTION TAKEN 12
7.0 CONCLUSION
16 Appendix Al Statistical Analysis of Strike Test Data T_ ables 1 Field Strike-Test Data and Evaluation for Areas which Exhibit Acceptable Stud Failure Rates 2 Field Strike-Test Data and Evaluation for Areas in which Studs are Non-Essential Structural Elements 3' Field Strike-Test Data for Areas Associated with Greater Than Allowable Failure Rates 4 Evaluation of Data for Areas Associated with Greater Than Allowable Failure Rates Figures 1 through 4, Control Area Steel Framing Plans Sketches 1 and 2, Repair Procedures i O \\ l so . - - - -. ~. - -, - -. y
l' I Oo oO ....? APPENDIX A NELSON STUD PROBLEM FINAL REPORT 5' SAN ONOFRE NUCLEAR GENERATING STATION, .A UNITS 2 AND 3 1.0 PURPOSE The purpose of this report is to provide final data and information as required by 10CFR50.55(e)(3) pursuant to notification of a reportable deficiency. The deficiency is related to a breakdown in construction practices and 1n a portion of the QA program at the San Onofre Units 2 and 3 construction site pertaining to the installation of welded studs.
2.0 BACKGROUND
Prior to the placement of concrete for building floors in certain locations, devices commercially identified as Nelson Studs are attached to supporting structural steel members. These studs are then embedded in the concrete and serve-to provide a shear connection between the steel and concrete, and thus develop a composite structural system. The studs are installed by a semi-automatic process with a welding gun connected to an electric power source. Inspections of the stud attachments, in the form of selective bend or strike tests, although performed to a limited extent, were not adequately documented or con-trolled in accordance with approved procedures and speci-fied requirements. This omission, found on October 8, 1976, constitutes a breakdown in that portion of the quality assurance program related to the attachment of these studs. .On October 9, 1976, inspection was performed on all-accessible studs. The inspection consisted of a hammer strike test performed to bend each of the accessible studs to at least 15 degrees from the original axis. It was revealed that the stud installation was potentially deficient and that inspection and welding. procedures had not been fully implemented. A Stop Work Notice was issued on October 11, 1976, stopping all concrete placements involving Nelson Studs and stopping further Nelson Stud installations. L y
U V o o oo l Nonconformance Report (NCR) W-021 was issued on.0ctober 13, l 1976, identi?ying specific areas where welded studs were in l question. Related lift numbers and drawing ndmbers were also identified. Exposed studs found to be deficient were replaced and confirmed by tect to be adequate.
3.0 DESCRIPTION
OF DEFICIENCIES 3.1 Procedures'Not Used 3.1.1 Crafts personnel were not consistently using available procedures for stud welding. 3.1.2 Construction Field Engineers (CFE's) were not using available procedures during stud welding inspection activity. 4 3.1.3 Quality santrol Instruction (QCI) had not been written for the corresponding stud welding process. As a result, inspection of stud welds and the associated testing were not fully implemented or documented. 4.0 ANALYSIS OF SAFETY IMPLICATIONS Field strike-test data on welded studs were obtained from beams with exposed studs at construction openings and at other accessible locations where concrete had not yet been placed. The field test data are sufficiently representa-1 tive to permit a statistical evaluation of the areas in question. At certain locations, the data indicate abnorm-ally high stud failure rates which deserve detailed analy-sis and corrective action. A statistical evaluation of the field test data has been performed for the purpose of categorizing the failure rates and proj ecting at various confidence levels, the number of studs that could be relied upon to perform as designed in the existent, installed beams. The statistical projection of the number of reliable studs for the various categories of beams, together with the calculated minimum number of studs required for. each beam, are the basis for verifying the adequacy of the composite structural system. Any deficiency which is identified by this method will be 0 e-e.- "q sI**m-g-MflP y*'**_"
- y.
e
i corrected by repair measures performed on the exIs't.ing beam installation to assure conformance with the criteria and bases of the Final Safety Analysis Report. Conformance is subject to the qualification that' the analytical treat-ment of '" incomplete composite action," as described in Sections 5.1 and 5.4 of this report, is in accordance with current research recognized by the AISC Specification Committee, but which has not yet been formally incorporated in the ~ AISC Specification as referenced in the SAR. Official adoption is expected in 1977. Based on the foregoing general criteria, the following categories were established: (A) Areas which exhibit acceptable stud failure rates: For these areas the test data on welded studs indicate that either one of the following conditions is met: 1. Stud failure rates are acceptable according to industry precedent and do not jeopardize t).e structural requirements. 2. The projected number of reliable studs exceeds the acutal minimum required according to structural design calculation. Consequently, in these areas the structural integrity has not been compromised, and the structural system is in full conformance with the basic design criteria. The Turbine Deck at Elev. 72'6", Units 2 & 3; the Radwaste Area Floors at Elev. 24'-0", 37'-0" and 50'-0"; and the Tankage Area Roof belong to this category, and their data are summarized in Table 1. (B) Areas which were completely open and accessible at the time of the reported problem: 9 In these areas,.the deficient stud installation is traceable to specific construction practices and/or operations which have been positively identified and subsequently eliminated. The studs in these areas were' inspected under. strict enforcement of revised procedures and repaired or replaced as required. New studs in these areas were installed and inspected according to proper procedures. The control Area Roof at Elev. 85'-0", North Side; and the Penetration Area Floors above Elev. 30'-0" belong to this category. i is 1 1 l l-
rG oo oo ay, (C) Areas in which the studs are non-essential structural components: The original structural design of the floor system in these areas included studs for developing composite two'way flexural action and for promoting a nominal anchorage between the steel beams and the thick con-crete slabs. It nust be noted,that this nominal anchorage is a matter of engineering judgment, and basically it is not intended to fulfill a specific structural-function derived from analytical considera-tions. Upon final analysis, it is evident that the reinforced concrete slabs acting independently in flexure are capable of developing the structural integrity of the floor, since only one way flexural action was used. In the few isolated locations where composite action is necessary to a very limited extent, the projected number of reliable studs is ample to develop composite behavior. These conditions mean that the primary function of the steel beams is reduced to supporting the metal decking and wet con-crete during concrete placement, and to facilitate the eventual vertical shear transfer at points supported by steel columns, both of which technically do not require studs. The secondary function of the steel . beams is to afford weldable steel surfaces for the possible future attachment of suspended loads below the floor. slabs. It follows from the re-assessment that the welded studs are therefore nonessential. In the limited locations where composite action is required, the projected number of reliable studs is adequate to conform with the specified design criteria. The only possible exception could be at points with heavy suspended loads' attached directly to the struc-tural steel beams. These heavily loaded points will be treated individually in order to ascertain that the reliability of the existing studs is adequate or that corrective measures to improve the beam-slab anchorage are provided where needed. The Containment Structure Floor at Elev. 45'-0", Unit 2, and limited portions of the Radwaste Area and Con-trol Area floors belong to this category, and their data are summarized in Table 2 (excepting Radwaste and Control Areas which are listed in Table 1). m. 6% %N 6 4W%S44M
l... o;O oo ~ ~ Areas associated with higher than normal fdffure rates: (D) A significant number of the steel beams in these areas are required to develop composite flexural action with For some of a one-foot thick concrete floor slab. liable studs is these beams the projected number of re insufficient with respect to the minimum required by structural design, and this condition has the follow-ing implications: l. For loading cominations incorporating the Operating Basis Earthquake (OBE), the composite action from the structural system imposes, load demands on the existing studs which are above the prescribed allowable load for studs. 2. The structural integrity of the system is not necessarily endangered because the inherent mar-gins afforded by the safety factors are more than adequate and the conservatism of the design Permits sufficient load carrying capacity demon-strable by analytical calculations. In particular, dead load plus live load combinations have not been found to cause stress conditions exceeding their corresponding allowables in any of the beams in question. 3. For-loading combinations incorporating the Design Basis Earthquake (DBE), the load demands imposed on the studs are below their corresponding allowable level, since for DBE cases the allowable load is substantially higher than for the OBE Cases. 4. The design bases stated in the Safety Analysis Report are not met completely due to the potential stud deficiency. Repair work must be undertaken to correct the defective installations and assure that there are no structural systems (i.e., beams) which do not meet the design bases. The Control Area Slabs at Elev. 50'-0", 70'-0" and 85'-0, and one beam in the Penetration Area at Elev. 30'-0" belong to this category. Their data are summarized in Tables 3 and 4 and Figures 1 through 4. 1 l 'ww 4-d 'b % 4
oo OO ~ 9+
5.0 TECHNICAL EVALUATION
OF THE DEFICIENCY 5.1 General bases of Structural Design: A' common approach in the design of structural' floor systems is to develop composite action between the steel framing beams and the reinforced concrete slabs. The composite action affords a flexural systcm super-ior to the beam or slab action 'alone and generally results in an economic overall design. Composite action is achieved by aroviding horizontal shear connectors welded to the top side of the beam and embedded in the concrete. These shear connectors can also be used to improve the anchorage of steel framing into concrete slabe to permit the transfer of horizontal loads from the framing to the slab diaphragm and to incorporate the slab in resisting heavy loads suspended from the beams. In regard to the primary function in flexure, the initial design of a composite system normally provides for full composite action. This is an expedient and conservative design practice obtained with nominal cost exposure. Structurally it means that either the effective concrete or the steel section, whichever governs, is fully developed by the shear connectors (even above the actual design loads). It must be emphasized that such a conservative design. praccice is not the only approach, much l'ess an absolute requirement, and that design according to incomplete composite action is an alternative provided by the AISC Specification and consistent with the levels of integrity specified in the Safety Analysis Report. The AISC Specification and its Suppliements(1) are the governing code for design. The Specification defines the allowable horizontal shear loads for studs, and also prescribes analytical procedures for evaluating incomplete composite action by the following expression: 8eff " 8s" h (S 8 ); equation tr s n (1.11-1) of the AISC Specification Where: Yh the lesser of the horizontal shear m associated with either the concrete or the steel section c -4 .-y .wy- ,-.r-- mv-
7- .v-oo oO V'h the shear value permitted by.the = number of connectors provided,.rel-evant 'for incomplete composite' action S section modulus of the steel beam = a ,a referred to its bottom flange section modulus.of the transformed S = tr composite section (full) referred to its bottom flange S,gf effective section modulus of the = incomplete composite section The equation is based on out dated research, and it representsalinearvariationofS*bbenumberofwith respect to V'3, the total shear allowed by studs provided. In addition, the Specification pre-scribes a minimum horizontal shear that must be developed by the studs in order to rely on any level of composite action. More current research recognized by the AISC(2) indi-cates that the functional relationship described above is more accurately expressed by introducing a cubic . radical for the shear ratio of equation (1.11-1). This modification represents a refinement on the analytical technique for the evaluation of incomplete composite action, and it results in a substantially higher composite system capacity than that provided by the previous, overly conservative linear expression. The AISC Specification Committee has approved a similar expression with a square radical instead of the cubif3) radical, and it is to be officially released in 1977 This forthcoming expression offers an analysis reflect-ing the current knowledge and by incor square radical (rather than the cubic)porating the and it pru-dently introduces adequate conservatism with respect to the research findings. Additional results of the research indicate that a value of 2 to 2.25 times the prescribed allowable loads is a realistic evaluation of the ultimate load ~ capacity of the studs. This finding means that the current factor of 2.5 mentioned in the AISC Specifi-cation represents a slight overestimation of the ultimate shear capacity of studs with respect to the specified allowables. This does not render the allow-able loads inadequate, and no revision of the AISC Spec. L allowables is anticipated. However, since the DBE requirements are based on ultimate load capacity, 1^ I l ~ .._m_
to ,~ 0 (} they have been checked to verify that SAR criteria would not be violated by defining the ultiddt'e capa-city as 2.0 times the prescribed AISC allowable loads. 4 In.the current structural design, the welded studs were conservatively designed to develop complete com-posite action as discussed above and the steel beam sections were designed according to the arbitrary overall floor loads prescribed for the various areas. However, in review of the anomalies in the installation of the welded studs, the structural design was reassessed with the intention of establishing the minimum stud requirements consistent with the basic design criteria. The first step in the reassessment was to review the loading associated with each of the floor beams. This was achieved by considering actual load distributions obtained from the equipment and floor occupancies, which at this date have been established more definitely than at the time of initial design. Another aspect of the load refinement consisted of a more detailed analysis of the tributary areas for each beam by recognizing load distributions derived from the one-way and two-way flexural action of the corresponding con-crete slabs. The second step in the reassessment was to refine the evaluation of the incomplete-composite-action sections according to the latest analytical criteria,- i.e., the AISC approved expression with the square radical described above. This analytical refinement provided a revised higher capacity for sectf.ons in which the projected number of reliable studs did not permit complete composite action. This analysis was performed discriminately, and the minimum number of studs required per beam was consistently selected by the criteria described in Section 5.3. 5.2 Outline of Statistical Analysis and Evaluation: 1 This section is intended to provide a brief descrip-i i tion of the statistical approach used in the projec-tion of the reliability of studs installed. A more detailed coverage of the subject, including definition of the statistical terms used, can be found in Appen-dix A1. The initial scope of the statistical analysis was to i segregate the field test data according to homogeneous entitles judged to be statistically compatible. This l judgment was based on the similarities of the stud l l l' e m4 4 % *@uyrA @ QQQW = -3
wy-w--r
- * = - -
T
.. w 0 0 ~ 00 ~ _9 failure rates and their distribution patt, erns quantified by Chi-square tests.and associative. studies. The first 1evel of segregation estab- ~ lished was according to the various areas within the.. plant. The Turbine, Containment, Radwaste, Tankage, Control and Penetration Areas were thus r6 cognized as separate entities.with their own characteristic sampling and correspondind statistical projections. Next, within the Containment and the Control Areas, secondary levels of segregation were escablished according to certain types of beams. In most cases the analytical segrega-tion can be correlated fairly well with discernible .i physical differences of the various entities. These physical differences pertain to conditions of stud welding and application, such as welding directly to uncoated-steel (Radwaste Area and certain beams along column lines of the Control Area and the Penetration Area), welding through metal decking to either uncoated or coated steel (Radwaste Area and interior beams in i the Control Area and the Penetration Area), welding to epoxy p' rimed steel (peripheral beams in the Containment Area), welding through holes burned in decking onto epoxy primed steel (shallower, interior beams.of the ' Containment Area), welding to zinc-rich primed sur-faces '(Penetration Area and some beams in the Control Area), etc. i i One definite exception to the correlation appeared in the Control Area for the case of.the North-South and the East-West main beams. The statistical. analysis had defined these as two separate encities, but in practice no physical, installation, or other differences were found to suggest such segregation. Nevertheless, the two entities were. preserved in recognition of the fact that the North-South data, by itself (see Table 3), would yield the most unfavorable projection of reliable studs, and that in the absence of any discernible phy-sical differences, such data was conservatively con-sidered to be applicable to both North-South and East-c West beams. Another anomaly in the correlation was found in the Penetration Area, where an isolated beam exhibited an l unusually high stud failure rate of 47% (see Table 3). The other two beams tested in this area, as well as the other beams with similar stud' welding condition, do not exhibit comparable failure races. The associa-l ] l tive studies thac form a part of the statistical analysis weakly suggested that the data from the s i i e h - ~.nm .+ n - -. -. - -. - - -.~.
. o. 0 0 OO ~ _10 ,~ three beams tested in the Penetration Area.'c'o"uld be-pooled into a single entity to project a number' of reliable studs applicable to the area. Nevertheless, as a subjective recognition of the significantly dissimilar failure rates, the most unfavorable pro-jection derived from the high single beam failure rate was preserved and applied to all the similar beams in the area. This was the extent to;which the anomalous failure rate was applied; it was not introduced or combined with the data of other areas for the follow-ing two reasons: 1. The associated studies of the test data in.the other areas yielded a characteristic beam entity. These were sufficiently well defined entities, so that arbitrary introduction of the anomalous test data will render them incongruous, and thus will represent an unwarranted distortion. 2 The time period, the welding operation set-up, and potentially the. field crews for the installation of studs in the Penetration Area were different from those for other areas, therefore there is no basis for extending the application of the single, anomalous data of the Penetration Area to other areas. The statistical evaluation proceeded with the develop-ment of reliable stud projections for each of the established entities. These projections are based on the failure rates derived from field strike-test data. Their development takes into account the number of studs tested with respect to the total number installed, and recognizes that the reliability of the studs must not be on an individual basis, but with due regard to stud groupings derived from the required number of studs per beam. The analytical bases of the statistical projections are closed-form mathematical solutions involving hyperbinomial distributions, without resort-ing to empirical idealizations. The fundamental assumption is that the sampling. gathered in the field is unbiased and applicable to the balance of the - corresponding stud entity. This assumption is justi-fied because the exposed areas where the sampling was obtained came into existence randomly, and due to reasons which are unrelated to the stud welding pro-cedures. The process for locating these' exposed areas and the resultant quality of the studs within these areas were not influenced by any bias in the stud welding procedures, and to that extent there is no interdependence between the two activities. 5 ees. A N
- W-
,9y s .,.,w.....
o o 0 0' _u_ The confidence level of the statistical projection oof reliable studs was conservatively set ats90%. This is an important paramet;_ zor any pr.ojection, and it is pertinent to justify the value selected. 1 Subjectively, this level of confidence is consistent with the criteria used by governing organizations d (e.g., the AISC) involved in,the preparation of codes of' practice, but more importantly, analytically its justification is derived by the resultant low pro-babilities of exceeding the specified design criteria ~ as discussed below. For the cases of beams with potential stud deficiencies, the annual probability of exceeding the ultimate load capacity of the studs was evaluated. The evaluation takes into account the magnitude and frequency of the seismic event, the ultimate capacity of single studs set conservatively at 2 times the allowable loads accoygyngtothelatestresearchrecognizedbythe AISC , and the 90% level stated above. The resultant probabiliyies of exceeding the ultimate capacity are below 10- -per year. Similarly, the annual probability of exceeding the design criteria (i.e., the factor of safety falling below 2.01 was evaluated and demon-3 strated to be less than 10 These probability values are considered to be sufficiently low,.and justify the 90% confidence level used in the calcula-tions. 5.3 Design of Shear Connectors: The shear connectors used in all instances were headed-welded studs, installed by a semi-automatic welding procedure. In conformance with the general structural design criteria specified in the Safety Analysis Report, the studs are proportioned to be within their allowable design load for loading combinations involv-ing the Operating Basis Earthquake (OBE), and just below the ultimate load capacity for loading combina-tions involving the Design Basis Earthquake (DBE). The allowable loads enforced were those prescribed by the AISC Specification, and the "below ultimate" design load limit was set at 80% of 2 times the allowable , loads. The latter provision conservatively accounts for the recent reassessment of ultimate loads ~ per AISC references and also. reflects the concept that structural elements, whose ductile, anergy-absorbing ~ character- ~ istics are' inferior to those of the main structural components, should be subject to lower capacity reduction factors. Thus 80% (which is consistent with the SAR requirements) was introduced instead of the 90% usually associated with steel framing members. i l l 9 ~ - -. ~ - w e- ,,,,, -, ~ ,.--->,-a ,-,,,,,,,.-w ,+
~ N.. 0 0 0 0 , ye The' statistical projection of.the number of reliable studs for the various types of beams, together-with the calculated minimum number.of studs required for each beam is the-basis for verifying the adequacy of the, existing installation as a composite system and determining the number of additional shear connectors that must be attached. The minimum number of studs required per beam will be governed by any of the following criteria: 1 1 The number of studs dictated by revised structural i design calculations based on reassessed loading analysis and on the inclusion of incomplete com-posite action for the cases of flexural behavior. The effective section modulus under incomplete composite action, and the corresponding stud re-quirement will be evaluated using the latest expression (square radical) recognized by the AISC Specification Committee. 2. The quantity of studs prescribed indirectly by the AISC through the lower bound limitation of V' This provision is to assure a minimum n5k6erofstudsbelowwhichrelianceonanylevel of composite action is not recommended. In this respect, it must be noted that the AISC Specifi-cation, through its Supplements, has reduced the original lower bound of V' from 0.50 V to 0 V, and that current reseahch suggests In upwa.25 rd i r vision to a probable value of 0.40 V - I" h recognition of this fluctuation,.the lower bounds 3 applied in the current design were conservatively set at 0.50 Vs for cases in which development of composite fleMural action is the primary function of the studs, and at 0.40 V f r limited cases h where diaphragm development or anchorage for suspended loads are the main functions of the studs. 3. The number of studs needed for development of a minimum moment of inertia. This lower bound for the moment of inertia of the composite section is i; dictated by minimum natural frequency considera-tions required to maintain the structural floor subsystems within the preferred range of seismic response dictated by the in-structure response spectra. The effective moment of-inertia for incomplete composite action-used for this purpose was evaluated using the analyticel expression (cubic radical) recommended in the reference lit- ,erature. This approach is justifiable because in
O O oO~ _13 frequency calculations accurate predienion of deflected amplitude is more important than arbi-trary conservatism with respect to ultimate. load capacity. 5.4 Cogservative Features Not' Resorted to in the' Design: This is a commentary on some features that could improve the calculated margin of safety of the design, but that were not formally incorporated into calcula-tions because fundamentally they are not covered by the basic design criteria. 1. Based on engineering judgment and some test data, the allowable loads for studs could be increased in proportion to the square root of the concrete compressive strength f' In the current design allowable stud loads fof. f' = 4000 psi, accord-ing to the AISC Specificati8n, were used without ta' king credit for~ actual f' 5000 psi concrete. 2. Typically, for vertical loading combinations in-volving a seismic event, the OBE case often governs the design and the studs are not an exception. Introducing the OBE of 33%'of the DBE (according to current NRC developmental criteria), instead of the 50% used in the current design, and also permitting an allowable stress increase for the short duration OBE loading combination, are two provisions that would relieye considerably the stud requirements if they were adopted.
References:
(1) American Institute of Steel Construction (AISC) Specification for the Design, Fabrication & Erection of Structural Steel for Buildings (Adopted February 12, 1969) Supplement No. 1, effective November 1, 1970. Supplement No. 2, effective December 8, 1971. Supplement No. 3, effective June 12, 1974. (2) AISC to Bechtel letter dated December 28, 1976. l (3) Draft of proposed revision to AISC Specification Sections 1.11.2 through 1.11.5, dated December 17, 1976. ^ l ~:7 .-.-.m
n( O '/) U ") O ' .y, 6.0 CORRECTIVE ACTION TAKEN The activity performed was for the purpose _of determining if there were any deficiencies in the structural systems erected to date. Many of the installed studs were already embedded in concrete. There were, however, exposed studs in construction openings located throughout the plant, and hammer strike tests were performed and documented on these exposed studs. 6.1 Corrective Action Specific corrective actions taken are as follows: 6.1.1 Sixty-three (63) Stud Welders performing this type of stud welding were requalified. This will be a continuing program. endix III was Bechtel procedure WPP/ QCI 201 App /QCI 202 Appen-6.1.2 revised and re-identified as WPP dix I. This revised Work Procedure Plan (WPP) and Quality Control Instruction (QCI) provides detailed sequences of operation and inspection that assures that welders and Quality Control Engineers are provided with adequate procedures and instructions. The following Bechtel pro-cedures and instructions also apply to stud welding operations: WPP/QCI 200, Appendices I through V, Control of Welding Filler Material. WPP/QCI 201, Welding Control for ASME, Section III. WPP/QCI 202, Welding Control for AWS Dl.1 Welding. The noted procedures are part of an integrated WPP/QCI program which combines work plans / pro-cedures and quality control instructions into one document and are incorporated in one manual. This manual is titled, " Field Construction and Quality Control Manual." The program of com-bining WPP's and QCI's is approximately 80% com-pleted. A " walk through" proofing of the above listed WPP/QCI's has either been completed or is in process of completion. A " walk through" pro-gram for all llPP/QCI's has been established. E t-.- M > ea, e# h-,p.<.*
( N ('} g 's -15 l ,9+ 6.1.3 'The requirements and documentation.-of. set-up inspections on each beam before stud welding are incorporated in WPP/QCI 202, 6. 1.4 Maintaining close and. strict surveillance of all stud welding operations by'the Field Welding Engineer and Quality Control Engineer with appropriate documentation is a requirement detailed in WPP/QCI 202 6.1.5 The removal of paint, rust, and other foreign deleterious matter from the studs or work area on the member to which the studs are to be welded are clearly defined in the noted appli-I cable WPP/QCI's and other associated documents. 6.1.6 A series of personnel training sessions address-ing Welding Procedures WPP/QCI's 200, 201, 202, 206 and QCI's 203 and 204 was attended by a total 1 of one hundred twenty-six (126) Supervisors, Field Engineers and Quality Control Engineers. 6.1.7 To date, forty-seven (47) internal audits have j been conducted by Bechtel in areas related to i this corrective action. Two hundred forty-five (245) elements of welding and related inspec-tions were audited. Nineteen (19) findings requiring corrective action were reported. Of l the corrective actions taken, ten (10) items were verified, and nine (9) items are still open. 6.1.8 As part of an overall program to provide greater assurance of procedure adequacy and product quality compliance, Bechtel site Quality Assur-ance at the jobsite has been increased by ten (10) persons with special emphasis directed to product audits. Eight (8) Bechtel Quality Con-trol Engineers at the jobsite have also been added. Weekly meetings for welding requirements are being conducted to assure thorough under-standing of related documents and requirements. Bechtel Construction has added four (4) Field Welding Engineers at the jobsite and also holds weekly procedure and instruction familiariza-tion meetings. 4 L--
Ob b O _16 6.2 Stop Work Notice A stop work notice, Number 4, was issued on Octiober 11,,1976, and resulted in the following action:
- 6. 2 /1 NCR W-021 was issued on October 13, 1976, defining the~ re-verification program.
6.2.2 Welders were requalified'and a revised welders list was issued on October 3, 1976. and updated on January 1, 1977. 6.2.3 WPP/QCI 201, Appendix III, Stud Weldings, Embedded in Concrete, was revised on October 12, 1976, and reissued as WPP/QCI 202, Appendix I, Stud Welding, dated December 13, 1976. 6.2.4 Field Engineers and Quality Control Engineers were reinstructed on October 13, 1976, and a continuing familiarization program, regularly scheduled, has been implemented. 6.3 Qualtiy Control Responsibilities As noted in Paragraph 6.2.4, regularly scheduled meetings have been established for a continuing familiarization program. (Reference paragraph 6.1.8.) 6.4 Advance Release of Work Plan Procedures Advance release of WPP 201, Appendix III, Weld Control, was initiated October 22,1976 (per QCI 002, aaragraph 9.6), and provided for immediate utilization by Quality Control Engineers pending formal release of integrated and approved WPP/QCI 201, Appendix III. This action precluded Quality Control acceptance of construction. work not covered by Quality Control Instructions. 6.5 Repair Procedures Implementation of the previously defined criteria indicates that some restitution of studs is necessary [ 'for some beams in the Control Area and to a very l limited extent in the Penetration Area. Two alter-nate repair procedures to achieve the required i restitution have been defined. They are' illustrated in Sketches Nos. 1 and 2, attached. The first alternative is a bolted-through approach applicable to cases where the steel decking corruga-tions cre parallel to the beam or where the beam is w
OD OO W 4 slightly embeded into the concrete slab. The concept of this approach is to develop a friction.-type' con-nection between. beam and slab through the pre-tensioned, high strength bolt. The grouting of the bolt in the drilled hold and the friction connection render the detail effective by minimizing the tendency of initial slip. The.second alternative consists of developing a hori-zontal shear key within the natural cavity that exists when the steel decking is placed over and across the top of the beams. .The shear key is well anchored to the steel beam by a welded attachment. 4 Positive engagement at the key-decking interface is gained by the bonding properties of the epoxy grout, and at the decking-slab interface is developed by the concrete engagement into the currugations on the j inclined surfaces of the decking. These conditions at the interfaces, even though-they are positively 4 resolved, are not crucial for the effectiveness of the 4 i system because under the actual shear transfer they are not highly stressed. Both al'ternatives have a design capacity of approx-imately 1.3 times that of the original stud, and are 4 l , calculated according to conservative practice. Never-i theless, it is recognized that the repair measures will be introduced subsequent to the initial stud installa-l tion at a time when some level of load may already exist procedure includes the following considerations: pair in the composite system. In that regard, the re 3 i 1. Both alternatives afford minimum initial slipping. 2. The composite beams were of un-shored construc-i tion..This means that the dead load of the j structure is not being carried by-the composite j system, and to that extent the studs are not i stressed. j 3. ' currently the only loading imposed on the composite system is construction loads and miscellaneous loads of equipment already installed. This load-ing represents only about 10% of the maximum i design load for the composite system. This [ governing maximum load is mostly due to future live load and the vertical seismic-amplification i of all gravity loads. p L l Consequently, the repair. procedures are adequate because of their conservative design, sensitivity to-initial loading, and more importantly, because they will be introduced at an early stage - of the loading sequence. N _ _,, #Pde pil MM
' I. ;,' OO OO -
7.0 CONCLUSION
<9* Thestatisticalprojectionofreliablestudsf$rthevarious types of beams, together with the calculated minimum number of studs. required for each beam, is the basis for evaluating the extent and effect of the defective stud weldments within each composite beam. In most of the areas involved, the proj' cted number of e reliable studs are sufficient to perform their structural function either as it was originally designed or as dictated by'a load analysis reassessment and the introduction of incomplete composite action. At several elevations in the Control Area, and to a limited extent in one floor of the . Penetration Area, stud deficiencies exist on various beams that will have to be corrected by repair procedures per-formed on the existing installation. The structural analysis and design was performed using adequate conservatism and in accordance with the applicable l codes of practice to assure that the existing installation will conform to the criteria and bases of the Safety Analysis Report. Conformance is subject to the qualifica-tion that the analytical treatment of " incomplete composite action," as described in Sections 5.1 and 5.4 of this 1 report, is in accordance with current research recognized by the AISC Specification Committee, but which has not yet been formally incorporated in the AISC Specification as referenced in the' SAR. Official adoption is expected in 1977. i 1 i I i -4 ygm --e-w w 4 _~-,-,-g ,,g w 9m-.--
s: 00 00 'I APPENDIX Al 6' STATISTICAL ANALYSIS OF-STRIKE TEST DATA 1.0 ANALYSIS OF STRIKE TEST DATA A total of 2,995 studs were evaluated by strike tests. Of this total, 2,829 passed and 166 failed the test for an overall success rate of 94.5%. It would be appealing to treat this data as a single aggregate sample since that would yield the greatest precision of the estimate of the success rate parameter, p. However, any particular treatment of the data mtsst be justi-fied in order to offer a sound estimate of success. There is a noticeable amount of variation in estimated success rates depending on location and the condition of the welding surface, as shown in the accompanying table. Each box displays the number of studs tested, the number passing the test (in parentheses), and the corresponding success rate (as a percentage): Containment Auxiliary Turbine (El. 45'-0") Buildings Bldg. (El. 72 '-6") Direct to 563 194 Painted (519) (183) Surface 92.2% 94.3% Direct to 902 1,243 (820) (1,233) Unpainted Surface 90.9% 99.2% Through Decking to 93 (74) Painted Surface 79.6% e The primary question is whether these different success rates are affected by location and/or surface or are they simply the variations to be observed in any random process. If the different rates are shown to lie within the realm of probabilistic " noise," then all 2,995 individual tests may be pooled into an aggregate sample and 94.5% used as the success rate in all the subsequent analyses. However, if this is not shown, then the data must be regarded as separate subsamples and an allowance made for the lower pre-cision which results. The following analysis shows that the latter alterna-tive is indeed the case, and Section 2.0 on the hyperbinomial distribution describes how the final recommendations incorporate this loss in precision to assure a rigorous and conservative analysis. The key analytic.questica is that of whether or not the underlying pass /f a'il ] The probability is the same regardless of location and_ surface condition. -_,~..,,;_
OO -. 0. OO principal statistic to be used is the X test of homogeneity. 'Io'r example, the effect of different locations may be isola.ted in the fo11owing'2-way contingency table: e Containment Auxiliary Turbine Total: Pass 519 1,077 1,233 2,829 Fail 44 112 -' 1U 166 Total 563 1,189 1,243 2,995 If the studs in all three buildings had a common success rate of 94.5%, (i.e., if homogeneity is the null hypothesis) then the expected number of " passes" in the containment building would have been 532,.with 1,124 and 1,175 expected in the auxiliary and turbine buildings, respectively. Similarly, the expected number of failures would have been 31, 65, and 68. The X test statistic is based on the differences between all 6 observed and expected values: test =.(519-532)2 + (1077-1124)2 + (1233-1175) 2 X 532 1124 1175 (44-31) (112-65)2 (10-68)2 31 65 68 = 94.05* 2 ThisteststatisticisapgximatelydistributedasaX random variable with 2 degrees of freedom for which there is only a 0.5% chance of exceeding 10.6. Since the test statistic is so much greater than this value, the idea of homogeneity must be rejected, and the data from each location treated as separate samples. Similarly, the question of whether the welding surface had any effect on the success / failure ratio may be analyzed from the following contingency table:
- The exact X value is 92.99.
The apparent difference is due to rounding off the expected values to intergers for narrative purposes. The exact . values were used in reaching all data-clustering decisions. (D A.,M. Mood and F. A. Graybill, Introduction.to the Theory of Statistics. McGraw Hill (1963) p. 318 6 e-. OOM MWN** Y A%- g
OO -- S. OO Through '9" Direct to Decking Direct to Paint Unpainted
- to Paint Total,,.
702 2,053 74 2,829 Pass Fail 55 l 92 19 166 Total: 757 2,145 e 9'3 ' 2,995 The value of X test here is 50.1, so that homogeneity must again be rejected and each surface condition treated as a separate classification. Similar analyses were performed for separate auxiliary buildings (i.e., control, penetration, tankage, radwaste) and also for individual beams and girders within these buildings. The final categories of homogeneous samples were established to be: Sample Location Condition
- Size (N)~
Successes (R) 1. Containment Direct to Paint 563 519 2. Tankage Direct to Paint 194 183 3. Control Through Decking 93 74 to Paint (beams) 4. Radwaste Direct to Unpainted 166 163 5. Penetration Direct to Unpainted 90 73 6. Control Direct to Unpainted 384 329 (N/S beams) 7. Control Direct to Unpainted 262 255 (E/W beams) 8. Turbine Direct to Unpainted 1,243 1,233 i Totals: 2,995 2,829 These categories were used as inputs to the hyperbinomial distribution to establish the probabilistic characteristics of beams and girders described in Section 2.0.
- These are the general conditions for the established categories and do not necessarily cover all of the differen.t conditions encountered through the installation.
4 a we d-M Ni _' gGe w + kmeerww.rq y-w
OO i OO 2~. 0 INFERENCES FOR INDUSTRIAL BEAMS AND GIRDERS '9" 3 ~ The results of the above analysis establishes the appropriate homogeneous - groupings of test data for quality characteristics of individual studs. The major concern here, however, is not only about individual studs but also the adequacy of collections of studs on entire beams and girders. This analysis proceeds by recalling the hyperbinomia'l distribution.( } The motivation is as follows. First, if the success parameter, p, were known precisely then the total number of good studs in a collection of h would vary according to a binomial distribution: P [k of h/p] = p (1-p) For example, if p =.6 and h = 5, then the numerical values of the resulting mass function would be: No. of Good Studs = k P [k of 5; p =.61 0 .010 1 .077 2 .230 3 .346 4 .259 5 .078 1.000 However, if p is not known but must be estimated, then such a binomial distribution assumes more precision than actually exists and makes things appear better than they are. For example, if n studs have been tested and only r passed, then the parameter p itself has a probability distri-
- bution, f(p) = r! n 4)! p (1-p) -
for 0 4 p 4,1 the familiar beta distribution ( ). Thus, while the expected value of p is r/n, other values of p between 0 and 1 may also have generated the sample, and these cannot be ignored in any subsequent inferences. (2)
- 11. Raf ffa and R. Schlaf fer. Applied Statistical Decision Theory liarvard University Press (1961).
- p. 237 (3)
A. M. Mood and F. A. Graybill, Introduction to the Theory of Statistics, McGray-liill (1963) p.129 ff. t
OO OO To obtain the probability of k good studs in a beam of h when r' bf n similar studs have passed a strike test, the unconditional distribution may be found by: 'r 1 P [k of h; r of nl = P [k of h/p]
- f (p; r, n) dp Jo e1 hl k
h-k. (n+1) ! r g_ y g_p)n-r dp k!(h-k)! r!(n r)! o Collecting constants: l n hl (n+1)! p #(1-p)" dp '#~ k! (h-k)! r! (n-r)! yo performing the integration, h! (n+1)! . (k+r)! (n+h-r-k) ! k! (h-k)! r! (n-r)! (n+h+1)! and rearranging terms in cominational notation yields the hyperbinomial distribution: r r+k \\lf n+h-r-k ) l } l for k = 0,..., h and 4 n = h i To gain a sense of the effect of this distribution, suppose that 15 studs have been tested and 9 have passed. The estimated value of p is 9/15 (i.e., still.6) as before. However, repeated evaluations of the above expression yicids the following distribution: No. of Good Studs (k) P [k; 9 of 15] O .023 1 .103 2 .227 3 .303 4 .246 1 5 .098 1.000 ,M^
OO. OO -e-Note that this distribution is more dif fuse than the simple binom'lal; i.e., the tails of the distribution are " fatter" and less probability mass is concentrated around the central value. The import of this is that when inferences are made about the adequacy (or inadequacy) of studs on beams or girders, a more stringent, conservative set of standards are applied than would result from the simple (and inappropriate) binomial distribution. The analysis described elsewhere in this report involves beams with 62 or more studs and girders with as many as 132, and the ~ values of n and r are on the order of several hundred in some instances. Thus, the evaluation of all the appropriate mass and cumulative distributions is a laborious and computationally demanding task. Accordingly, a computer program was developed to assist in these studies. The two key ideas are: i) all probabilities are carried in logarithmic form until the final printout to guard against round-off error and assure the j requisite level of accuracy. j ii) each ralue of the mass function is related to the previous one, so that once p(0 of h; r of n) is found, the other values may be calculated recursively. This reduces the number of factorial evaluations and aids the computational efficiency of the total program. Execution of the computer program yields the density and the probability functions derived from a given set of field test data for a given total of studs grouped according to the number of studs per beam. Next this output is reduced to obtain the probability of exceeding the prescribed design criteria as a function of the number of reliable studs which exist or which are to be provided in a given beam. From this information, the projected number of reliable studs for a given type of beam is derived observing the stipulated 90% confidence level. The corresponding projections of reliable studs are included in Tables A1, A2 and A3. I I e r
- i.____-._-.
.e
s TABLE 1 FIELD STRIKE TEST DATA AND EVALUATION FOR AREAS WHICH EXHIBIT ACCEPTABLE STUD FAILURE, RATES No. of Type Studs Total No. Percentage of Beams of per Code of Studs No. No. Reliable Studs at Location Tested Beam Beam per Floor Tested Failed Failed 90% Conf. Level Remark's Radwaste 1 Main 34 1 415 34 1 2.9 93 Main beaw@ W side, , Area-- between 16.9 &l6.4 Elev. 24'-0" 34 0 0 93 '. Main beam @ E side, 1 Main 34 1 between 16.9 & 1 Radwaste 1 Main 34 1 290 18 0 0 93 Main beam @ W side, between 16.9 & g Area-- Elev. 37'-0" 1 Main 36 1 17 0 0 93 Main beam @ W side, between 16.9 & 18.4 2 Main 36 1 35 0 0 93 Radwaste 1 Main 34 1 220 14 1 7.1 93 Main' beam @ W side, Arma-- between 16.9 & 18.4 Elev. 50'-0" 1 Main 34 1 14 1 7.1 93 Main beam @ E Side, between 16.9 & 18.4 Trnkage ' 1 Main 97 2 970 97 6 6.2 91 Main beam @ 15.6 Arca-Elev. 63'-6" 1 Main 97 2 97 5 5.2 91 Main beam @ Nort 15.6 Turoine 1 Main 75 6 6760 75 5 6.7 98 Main beam @ E of Building between#7&8 Elev. 72'-6" ~ 1 Main 115 6 115 1 0.9 98 Main beam Q,N of'8, between'J &*H (Unit 2) 1 Main 40 6 40 2 5.0 98 Main beam @ S of'9, between J & H CSte Table 3 for Code data I
'T l I TABLE 1 (Continu';d) ( No. of Type Studs Total No. Percentage of i i Beams of per Code 'of Studs No. No. Reliable Studs at Location Tested Beam Beam per Floor Tested Failed Failed 90% Conf. Level Remarks. I, Turbine 1 Main 75 6 75 1 1.3 98 Inter. beam @ E of
- Euildinr, "C", between 7 & S Elev. 72'-6" 1
Main 50 6 50 1 2.0 98 Inter. beam @ N of '., 60, be tween "d' & "D" 27 Misc Varies 6 406 0 0 98 g 24 Misc varies 6 482 0 0 98 Misc Varies 3 or 5 175 0 0 90
- See Table 3 for Code data 5
e o g 6 O 9 1 'b
I' TABLE 2 FIELD STRIKE TEST DATA AND EVALUATION FOR AREAS IN WHICH STUDS ARE NON-ESSENTIAL STRUCTURAL ELEMENTS 3 No. of Type Studs Total No. Percentage of Eeams of per Code of Studs No. No. Reliable Studs at Location Tested Beam Beam per Floor Tested Failed Failed 90% Conf. Level Rematks* Containment 1 Main 92 3 2020 92 7 7.6 85 For the containment Building area, elev. 30.'-0", Elev. 4 5'-0" 1 Main 86 3 86 10 12.0 85 at a confidence (Unit 2) lev ( '. of 90%, it is 1 Main 92 3 92 23 25.0 85 projected that 93% of the 880 ins ed 1 lbin 98 3 98 2 2.0 85 studs are reli 4 1 25.0 90 1 Inter 4 5 16 1 6.3 90 1 Inter 16 5 CSee Table 3 for Code data O 4 s 1 'A ti
FIELD STRIKE TEST DATA FOR AREAS ASSOCIATED WITH GREATER YHAN ALLOWABLE FAILURE RATES No. of Type Studs Total No. Beams of per Code of Studs No. No. Location Tested Beam Beam per Floor Tested Failed Failed Remarks Mainbeam@"L"betweenk.1&b2.4 Control 1 Main 135 1 4576 132 34 26.0 ' ~ Area (MK3) Elev. 50'-0" Main beam @ k.'1 & '.' of"L" 40 4 10.0 1 Main 130 1 N (MK8) Main beau @ 2S1 & E. of "L" 23 0 0 1 Main 126 1 V (MK2) Inteqaediate beam @ E of L Letween 21.f 1 Inter 31 4 4706 31 6 19.0 ,'g i (MK7) & 22.4J Main beam @ "L" between b & 2.4 Control 1 Main 135 1 4294 132 16 12.0 V I Area (MK1) l Elev. 70'-0" 44 0 0 Main beam @ 21.1 & W of "L" l 1 Main 126 1 (MK10) j 43 0 0 Main beam @ b & E of "L" 1 Main 126 1 D' (MK4) l 24 0 0 Main beam @ & W of "L" 1 Main 126 1 (MK4) 13 0 0 Main beam @ 25.'1 & E of "L" 1 Main 126 1 g (MK3) W e beam @ E of "L" betweh 1 Inter 31 4 3860 31 6 19.0 rmed (MK7) V v' 1 Control 1 Main 120 1 2560 120 5 4.2 Main beam @ "L" between 2 &22]. Area (MK7) Elev. 85'-0" 29 1 3.5 Main beam @ 21.1 & w of "L" 1 Main 112 1 (MK1)
No. of Type Studs Total No. Beams of per Code of Studs No. No. Location Tested Beam Beam per Floor Tested Failed Failed Remarks r. 9 1 11.0 Main beam @ 21.1 h E of "L" s Control 1 Main 112 1 Area (MK2) v e-Elev. 85'-0" 8 0 0 Main beam @ h2.4 & E of "L" 1 Main 112 1 l (MKl) l 1 Main 112 1 29 1 3.5 Main beam @ 4 & W o f "L" ) (MKl) 's..' 1 Main 31 4 1650 31 7 23.0 I germediate beam @ E of "L" cenl N. (MK10)
- 21.l' & 22.4, w-Penetration 1
Main 114 1 1620 30 14 47.0 Main beam @ col. "N" S of,12.5 l Area ,.e-l 38 2 5.3 Main beam @ col. "S" between p2.6 & l Elev. 30'-0" 1 Main 66 1 (Unit 2) 13.6 's _. 22 1 4.5 Main beam @ col. R4, south of'12.3 1 Main 132 1
- TEST DATA CODES 5
1 - Uncoated Steel 2 - Coated with Zine Rich Primer 3 - Coated with Epoxy Primer h 4 - Through Decicing to Coated Steel 5 - Through Burned Hole in Decking onto Epoxy Primed Steel 6 - Epoxy Primer, Intermediate and Final Coats Ground off Prior to Welding A. e f
EVALUATIONOFDATAFORAREASASSOCIATEDhTHGREATERTHANALLOWABLEFAILURERATES Studs Minimum Percentage Percentage of No. of Beam per of Studs Required Reliable Studs at Number of Studs Location Beams MK Beam For Composite Action 90% Conf. Level to be Restituted Remarks ~ Control 13 1 120 65.0 80.0 None Area l F Elev. 3 2 120 65.0 80.0 None ll . [ 30'-0" J 5 3 120 65.0 80.0 None e -- 75 4 60 50.0 68.0 None i ~ h, i 2 5 60 30.0 68.0 None lLq l g (* 6 6 40 32.0 68.0 None 1 l 5 7 120 52.0 80.0 None j ljI-15 8 120 53.0 80.0 None F i b Control 8 1 126 56.0 80.0 None I b Aren l 5 Elev. 6 2 126 52.0 80.0 None l l 50'-0" I % +: 11 3 135 83.0 80.0 50 ~ 6 4 135 49.0 80.0 None 30 5 60 67.0 68.0 None 27 6 60 60.0 68.0 None 18 7 30 73.0 68.0 None Composite action not requit 1 8 130 80.0 80.0 None 3 1 9 126 52.0 80.0 None n f i s-
[ I l 'J.- Studs Minimum Percentage Percentage of ~ No. of Beam per of Studs Required Reliable Studs at Nu=ber of Studs Location Beams MK Beam for Composite Action 90% Conf. Level to be Restituted Remarks Con trol-1 10 135 58.0 80.0 None Area Elev. 2 11 20 80.0 68.0 None Composite action not requi 50'-0" g d 1 12 46 100.0 80.0 None Composite action -not requi l Il i 1 13 120 53.0 80.0 None (. Control 11 1 135 90.0 80.0 160 ll h l Area l l Elev. 6 2 135 54.0 80.0 None 1 6 3 126 52.0 80.0 None 12 4 126 68.0 80.0 None 1 26 5 60 67.0 68.0 None 53 6 30 73.0 68.0 None Composite action not requi 27 7 60 60.0 68.0 None 1 8 120 95.0 80.0 20 1 9 46 100.0 80.0 12 1 10 126 52.0 80.0 None c 1 11 126 52.0 80.0 None 1 12 135 58.0 80.0 None 6 13 20 60.0 68.0 None i
- 'l,_',.f.\\'lQ.f*G,f,'.'~*.?l~;
l',,:...',l.,-{,,=R.;4f _ ~l?.ljl.~ ~[ 7l'. lA,~ ~ ~ A. - 'Il > }...h, fl.e [. ,q> *., : - ;:
- J ; ;. a 4. -- :V,,v
'..q. - V 3,., +- - f.
- i,
. 9. , *;.y:.. e - ;, -4 a-e ~: qa ' t
- c.. *...
t. +, , y ; y?.... .. W. k,
- ;,:, y.
.x 7 .?.,,
- t.. -_f.;.,.~
y. .x-Q.,,. i, r... a._ s - - ;.. - -a y .v y
Studs }iinimum Percentage Percentage of No. of Beam per of Studs Required Reliable Studs at Number of Studs Location Beams MK Beam for Composite Action 90% Conf. I.evel to be Restituted Remarks Control 13 1 112 45.0 80.0 None [ Area Elev. 7 2 - 112 48.0 80.0 None 85'-0" 2 3 177 63.0 80.0 None 1 4 174 50.0 80.0 None 3 5 112 62.0 80.0 None 91 14 6 120 55.0 80.0 None O 5 7 120 45.0 80.0 None 2 8 94 51.0 80.0 None 2 9 120 55.0 80.0 None 94 10 30 73.0 68.0 None Composite action not requim 4 11 30 100.0 68.0 44 Composite action not requim 8 12 30 100.0 68.0 None Composite action not requir O 4 13 30 100.0 68.0 44 9 on col. line f ], Penetration 1 114 47.0 40.0 10 Area x Elev. 1 132 38.0 40.0 None On col. line R4i ~- 30'-0" t
~" ~ o. e0 0 'sc 0< ? s-4f4 >//2"$. WASHERS l ,4STM A 440 L l pymq j + n c - T a- <t . ;/.:- .,;; g.i. I 1 .,.cc q T.O 3 f i Y 3' A/OH.SNR/NK ~ H/GH STRENGTH . !.;i. :: . ~. GRQu?7* f W::
- .i ;.
. :l: ~ ..g.. u - e.e = y v t Ij t j = g m s g ) i v- ~ ? -/f]$ 7HREADED RCD IWTH /-NUl* EA. END ASTM A 525 70200ED W30 CE ? W55 FOR FR/CT/ON CONNEC7/ON: //pg"0 VERS /ZE HOLE /N PLATE WASHER 5 AND 8EAM FLANGE: Y '}. 9/g' OVERS /2E HOLE /N CONCRETE 5L48. REPA/R PROCEDURE o 5 KETCH / p,toprgg:
- l.PR/OR 70 DR/lL1NG CHECK HOLE LcCAT/CA/ AS /~0LLOWS:
- WITH RESAR DETECTOR A5 CERTA /N TH47 Top LAYER OF RENFORCEMENT AND ANY EX/ST'/NG STUD 5 ARE C.S'E OF HOLE - PREFEERED LOCA7/ON /5 AT VALLEY0F DECK /NG CCRRUGA7/CNS, LU NOT L0CATE 7M'u S/ DES OF C5:%/NJ.
- 2. WHERE BEAM FLANGE /6 NOT &xPOSED ATlow2R END OF 80D" PKoV/DE A 5EcoND PLATE IVAtNEK F/77EO 70 EEAR ON El AE) / DECR.
B. F/REPRCCF INS BCL7ED CONNEC/?CN AT UNDERS/DE OF BEAM FLANEE -~~ p J s
Oo _Oo 099 e e sN/GH STRENGTH EPOXY GRour SouDL Y PACKED AROUND 8v2 35 TUBE AND AGA/NST DECK /hG /N VO/D Bs7 WEEN DECK /NG AND TOP FLAAGE of 8EAM FOR FULL W/DM OF &EAM. PR/oK id GROU7/MG ME SURFACE OF 5a2 Cb TUBE OECX AND TOP OF DEAM FL ANGE SHALL 82 'CLE4NED AND F/RE-PROOF /NG REMOVED. 1 f ,.5. ' ? m .Y.,'-i 3 e:: y . D .9jt.,, "kbI b: c;.Q:. _ ' 4 /4 _- . i.' O 4: 4::l2:. n / //2x 04 ~~ CUT- 0u7' fO' ,jP f T ,62. % C: e a. .e::.t. .~,.: :;.- [f j u, w e ,:...2- 'n ' . m. W... -P/ECE OF708/hG q r:.h,.. S~ UNCOA W/7H cut-our 7CP C'&^ ~ I FAC"' Bs 2< $2 0 #0 W" ~ A S TM A 50!. ? ~. L y- }r 7ki EA.E/DE REMOVE fiasmo,Onua ve-AND CLEANEC3E5 // OF BEAM FLAH3=- WQ- ~ /$7E 'F/2E/TECOF AN ll4 W.. r I ExfossO WE.' vs .I I OR 708.W3 JFTER !f;; W2/ ogw24 Q[: '" MLATOk REPA /R PROCEDURE &,3 *E- .q.. SKE7~CH 'l l~,Q 0 ** i A4.- A,g. s e f l \\ I}}