ML20070T136

Forwards Responses to IE Bulletin 79-02, Pipe Support Base Plate Designs Using Concrete Expansion Anchor Bolts. Final or Status Rept Will Be Provided by 830731

ML20070T136
Person / Time
Site:	Midland
Issue date:	01/28/1983
From:	Jackie Cook CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
REF-SSINS-6820 20684, IEB-79-02, IEB-79-2, NUDOCS 8302080216
Download: ML20070T136 (125)

Text

{{#Wiki_filter:e Consuniers Power James W Ceek Vice President - Projects, Engiaresing and Construction General offices: 1945 West Pernen moed, Jockeon, MI 49201 * (617) 788 0453 January 28, 1983 Mr J G Keppler, Regional Administrator US Nuclear Regulatory Commission Region III 799 Roosevelt Road Glen Ellyn, IL 60137 MIDLAND ENERGY CENTER PROJECT DOCKET NOS 50-329 AND 50-330 IE BULLETIN 79-02 FILE: 0505.12 SERIAL: 20684

References:

CPCo letters to J G Keppler; Midland Project; Docket Nos 50-329, 50-330; IE Bulletin 79-02: 1) Serial Howe-195-79; dated July 3, 1979

2) Serial Howe-233-79; dated August 15, 1979

3) Serial Howe-84-80; dated May 7, 1980 4)

Serial 9107; dated June 9, 1980 5) Serial 10049; dated October 31, 1980 6) Serial 11505; dated February 26, 1981 7) Serial 14636; dated December 15, 1981 8) Serial 17510; dated June 1, 1982 References 1 through 8 are correspondence which address IE Bulletin 79-02. References 3 through 8 reported that further evaluations and corrective actions were required to completely address 79-02. to this letter provides the latest status of those remaining action areas. The paragraphs of enclosure 1 are numbered the same as refeences 5 through 8. provides revised answers to Bulletin 79-02 questions 1 through 5. Enclosures 3, 4, and 5 provide test reports concerning expansion anchor static tension tests and relaxation tests with strain gauge load cells, respectively. Either a final response or a status report will be provided by July 31, 1983. aw W& WRB/lr B302080216 830128 PDR ADOCK 05000329 O PDR ] 2 fjd3 OC0183-0054A-MP01 T_6 / l

i 2 Serial 20684 I I

Enclosures:

(1) Report for I & E Bulletin 79-02 (2) Response to IE Bulletin 79-02, Questions 1, 2, 3, 4, & 5 (3) Concrete Expansion Anchor Static Tension Tests for Bechtel Power Corporation, Bechtel No 7220, WJE No 80804Q, March 9, 1982 (4) Concrete Expansion An' hor Relaxation Tests for Bechtel Power c Corporation, Bechtel No 7220, WJE No 80804Q, May 28, 1982 (5) Concrete Expansion Anchor Relaxation Tests for Bechtel Power Corporation, Bechtel No 7220, WJE No 80804Q, Addendum, i June 11, 1982 i CC:- Document Control Desk, NRC Washington, DC RJCook, NRC Resident Inspector Midland Nuclear Plant d 1. 1 k OC0183-0054A-MP01

3 Serial 20684 BCC: RCBauman, P14-314B (w/ene 1&2 only) WRBird, P14-418A (w/all enc) NRC Corres File, P24-517 (w/all enc) FWBuckman, P14-113A (w/ene 1&2 only) MLCurland, Midland LHCurtis, Bechtel AA (w/all enc) LEDavis, Bechtel-Midland (w/ enc 182 only) MADietrich, Bechtel-Midland. GREagle/DNReia, TASK-AA JFFirlit, JSC-230A WJFriedrich, Bechtcl-Midland DFJudd, B&W JNLeech, P24-507 HPLeonard, Midland BWMarguglio, JSC-220A DBMiller, Midland JAMooney, P14-115A GLRichardson, Bechtel AA (w/all enc) JARutgers, Bechtel AA (w/ene 1&2 only) PSagooleim, P14-210 RAWells, P14-113A REWhitaker, Midland l t-i OC0183-0054A-MP01 Serial 2068h Pcg2 1 IO2499 REPORT FOR I & E BULLETIN 79-02

SUBJECT:

I & E BULLETIN 7",-02 " PIPE SUPPORT BASE PLATE DESIGNS USING CONCRETE EXTANSION ANCHOR BOLTS" INTERIM REPORT DATE: 1/19/83 1. Anchor Bolt Use Prohibition a. An agreement has been reached with the original pipe support design agency to allow a one-time design deviation that permits expansion anchor bolts to remain in the first two supports on either side of a pump, provided the calculated bolt load is 25% or less of the anchor bolt specification allowable. Pipe supports that do not meet these criteria will be reworked. b. The primary design agency has prepared a report to document their design methodogy. The method has been determined to be acceptable. However, there are some Bechtel comments on this report currently being addressed by the design agency. The project plans additional reviews of the application of the methodslogy by the design ager.cy. c. Discrepant pipe support designs issued by the pipe support design agency and documented on nonconformance reports have been dispositioned. All pipe support designs issued by the primary pipe support design agency are being rereviewed for proper anchor bolt usage. This review is now being conducted to a priority based on the seismic reanalysis schedule rather than the turnover schedule as previously stated. The new schedule for completion of the review is now anticipated to be November, 1983.

Pcga 2 l0t'499 2. Determination of Proper Embedment Depth a. The inspection of anchor bolts used for pipe supports is addressed in Section 6. b. It has been concluded that no further testing and inspection for l embedment depth is required for expansion anchors used on non pipe support applications. This conclusion is based on the results of the reinspections of expansion anchors used on pipe supports, and heating, ventilating, and air conditioning Seismic Category I support applications. Reinspection and testing in accordance with project specifications (issued to satisfy the requirements of IE Bulletin 79-02) for pipe support anchors installed before May 30, 1980, was completed. Of 1,631 expansion anchors reinspected, 65 did not meet the requirements for embedment depth. These results indicate, with a 95% confidence level, that over 95% of the anchors satisfies the criteria for embedment length. Additional reinspection and testing of expansion anchors used for heating, ventilating, and air conditioning Seismic Category I supports was performed. Of 4,565 expansion anchors reinspected, 72 did not meet the requirements for embedment depth. These results also indicate, with a 95% confidence level, that over 95% of the anchors satisfies the criteria for embedment depth c. The controls initiated in May,1980 for length marking and quality control inspection (MCAR 31) provide assurance that embedment depth will not pose a problem. 3. Demonstration of Achievement of Required Factor of Safety a. Midland-specific tests to determine the amount of preload remaining in the bolt indicate that an average of 37% of the original preload remains in the bolt af ter 1 year. Other tests (References A and B) have established that the amount of preload on the bolts will not affect the performance of the anchorage. If the initial installation torque on the bolt accomplishes the purpose of setting the wedge, then the ultimate capacity of the

Pega 3 l02499 bolt is not af fected by the amount of preload present in the bolt at the time of cyclic loeding. These tests (References A and B) indicate no anchor pullout failures occurred as a result of cyclic loading and that preload is not required to withstand cyclic loading. A reviseri response to IE Bulletin 79-02 has been completed. b. An additional static tension test (Reference C) to supplement the manufacturer's data was completed. The final report, l combined with the manufacturer's data, establishes that all i sizes of expancion anchors used for pipe supports under the scope of IE Bulletin 79-02 on the Midland project met the j required factor of safety for pullout. j 4. HCAR Status a. MCAR 34: l Bechtel Management Corrective Action Report (MCAR) 34 final report, concerning installed drop-in anchors, has been issued. e No further new corrective action is required. Required rework has been completed and MCAR 34 was closed on December 14, 1982. b. MCAR 31: i MCAR 31 revised final report, concerning embedment depth of l expansion anchors, was issued May 10, 1982. Corrective actions associated with MCAR 31 are complete and the MCAR was closed on - May 24, 1982. 5. Reportability Review Review of the results of the inspection and tests identified no items with a safety impact. Results of 'uture analyses will be reviewed for reportability under 10 CFR 50.55(e). l t I -. _ _ _ _ _ - _, _ ~

1 ( Pes. 4 10249S 6. Additional Expansion Anchor Inspections Inspection of 100% of the accessible pipe support expansion anchors installed before May 30, 1980, is complete. An evaluation of the adequacy of the inaccessible anchors (less than 6.7% for any parameters), based on the inspection results of the accessible anchors, is now complete. As a result of this review, it has been determiced that no rework of the inaccessible anchors will be required. Identification and completion of the rework for the inspected anchors is now approximately 90% complete. The remaining 10% will be reworked as part of the Construction Completion Program currently being implemented at the jobsite. I l 0 l ( References A. Teledyne Engineering Services Technical Report 3501-2 for Utilities /TES Owners Group Summary Report Generic Response to US NRC IE Bulletin 79-02. l B. Commonwealth Edison Company Summary Report, Static, Dynamic and Relaxation Testing of Expansion Anchors in Reponse to NRC IE Bulletin 79-02. C. Final Report on Concrete Expansion Anchor Static Tension Tests for Bechtel Power Corporation, March,1982, Wiss, Janney, Elatner and Associates, Inc. Serial 20684 Bechtel Associates Professional Corporation 102353 777 East Eisenhower Parkway Ann Arbor, Michigan wamess P.O. Box 1000. Ann Arbor, Michigan 48106 January 24, 1983 BLC-15834 Consumers Power Company 1945 West Parnall Road Jackson, Michigan 49201 Attention: Mr. R.C. Bauman Design Production Manager

Subject:

Midland Plant Units 1 and 2 Consumers Power Company Bechtel Job 7220 RESPONSE TO IE BULLETIN 19-02 File: C-305 w/a

References:

See Page 2 Three copies of the revised responses to Questions 1, 2, 3, 4, and 5 of NRC IE Bulletin 79-02 on Pipe Support Base Plate Designs Using Concrete Expansion Anchor Bolts are attached. This supersedes the original response to IE Bulletin 79-02 for Questions 1, 2, 3, 4, and 5 provided to the NRC by Consumers Power Company in References A and B. This will l also supersede the response provided by Bechtel to Consumers Power Company in Reference G. No change is required in the response to Question 6. This closes Bechtel's action item stated in Paragraph 3a of Reference F. Very truly yours, E.M. Hughes Ann Arbor Project Engineer SLS/HND/jsh(C) Attachments: 1. Revised Responses to Questions 1, 2, 3, 4, and 5 of NRC IE Bulletin 79-02 (three copies) 2. Specification 7220-C-305(Q), Table 3.2, Rev 15 3. Midland Concrete Block Wall Design Criteria with NRC's Interim Criteria for Safety-Related Masonry ( Walls l 4. Specification 7220-C-305(Q), Appendix D Piping List 1560a l L

s s Bechtel Associates Professional Corporation BLC-15834 -l02353 3*"uary 24, 1983 Page 2

References:

A) CPCo Letter Serial 7234, S. Howell to J.G. Keppler. 7/3/79 B) CPCo Letter Serial 8163, S. Howell to J.G. Keppler, 1/7/80 C) Summary Report Generic Response to U.S. NRC IE Bulletin 79-02, 8/30/79, Teledyne Engineering Services D) Commonwealth Edison Company Summary Report Static, Dynamic, and Relaxation Testing of Expansion Anchors in Response tr NRC IE Bulletin 79-02,7/20/81 E) L.H. Curtis t o R.C. Bauman, 9/22/80, Report on Testing of Concrete Expansion Anchors Installed in Block Walls F) CPCo Letter Serial 17510, J.W. Cook to J.G. Keppler, 6/1/82 (Com 072422) G) BLC-14928. E.M. Hughes to R.C. Bauman, 8/2/82 (Com 079523) cc (all w/a): W.E.-Bird ; F.W. Buckman D.B. Miller P. Sagooleim i i Written Response Requested: No l l 1560a

. to BLC-15834 Page 1 of 4 102353 PEVISED RESPONSES TO QUESTIONS 1, 2, 4, AND 5 0F NEC IE BULLETIN 79-02 Question 1 Verify that pipe support base plate flexibility was accounted for in the calculation of anchor bolt loads. In lieu of supporting analysis justifying the assumption of rigidity, the base plates should be considered flexible if the unstiffened distance between the member welded to the plate and the edge of the base plate is greater than twice the thickness of the plate. If the base plate is determined to be flexible, then recalculate the bolt loads using an appropriate analysis which will account for the effects of shear-tension interaction, minimum edge distance, and proper bolt spacing. This is to be done prior to testing of anchor bolts. These calculated bolt loads are referred to hereaf ter as the bolt design loads.

Response

In the design of pipe support base plates, plate flexibility was considered if the unstiffened distance between the member welded to the plate and the edge of the base plate is greater than twice the thickness of the plate. For base plates determined to be flexible, an analysis has been performed on the base plate and expansion anchors that accounts for the effects of shear-tension interaction, minimum edge distance, and proper bolt spacing. Question 2 Verify that the concrete expansion anchor bolts have the following minimum factor of safety between the bolt design load and the bolt ultimate capacity determined from static load tests (e.g., anchor bolt manufacturer's) which simulate the actual conditions of installation (i.e., type of concrete and its strength properties): a. Four - for wedge and sleeve type anchor bolts b. Five - For shell type anchor bolts

Response

A minimum safety factor of four between the bolt design load and the bolt ultimate cap. city was used for wedge type expansion anchors. The ultimate capacity was determined from manufacturer's data. Additional static load tests simulating the actual conditions of installation were performed to supplement the manufacturer's data. Sleeve type anchor bolts are not used on the Midland project. A safety factor of five between the bolt design load and the bolt ultimate capacity was used for the shell type anchor bolts. The ultimate capacity was determined by static load test at project site simulating the actual conditions of installation. The project specification was the basis of design for all pipe supports and '1561a

Attschm nt I to BLC-15834 102353 Pese 2 of 4 hangers. Use of shell type anchor bolts was prohibited in August 1979, and therefore design requirements were deleted from the specification. Question 3 Describe the design requirements if applicable for anchor bolts to withstand cyclic loads (e.g., seismic loads and high cycle operating loads).

Response

Expansion anchors are not used on the supports of pipes subject to high cyclic operating loads (continuous vibration from valves, pumps, and machinery of flow-induced vibrations). These lines are identified in the project specification (Attachment 5), and it will be verified by field inspection that no expansion anchors were used for these lines. Question 4 Verify from existing QC documentation that design requirements have been met for each anchor bolt in the following areas: a. Cyclic loads have been considered (e.g., anchor bolt preload is equal to or greater than bolt design load). In the case of the shell type, assure that it is not in contact with the back of the support plate prior to preload testing. b. Specified design size and type is correctly installed (e.g., proper embedment depth). If sufficient documentation does not exist, then initiate a testing program that will ensure that minimum design requirements have been met with respect to Items a and b above. A sampling technique is acceptable. One acceptable technique is to randomly select and test one an hor bolt in each baseplate (some supports may have more than one base plate). The test should provide verification of Items a and b above. If the test fails, all other bolts on that base plate should be similarly tested. In any event, the test program should ensure that each Seismic Category I system will perform its intended function.

Response

a. Preload is lost over the life of the plant because of creep and other similar phenomena. However, it is not necessary that the bolt preload be equal to or greater than the bolt design load (References C and D). Pipe supports and anchors are subjected to static and dynamic (cyclic) loads. The cyclic loads are seismic loads and pipe transient loads. Expansion anchors tested in concrete exhibited insignificant anchor displacement when subjected to seismic and pipe transient loadings (Reference D). Therefore, if the initial installation torque on the bolt accomplishes the p rpose of setting the wedge, then the ultimate capacity of the bolt it not 1561a

___ to BLC-15834 1023_a3 Page 3 of 4 1 affected by the amount of preload present in the bolt at the time of f cyclic loading, All concrete expansion anchors are designed and installed according b. to Specification 7220-C-305(Q). Verification and testing procedures along with acceptance criteria are given in Specification 7220-C-101(Q) for anchors installed before May 30, 1980, and are given in Specification 7220-C-305(Q) for anchors installed af ter May 30,1980. These procedures require all expansion anchors to be verified for location, number of anchor bolts, spacing, and edge distance as shown on design drawings, embedment length, and projection of anchors, washers, age of concrete, anchor bolt diameter, type of anchor, anchor bolt length, and thread engagement. All expansion anchors are tested to ensure proper installation as specified in Specification 7220-C-305(Q) The proper documentation, indicating the location of or 7220-C-101(Q). the anchors, test results, type of failure when applicable, and date of along with name and signature of the inspector, are maintained at

test, the jobsite.

Question 5 Determine the extent that expansion anchor bolts were used in concrete block (masonry) walls to attach piping supports in Seismic Category I systems (or safety-related systems as defined by Revision 1 of IE If expansion anchor bolts were used in concrete block Bulletin 79-02). walls: Provide a list of the systems involved, with the number of support 9, a. type of anchor bolt, line' size, and whether these supports are accessible during normal plant operation. Describe in detail any design consideration used to account for this b. type of installation. Provide a detailed evaluation of the capability of the supports, c. including the anchor bolts, and blockwall to meet the design loads. The evaluation must describe how the allowable loads on anchor bolts in concrete blockwalls were determined and also what analytical method was used to determine the integrity of the blockwalls under Also describe the acceptance criteria, including the imposed loads. Review the the numerical values, used to perform this evaluation. deficiencies identified in the Information Notice on the pipe supports and walls at Trojan to determine if a similar situation exists at your facility with regard to supports using anchor bolts in concrete blockwalls, Describe the results of testing of anchor bolts in concrete d. blockwalls and your plans and schedule for any further action. l 1 1561a

Attacharnt I to BLC-15834 102353 erga 4 of 4

Response

Expansion anchors have been used to support field run small Seismic Category 1 piping on blockwalls, a. A list of systems, number of supports, type of anchor bolts, line size, and accessibility is not available because the routing of these pipes is not completely established at this time. Less than 100 attachments will be made to blockwalls, primarily in the auxiliary building between elevations 610' and 640'. For pipe supports not yet installed, the blockwall permit provides the necessary documentation and approval required before installing expansion anchors in blockwalls, b. Block wall attachments will be used only if there is no other feasible alternative for support. The design allowables and limitations for the use of expansion anchors in masonry walls are provided in Specification 7220-C-305(Q), Rev 14 (Attachment 2). Allowable loads on anchor bolts in concrete blockwalls were determined from the test report of concrete expansion anchors in blockwalls (Reference E), The concrete blockwalls (masonry) are designed as provided in ACI c. 531-79 with the additions described in Attachment 3. contains a comparison of Midland concrete blockwall design criteria with the NRC's interim criteria for safety-related masonry walls contained in Appendix A to the Standard Review Plan (NUREG-0800), Section 3.8.4. Allowable design loading for support of attachment to the masonry walls (block) are indicated in Drawings 7220-C-2051 through 7220-C-2097,'as applicable. d. A testing program was conducted at the jobsite and verified the acceptability of the expansion anchor design loads used in blockwalls (see Reference E). I l l , 1561a 2 ,,n,,

__ __ to BLC-15834 Page 1 of 5 Spacificatien 7220-C-305(Q), rov 15 l 102,_3 Appendix C oa TABLE 3.2 V ALLOWABLE DESIGN LOADS FOR WEDGE AND STUD TYPE ANCHORS IN MASONRY BLOCK WALLS } Minimum I Embedment Anchor i Before Tension Shear Diameter l Torquing * (T) (S) Spacing! (in.) (in.) kipsO) kips (1) (in.) I I 3/8 1-5/8 0.4 0.5 4 { i 3/8 I 3-1/8 0.6 0.8 4 1/2 2-1/4 0.7 1.4 5 1/2 3-3/4 0.8 1.6 5 3 5/6 4 1.2 2.2 6 5/8 5-1/2 1.6 2.4 6

• See Table 4.3 for embedment after torquing A

z!L (1) Values based on " Report on the Testing of Concrete Expansion Anchors in Masonry Blockwalls" at Midland, Michigan NOTES: A. When only one expansion anchor is used for attachment, allowable design load shall be two-thirds of allowable design load indicated in Table 3.2 above. B. Use 50% of allowable design load values uhen designing for seismic loads for pipe supports. C. When an anchor is subjected to both tension and shear loads, the loads shall be calculated on a straight line interaction diagram as follows: S T applied + applied S T l.0 allowable allowable C-10

Attachmnnt 2 to BLC 15834 Page 2 of 5 Spscification 7220-C-305(Q), Rnv 15 ,. j5 Appendix C IO2353 V D. Reduction in the minimum spacing and minimum edge distance is not permissible. E. Table 3.2 with notes is applicable for Q walls only. F. Allowable design loads shall not be increased for ,)y accident or severe environmental conditions. I i i 1 C-11 l 1

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - to BLC-15834 Pago 3 of 5 102353 Sptcification 7220-C-305(Q), RLv 15 (25 Appendix C TABLE 4.3 U WEDGE AND STUD TYPE ANCHORS IN MASONRY BLOCKWALLS Torque at In-stal-lation (ft-lb) Minimum Minimum Threads Test Torque Embedment Embedment Minimum Minimum Anchor Not after (in.) (in.) c/c Edge Diameter Lubri-Installation Before After Spacing Distance (in.) catedH3 (ft-lb) Torquing Torquing (in.) (in.) 3/8 10-15 6 1-5/8 1-3/8 4 2 3/8 20-25 10 3-1/8 2-7/8 4 2 1/2 25-35 13 2-1/4 2 5 3 1/2 25-35 15 3-3/4 3-1/2 5 3 5/8 55-65 38 4 3-3/4 6 3-1/2 5/8 65-80 50 5-1/2 5-1/4 6 3-1/2 '3 These values give a torque range and minimum embedment before torquing which were determined by testing expansion anchors in masonry blockwalls. (Hilti anchors were tested.) NOTES: A. Prior project engineering approval is required before attaching "Q" and "non-Q" applications to Class I masonry blockwalls. B. All reinforcing steel that may interfere with expansion anchors must be located prior to drilling. Reinforcing steel in blockwalls shall not be cut. C. Expansion anchors are allowed only in Zone A (hatched) as shown in Figure C-2, provided that cells are fully grouted. D. Reduction in the minimum spacing and minimum edge distance is not permissible. C-16

- - - to BLC-15834 Page 4 of 5 fs lOL'dOd' Spscification 7220-C-305(Q), RLv 15 /7 5* ' '~ ~ ~ Appendix C E. A maximum of two anchors is permitted per block. F. When an installed expansion anchor has been loosened and then retightened, the torque at installation shall apply for retightening the anchor. Embedment length after torquing shall then be verified. G. Expansion anchors are not permitted in the following locations: a) Block at the end of a masonry wall or 16 inches from the end of a masonry wall with all provisions of Note 4 b) Block in the top course of a masonry wcll (The course under the soffit of a steel beam should be treated as a top course.) H. A door opening should be treated as a discontinuous end and, hence, as the end of a wall. I. The corner of a blockwall should be treated as a discontinuous end and, hence, as the end of a wall. J. In the case of a small penetration up to 6 inches in diameter or a 6" x 6" square opening, expansion anchors shall be placed 6 inches away from the edge of the opening. K. A large opening should be treated as the end of a wall. i l l l l l C-17 l to BLC 15834 Page 5 of 5 Specificatien 7220-C-305(Q) Apptndix C, Rav 15 b .I f) ') J J d 9-* iUL Figure C-2 V 7.... ,.;'y. y 9..o 9: ? ?,. _c_.,' p, : .6 _..g , p. g '.. .c" - ;. 9: .),- ' t- ,o . r -r,..- l n.- .p p..... r <, y y i I 'Embedment Lenath (E) I 1 16" c % Mortar Joint 5-1/2" /l 5-1.12" i l ..f g Block (Typ)Q j i l JL.... L.i..;.:..o-,..~ s '.' _l : ~~ .~.t ., ~, n - I l l f/,/,' ' ' ' '/ 'l,,' l l / ) k ,7 .g-l .l - I .,/ t/ / // / s l ~ l / v ~~ -y i .p. ,/ .,/ I ,7(. \\ 1.1.t l -Zone I / A /. I t n ,/,/p/. l ,/, >. o l ,- L //;l h { o s k l t l y l .f .,_. a, '_ T _.., 4.,.e s. _ ? -- -.~...~-,.s_;~_..~ w w

• ,y J L

^ _

7. +

l Mortar Joint Q Block " -Joint l Partial Masonry Wall Elevation Showing Anchor Locations C-19

Attachsint 3 to BLC-15834 Pag 9 1 of 4 \\ COMPARISON OF BLOCKWALL CRITERIA 10i353 n { NUREG-0800 Standard Review Plan Description Midland Design Criteria (SRP) Section 3.8.4 Remarks 1. IAAD COMBINATIONS Normal Conditions Service Load D+L D+L D+L+W D+L+W D+L+E D+L+E l D+L+T +R See Note 1 D+L+T +R +E See Note 1 D + L + T* + R +W See Note 1 Severe and Extreme Environmental, and Abnormal Conditions D + L + E' D+L+T +R + E' See Note 1 D + L + W' D + L + T* + R +W See Note 1 D+L+R D + L + T" + R" + See Notes 2 and 3 1.5P, D+L+R+E D+L+T +R + See Note 2 I# 1.25P 1.0 a difference in (Y +y + Y,) load factors; + 1.2SE Notes 4 and 5 D + L + R + E' D+L+T +1 See Notes 2, 4 and 5 + 1.0P,+ 1.0 (Y +Y + Y,) + 1.0E' 2. STRESS INCREASES Normal Conditions No increase of stresses For wind or opera-ting basis earth-quake loads, no in-crease is allowed (Section 3a). Inspection required Inspection required i e

WG Lia]EB U Ge LLa gg@$6 Paga 2 of 4 NUREG-0800 Ot353 St.ndard Revie, plan o Description Midland Design Criteria (SRP) Section 3.8.4 Remarks \\. STRESS INCREASES (continued) 4 l Direct tension perpen-Tension perpendicu-dicular to bed joints lar to bed joints racisted by reinforce-requires testing for ment; no unreinforced unreinforced masonry. masonry is used on the For reinforced proj ect. masonry walls, all the tensile stresses shall be resisted by reinforcement I Severe and Extreme Environmental, and Abnormal Conditiocs Axial or flex-2.5 2.5 ural compres-sion Bearing 2.5 2.5 Shear and bond 1.33 1.3 Masonry tension 1.5 1.5 g s parallel to bed joints Shear and ten-1.33 None specified sion at block wythe to concrete or grout core interface Masonry tension 0 Reinforced masonry 0 perpendicular to Unreinforced mason-bed joints ry 1.3 Reinforcement 2.04,0.9Fy 2.0 1 0.9 Fy stress excepc shear / e

At?.cchment 3 to BLC-15834 Pags 3 of 4 I02353 NUREG-0800 Standard Review Plan Description Midland Design Criteria (SRP) Section 3.8.4 Remarks 2. STRESS INCREASES (continued) Shear - 1.5 1.5 Reinforcement Bolts 1.5 1.5 (No increase is allowed for expansion or grouted anchors.) 3. ALLOWABLE STRESSES As provided in ACI As provided in ACI 531-79 with the 531-79 following additions Dircet Tension and Shear at: 'o) hortar collar 0 Not specified joints b) Block wythe 12 psi Not specified See Note 6 concrete or grout core interface Dirset and flex-2.5 /f c Not specified See Note 7 I urcl tension for call and core ccacrete acting cicne 4. DESIGN AND ANALYSIS CONSIDERATIONS Interstory drift Consideration required Consideration required effects Unreinforced None used Not allowed in new construction masonry walls Masonry shear None used Minimum reinforcement walls per ACI-531 Special protec-Considered Required to meet See Note 8 tion for masonry SRP 3.5.3 walls from acci-dent pipe reaction, jet impingement, and missile impact y ,,,_-_-.w-

' ** "'"-15334 l02353 $ "4 #'4 NOTES: 1. Thermal effects T are not considered in the analysis of masonry walls. Thermal gradient Effects do not have a significant effect on reinforced masonry because sustained temperatures are limited to approximately 150 to 200F. The predicted thermal compressive stresses are conservative because assonry creeps under sustained loads and thus reduces the effec-tive modulus and stress. Thermal loads are therefore not considered in the analysis and design of masonry walls. Piping reactions (R ), in-cluding thermal effects, were added into the design live loads,(L). 2. Thermal effects T are short-term transient loads and are not considered in the analysis of masonry walls (see Note 1). Piping reactions (R ), including thermal effects, were added into the design live load (L),. P,is included in R as a pipe break effect. 3. For concrete structures (strength design), the SRP criteria include 1.25 P and 1.5 P under certain load combinations, but the Midland FSAR ($ections3.$.4.3and3.8.6.3.3)doesnotir.cludetheseincreasedload factors. The same condition exists with the masonry walls (i.e., the masonry walls are designed with a 1.0 load factor applied to P,). 4. The 1.25 E load requirement in the SRP load combination for OBE is enve-loped by the Midland criteria load combination for SSE because E' = 2.0.E. 5. Masonry walls shall not be used to resist jet forces or pipe whip re-( straint supports; therefore, Y, Y, and Y,are not considered. 6. Allowable stress for block wythe to conrete or block wythe to grout core interface: The Midland criteria allow 12 psi; Uniform Building Code, 1979 (Page 180, Table 24-B) allows 12 psi for a hollow unit net mortar area (unreinforced) and also allows 25 psi for grouted masonry (unrein-forced). 7. Stress due to direct flexural tension for cell and core concrete acting I alone is considered to be one-third of the value of the modulus of rup-ture for concrete, which ensures a safety factor of 3. Modulus of rup-l ture for normal weight concrete is F = 7.5 Jf'c (ACI-318-71). 8. Barriers and/or pipe whip restraints are provided to protect blockwalls from jet impingement, missile, and pipe whip loads when the failure of the blockwalls could prevent other safety-related structures, systems, or components from performing their intended safety functions. l ('

- to BLC-15834 Page 1 of 2 _.iiN 102353 Specification 7220-C-305(o), aev. 13 s Appendix D The following piping shall not be supported by expansion anchors: 1. Pressurizer Relief 4"-2 ECD-5 6"-2 ECD-6 6"-2 ECD-7 14"-2 ECD-5 4"-l'dCD-5 6"-1 ECD-6 6"-1 ECD-7 14"-1 ECD-5 2. Pressurizer Spray 2-1/2"-2CCA-15 ~ 1/2"-2CCA-15 4"-2CCA-15 2-1/2"-1CCA-15 1/2"-1CCA-15 4"-1CCA-15 3. Makeup and Purification 1-1/2"-2HCC-35 1-1/2" 2CCC-3 2-1/2"-2CCB-12 2-1/2"-2CCB-2 2-1/2"-2CCB-8 1-1/2"-1HCC-35 1-1/2"-1CCB-3 2-1/2"-1CCB-12 2-1/2"-1CCB-2 2-1/2"-1CCB-8 4. Service Water The first two supports on inlet and outlet of auxiliary building chillers on following lines: 6"-1HBC-304 6"-1HBC-305 I 6"-2HBC-304 6"-2HBC-305 5"-1HBC-307 6"-1HBC-308 6"-2HBC-307 6"-2HBC-308 D-1

_ to BLC-15834 Page 2 of 2 A Sptcificaticn 7220-C-305(Q), Rsv. 13 Appendix D 5. Chilled Water l02353 The first two supports on inlet and outlet of chillers on following lines: 6"-lHBC-233 6"-2HBC-233 4"-1HBC-233 4"-2HBC-233 6"-lHBC-216 6"-2HBC-216 6"-1HBC-235 6"-2HBC-235 4"-1HBC-235 4"-2HBC-235 6"-1HBC-218 6"-2HBC-218 6. The first two supports on suction and discharge of pumps 7. Reactor building spray system between the spray noz les e and the steam lines to the auxiliary feed pump turbine D-2

e Serial 2068h t CONCRETE EXPANSION ANCHOR STATIC TENSION TESTS W FOR i BECHTEL POWER CORPORATION J BECHTEL NO. 7220 a WJE NO. 80804Q e Y' March 9, 1982 E s A o e a n d A En 1 sure 3 contains the Test Report h s and Appendix J. s Appendices A through I and E to this i Test Report are available at the Consumers a Power Offices in Jackson, Michigan. e 3-i (NOTE: Page renumberinc at the bottom I of this report was done by CPCo D microfilming dept. The first three pages of the report were merely identification pages used for microfilming purposes.) 1 v WISS, JANNEY, ELSTNER AND ASSOCIATES, INC. 330 Pfingsten Road Northbrook, Illinois 60062 (312) 272-7400 1 141 '" " " ~ '""~ 5 pa

( i CONCRETE EXPANSION ANCHOR STATIC TENSION TESTS FOR BECHTEL POWER CORPORATION W BECHTEL NO. 7220 i S 5. WJE NO. 80804Q J a n March 9, 1982 n e Y. E INTRODUCTION I s O { Wiss, Janney, Elstner and Associates, Inc. (WJE) was retained by C Er Bechtel Power Corporation to perform a series of static tests on concrete O a n expansion anchors. The anchors are intended for use at the Consumers O d A Power Company Midland Nuclear Power Station Units 1 and 2. The test S series was performed in order to quantify the structural behavior char-c { acteristics of the anchors. The tests described in this report were t Y performed at the WJE laboratory in Northbrook, Illinois during the per-e S. ~ l iod from March 19, 1981 through September 11, 1981. n ~ c. REFERENCE DOCUMENTS The tests on the concrete expansion anchors were performed in accordance with the following specifications and procedures: Consumers Power Courpany Specification 7220-C-116(Q), Revi-sion 1, " Technical Specification for Testing Stud-Type Con-crete Expansion Anchors for the Consumers Power Company Mid-land Plant Units 1 and 2, Midland, Michigan" Consumers Power Company Specification 7220-C-305(Q), Revi-sion 13, " Technical Specification for Design, Furnishing, Installation and Testing of Expansion-Type Concrete Anchors for the Consumers Power Company Midland Plant, Units 1 and 2, Midland, Michigan" 7t 72&o-c //5- //- S 1141 es

r Consumers Power Company Specification 7220-G-23, Revision 8, i " General Requirements for Supplier Quality Assurance Programs for the Midland Plant Units 1 and 2 for Consumers Power Company" Wiss, Janney, Elstner and Associates, Inc. Procedures Manual OP-29, Revision 1, " Procedures Manual for Static, Torque-Ten-W sion, and Relaxation Testing of Expansion Anchors Embedded in i Concrete for Bechtel Power Corporation" 5 s. I The Consumers Power Company Specifications 7220-C-ll6(Q)-Revi-a n n sion 1, 7220-C-305(Q)-Revision 13, and 7220-G-23-Revision 8 are e Ye 4 enclosed in Appendices A, B and C, respectively. These specifications O s are hereafter ' referred to as Specifications C-ll6, C-305 and G-23, t C D respectively, or simply as the Specifications. I m~ The WJE Procedures Manual 0P-29 is enclosed in Appendix D. This a n ~ d operating procedure provides an outline of the steps to be taken in f accumulating the physical test data for the test program. s 0 t i SCOPE T f e s. The expansion anchors which were tested are generically identified I ~ n as drilled-in expansion anchors, All anchors used in this test program c. were manufactured by Hilti, Inc. of Tulsa, Oklahoma, except the 7/8 in. diameter anchors which were manufactured by ITT-Phillips Drill Division of Michigan City, Indiana. Expansion anchors are sold under a variety of trade names, each being associated with a different manufacturer. The anchors achieve load carrying capacity through a wedge mechanism located at the bottom of the anchor. The wedge is expanded against the side wall of the embedment hole by tightening a nut which tends to pull the anchor from the hole. When the wedge jams against the side of the hole, further tightening of the nut establishes a tensile load on the anchor bolt. h M- ~742."2 0-C //G//- 3 @b

The Hilti wedge anchor system uses different steels in the bolt shank and in the wedge. The bolt shank is made of AISI 11L41 or 1144 steel meeting the chemical requirements of ASTM A108. The two inde-pendent expansion wedges near the bottom of the anchor are made of AISI W i 1050 spring steel. The nuts supplied with each anchor meet the require-5 ments of ASTM A307 Grade A. The bolt shank material has an approximate J minimum yield and tensile strength of 95,000 and 105,000 psi, respec-n e, giy,1, 7 'N E g The wedge-anchor system manufactured by ITT-Phillips is a two part C s t anchor system. The bolt shank is an AISI 1213 steel havin6 an approxi-O D mate yield and tensile strength of 63,000 and 90,000 psi, respectively. j The one piece wrap-around expander clip (wedge) is made of Type 302 7 d stainless steel. S s All testing was performed in accordance with the Consumers Power O C i Company Specification C-ll6. This specification was provided by Bechtel a Power Corporation. e } All anchors were installed in accordance with the anchor manufac-n C turer's recommended procedures and Specification C-305. All testing conditions were in accordance with the Specifications. The testing program. included: (1) establishing torque-tension relationships; (2) static load tests of single anchors to failure; and (3) measuring relaxation of anchor preload with time. A report icsued by WJE on July 2, 1981 discusses the results of the relaxation test. This report discusses the results of the static load tests. A total of 156 samples were tested in the program. Fig. I shows the overall grouping and sequence of tests for each group. Table 1 gives the specific groups for each test configuration, which included different 7DD o-C.n s-//-3 1141 A3 1

s i ) . cc b C3 C3 Test Program Start 156 Samples 7, 73 I "T ~~ Relaxation Test Torque Tension Test 36 Samples 72 Samples g Specified Torque Per C-115 Specified Torque Per C-116 i 4000 psi concrete 4000 psi and 5000 psi concrete j --L s 4hs L %d i $3 Static Pullout Test $3 Static Pullout Test 48 Sem,nles tg 24 Sampics Minimum and Maximum e Average Torque Per C-305 Torque Per C-305 l f4 4000 psi and 5000 psi concrete 4000 psi and 5000 psi concrete li ik i 04 {gs Fig.1 - Overall grouping and sequences of test A3 i

TABLE 1 - SPECXFIC GROUPS OF THE TEST SAMPLES 4 No. of Samples Manufacturer Anchor .Embedment Concrete Torque-l Static Pullout Test diameter depth category Tension -305 applied torque Relaxation (in.) (in.) (psi) Avg.* Min. Max. Hilti 1/4 1-1/8 4000 3 3 5000 3 Hilti 3/8 1-5/8 4000 3 3 5000 3 Hilti 1/2 2-1/4 4000 3 3 5000 3 3 Hilti 5/8 2-3/4 4000 3 3 3 3 3 o 5000 3 4 4000 3 3 g 5000 3 Hilti 3/4 3-1/4 4000 3 3 3 3 3 ~ 5000 3 3 3 3 4 4000 3 3 I 5000 3 1 5 4000 3 3 5000 3 Phillips 7/8 4 4000 3 3 3 3 3 5000 3 3 3 3 5-1/2 4000 3 3 3 3 3 5000 3 3 3 3 Hilti 1 4-1/2 4000 3 3 3 3 3 5000 3 6 4000 3 3 5000 3 Total 72 72 36

• Samples had been subjected to torque tension test with torques specified in C-ll6.

These torques were greater than the maximum installation torques specified in C-305. ( 1145 743420-e /f 5-N-3 P39 y -yn ,-y .m 1

manufacturers, anchor dia.se ters, embedment depths and concrete cate-gories.. Three expansion anchors were tested in each test configuration. The scope of work originally required static pullout tests on 24 anchors. These 24 static pullout test sampi 4 first were subjected to W i the torque-tension test, as shown in Fig. 1. The torque tension tests S s. involved installation torques higher than the maximums allowed in Speci-1 g fication C-305. An evaluacion of the static pullout test data derived n { from the first anchors tested indicated that the strength of the anchors E may have been si nificantly reduced as a result of the high torques 8 O 1 s t imposed during the torque-tension tests. Additional static pullout n e o tests were therefore performed on 48 expansion anchors. Half of these O j anchors (24) were torqued to the minimum installation torques allowed by d Specification.C-305. The other half were torqued to the maximum A s s installation torques. 0 ( t 1 a IDENTIFICATION SCHEME FOR INDIVIDUAL TESTS t e s, All test y samples were coded by a unique test mark which identified, n t-all the characteristics indicative of the test. The test mark was composed of an eight part alphanumeric name. The name parts were separ-ated by dashes and identify: (a) type of experiment, (b) specimen name, (c) expansion anchor type and bolt diameter, (d) depth of anchor embed-ment, (e) installation torque, (f) test torque, (g) test number in the test series, and (h) concrete strength and client identification. The WJE manual OP-29, which is enclosed in Appendix D, gives addi-tional details regarding the identification scheme. An example of the identification code is given below. ,..x _g_ '? M o-C //S-//. 3 1141 ps -_..,_,-.___.,_,,,y_ ,.c __,,.,,_.,,-.-y--_%,..,c_ y

f Example: S -C1/9-H/W(5/8)-2 3/4-250-145-1-4000(BPC). 1 2 3 4 5 6 7 8 9 1. Static pullout test 2. Concrete Specimen No. 9 W { - C1 designates nominal 4000 psi concrete S. - D1 designates nominal 5000 psi concrete y a n 3. Hilti: Wedge anchor with 5/8 in. bolt diameter n e y, 4. An embedment depth of 2-3/4 in. before torquing E g 5. An installation torque of 250 ft-lbs S t 6. A maximum testing torque of 145 ft-lbs O !r 7. Test No. 1 of the test series _~ i a n 8. Nominal concrete strength of 4000 psi. d A 9. Client is Bechtel Power Corporation (BPC). ~ S g a c TEST SPECIMEN MATERIALS ~ i a t e Concrete test specimens and expansion anchors were supplied by S. I Bechtel Power Corporation and delivered to the WJE Lab in Northbrook, n Illinois. Certificates of Conformance with the material requirements of the Specifications were furnished by Bechtel to WJE. Concrete slabs were stored inside the laboratory under ambient laboratory conditions. Expansion anchors were segregated and stored in the laboratory to pre-clude unintentional use of the anchors or substitution of other anchors. Structural plates and shapes used as parts of testing assemblies i and whose properties are cited in the Specifications meet the require-ments of ASTM A 36, Specification for Structural Steel. The material certificates for the steel plates are found in Appendix E. I 1141 ^ j 2.;L C-C HS-N-3 p3) 1 l

E The quantity, size, strength category and designated use of the concrete slabs supplied by Bechtel Corp. are given below. Strength Quantity Size Category Designated Use W i 2 4 ft x 11 ft x 1-1/2 ft 4000 psi Torque tension and pullout tests s 2 4 ft x 11 ft x 1-1/2 ft 5000 psi Torque tension and pullout tests Jj 8 3 ft x 3 ft x 1 ft 4000 psi Relaxation tests n e Y. A maximum of 23 anchors were installed in each 11 ft slab. A E maximum of six anchors were installed in each 3 ft slab. s t B Compressive strength data, obtained from 6 in. x 12 in. cylinders, O r were also received from Bechtel Corp. An evaluation of the data indi-n d cated that both concrete categories have a 90 day compressive strength f of approxima. ly 7000 psi. The test data and compressive strength ver-5 ( j sus age graphs are enclosed in Appendix F. 1 a SPECIMEN TRANSPORTATION AND IDCATIONS OF EXPERIMENT 3 e 5. I n Criteria for Transportation c. Expansion anchors were to be installed in concrete slabs containing no significant defects such as cracks, honeycombs or internal voids. Specimen stresses during transport were to be checked to ensure that maximum stresses were below the point at which cracking could occur. The slabs were received in good condition with no cracking observed. No inspection was performed by WJE to determine if the slabs were free from internal defects, but the surface appearance of the slabs indicated the concrete was dense and well consolidated. ( i M- '141

• # 4"5 '"' "

Psi?

Lifting Inserts for Specimen Handling k Lif ting inserts were cast into the specimen to facilitate trans-portation and handling. The lif ting inserts were positioned in each test \\y specimen so that they caused minimum restraint on the testing of anchors 1[ embedded into the test specimen face. Jj Location of Anchors n e Test anchors were located to satisfy the spacing requirements of E f ASTM E 488 and the Specifications. Minimum center-to-center and edge M t n spacings listed in the Specifications were followed in locating the C3 r I 'chors. O U ~ EXPERIMENT PREPARATION A ~ s 5 ( j All anchors were installed using the manufacturer's recomunended i a installation procedure. These procedures are included in Appendix G. t 7 2 5. g Expansion Anchor Installation n C. Drilling of embedment holes - Test expansion anchor embedment hole locations were marked on each test specimen prior to drilling each hole. Holes were spaced to meet. the requirements of ASTM E 488 and the Speci-fications. The embedment holes were located and drilled to minimize the effects of restraint of the lifting inserts. ) Holes for the test expansion anchors werc drilled af ter the con-crete had cured for a minimum of 45 days. The diametcr of the expansion hole was in accordance with the manufacturer's recommended procedure. 1 l Each hole was drilled with a carbide tipped bit, meeting the requirements 1141 , a n o-ws-//-3 b,p ---e -,-r-.m--- y m-,

of ANSI B94.1.!' (1968). The recommended diameter drill bit was installed in a roto hatsner drill corresponding to the reconsended drill for the size and diameter of the expansion anchor to be installed. Each time the drill bit was used, its use was recorded. When the bit had been used 100 { times, the bit was discarded. The depth of the anchor hole was deeper 5. than'the specified embedment depth. y a n l Anchor installation - Before each a:2chor was placed into the embed-Y. T E ment h le, 1 se dust resulting fr a the drilling operation was removed 1 s by air pressure. Each anchor was inserted into the cleaned hole and i $ driven into the hole to the specified embedment depth. O I O a The anchors require rotation of a nut (turned element) on the anchor n ~ d stud to expand the wedging mechanism against the concrete. A torque in ~ As accordance with the Specifications was applied. The application of this S 0 (, e torque was termed the installatica torque. i 3 t e Measured Installation Characteristics 5. I j, During anchor installation, measurements relating to the conditions of the installation were recordad on data sheets. The data sheets are found in Appendix H. Basic information regarding installation included verification that the manufacturer's equipment and recommended installation procedure were followed. If the manufacturer's recommended installation procedure varied from that in the Specifications, the manufacturer's procedure was waived and the Specifications' procedures followed. All anchors were installed with an out-of plumbness of less than three degrees. A photograph of the instrument used to measure the out-( of plumbness angularity is shown in Fig. 2. y g yg) [//f* - //* S 1141 ga

3..,!'+ ... j

.y,

L g. ; ^, j[ T ~ ' ~'~ %'.J'.* . ;g J.. ' ; r ,: s . '+ q m.: 5,g p l .... 9..,. m., ' %.,.x..s* t y 5. g,., '., ' i g .- q g y .. y.. . h. .a .,.s + ,,u te+ + ~ .~<..,n s... r. g, m, . c, p,. 9

.y

... - 's.,.M yyG ,, {..,,. a, +. _1 't . pg

. 5

? :,.. = . i..?. v.'. v .s v. ' 5 ' a. ..e.. i'. s = y [ s.' 4[.

• (

s.)]- * -,$

4 74., &. 'k e 9 v..., 4 .x N...+.* , m ~,. -.y\\....-. $. <-., Oa. #. n.,. n-

~, A_ p

. s, + g %, s *. Wi.. ,) f,- *y". a v.- -..? M..., .O *, j c s 4 +., _',, -j, .g

y

..y - +.y y/. (, T Jg.,.p;. 7.<% 4. 4 ,e 4, f 'y i h ,;*-,._ r y " ~un. ;- ., 9 ,., 'ag~.'_ y,,,Q t }p,bj. e..' + y g<

j. -.,

5 + .s .g s,. ' -*' -.'y ? p 4, 2 y. y % l r - ., r [j 4,2 P-q +. T N ?. ~ pp - (. 'y 's l .c.- -.s . /7 ~- 94~.'.. y, 4. - + \\ 4 l s - *Q ' ' z' 4 .6

• b

- + p v l %.. 'e.N '.', pl, ? a g h% l . ?(

• j

. i e'

L,,,c ", o

.v .,, %;,_.4[* - - m ' e., rf 4.3 ._{ \\ y' ~'.' . f.., * *..s L ~ l ....

• w -

g as.....' [- + F 4 g g g i +. ~ n 's l n qs. ..,'x .,; ;9- , e k" 1 p' ',[ w,. ~ - 4 e..

s. -

n.: ~ en., , a; y .-s ;

• '. M.. p '.s. +, e,.,' -

m,.. v. y .~ l p'. , - 45 .- 4 g 4; ,3 f. ? ,g p e O v. C 2 -s~ ,,;j g 3 ,' T a 4 a W W

s. *epaus-

... ~ - t#m i,.: ? [- O UJ A. =; t .-( . + 4 =-- 1 C .J n, -> e 4. *' y my j'.. I' *,,', p [ C 2 g Z -c'

y

• . g

,y t. e Q

• assem...

Iw., .+ f4 g = 1sess - f. , g .e.. O s, A ? .'Y [ ' "'. ,,'[ -) * ' c. ?

• f',..

~~ _ + Q Q

...,y-?

.s .e ..s. , f y., _ + w.. 2 O~ Q q ^Q,,,g% , z,. .,' s q -g. 9 4 s, __ z c

1. ',.i

3. p M.

• u

-.. em . ~.; [.' ~ 7.. .,Y * - O N Ej N p.'$ ' ~ ' " f g,,., y.

• A4 T

) 2 ed.4}w.. -h4j.. y1 vec,. * .~.;._. -.p -.;p p

• A.

.p .46 6KW' T V,. _ a g 4 A* ', .,_w E

• {,.-.#

q/. y,_

< j.N,.....,.

....A . 'w p-

's rQ,,g"gle) ** '

.'3g l p "y; W,. ^ Me ~ l ' &gh - ~ 3 M. l Y h},f.. ,:,. M A. '1 ..s .e' 4 7 YJ ^ ~ .( Eh'

1. k.(d A fq.#. /*g '.,

• [

g

• . e * * '. =

Mg.Caii k f,j .e ' i : ,-] n' o h. J QQ V e,3r% .+. 4htqs ' *:. :. -- ' v. .~ s s v

a... -

.-7.. .s a

• ?

MAMw s q [ ; T l @,,aMMh, N.h,h.,.$M; y '.f g f', 8k"y., " g th.. J. *

• k,j@, - _. - *.

,, g..,,,,Q,.' j M g *h, a % Q ' f ', 7 5. y *

• ~.;

., h ? ,' q j, ' e T y ,,, 4i Q...'- =%. 4 ,f

.4, f*?,,c y,,.

j

• y f. t-q"i!sU..Qs

/.i >r y 4 +- 9f-9 t .a ?.- i. ? r-.-% 5., N +. 4 4,$4j% &p6 W_;w f'%.'?& :.n w&P p%:?%i.\\ ' 4 9." P ' * / ThMW PL%n T O .'.Q'v %;p'+%... a/ - 24 Tig. 2 - Instrument to measure out-of-plumbness AA ?.M O-CH S-N -h. Mi

y The objective of the instrument is to measure the angularity of the 7 installed expansion anchor with respect to a flat surface representing the surface of the embedding material. The flat surface representing the surface of the concrete slab is represented by a smooth steel plate and W i is re ferred to as the reference plate. A perpendicular plate was s s. attached to the threaded portion of the anchor and established the posi-Jj tion with respect to angularity of the installed anchor. A turning n { plate, as shown in Fig. 2, was rotated through 360 degrees while bearing E against the perpendicular plate. The dial gage attached to the edge of aO l s t the turning plate measures tha distance changes between the reference n o plate and perpendicular plate as the turning plate is rotated and the i 7 j stem of the dial gage slides on the reference plate. d The out-of-plumbness of the anchor can be determined from trigon-s s ometry knowing the minimum and maximum dial gage read'ag and the distance O t i from the center of the anchor to the stem of the dial gage. a f This instrument provides a relative measurement of the out-of-s. g plumbness. Possible sources of error are (1) the dial gage is not n C-mounted perpendicular to the turning plate, (2) c. tere is a gap between the anchor stud and the hole in the turning plate where the anchor stud goes through it, (3) the perpendicular plate may not be exactly per-pendicular to the bolt axis because of machining inaccuracies, and (4) high and low spots on the embedding material caused the reference plate to be unrepresentative of the reference surface. Considering these possible sources of error and that maximum and minimum dial gage readings are used in calculating the eut-of plumbness angle, it is reasonable to assume that if the calculated out-:,f-plumbness is less than 3 degrees, the anchor itself probably has a true ( -h-g 7 210 -cas ^3 W

a 4 out-of-plumbness with respect to the embedding surface of less than 3 degrees. The out-of-plumbness for all anchors installed during this test program was less than 3 degrees. W SINGU ANCHOR STATIC TENSION TEST i S Loading Equipment J a n n A framework to apply a tension load to the test anchor was posi-E tioned over the embedment anchor such that the anchor, hydraulic ram, 3 [ load cell and tensioning rod were vertically aligned. Fig. 3 shows a j t O n photograph of the loading fixture and the test setup. The location of B r the pedestals that separated the test specimen and the reaction beam were ~ a 3 n d 8 Paced approximately 2 f t from the loading location which is far enough f from the test anchor location to avoid compression restraining effects S y of the reaction frame. The couple link was designed to pivot and ensure ia that the~ applied tension load remains parallel to the axis of the test E s. anchor. ~ I n t. Instrumentation l The instrumentation used in perfofming the static load test included a strain gage load cell, an analog X-Y recorder, two linear variable differential transducers (LVDT) and a data acquisition system. A desk top mini computer was interfaced to the data acquisition system and was used for data retrieval and storage. Descriptions of the equip-ment and calibration certificates are included in Appendix I. The applied test load was measured with the electrical resistance strain gaged calibrated load cell. The load cell was mounted co-axially (' with the tensioning rod. 1141 mo -casa-3 61 yw y, w--w- -,r -vyy ri- -ym-4w-----n.-~-'y -s---e--,*-y--- w --r w

---

e

rn ~ $I~. ',a; 4%.Qp k..l,,tt; '.'. '. :*.].,... '..('Q. ,, b ?*

.v. % :

.,y q '.> .? Q - o ':a 5.~. s +. a ~, , j, , t 7*, .%*v u, e - g, a ~.... .s,

q. * ;. ', ' ;'. '.;". ;' 7,....

.3.,, s R. '. 4 2 ?,y .c..~ ~

L. *.,n '*

.,,.,.

1. 'y., 4- ..s m V ' -

• g q.pi.)

&_"7 A,..'n.. 5 }. ~. + %.L + r

( _,..g.[ '. ((" [. ' ;f.l'- l ;,,. ' A (..

'..._lle V': A 4 1 5; / 4 ' h -} ~:.j, ,. I:_ l,7 , k ;,, :. ^3 .t r 5 g.'*, 7 :'] -} ,' ~J ) ~,.'l w, 9 9,, f.. 3. -T ",. g };.p.,, 3 [3..,. p. / 4-s' 3. f.4 p[; ' 15 4., g..i g 1, ~ l b. D[ h '.,, " I $ 4'b-' .[. -.. ' W ~ ~. ;,.,, [ 'y .). Y [*,[ -Y g y ~ hg.$jj{ a@:'1pMM .hgN. [f.,: 4 ig

1. .I

'c

. - l4.j l f

.[f? .f f. "r l &' *' N f s.l \\*. .fi - l ? 'l ,; g.; pg h ', _ .y f. ,e r , f,yg g Qf.W' ~., } my 4.q-. ?; g g* s ~ ;k. fag [ _,z 3 _ {... p . ' '$.. ) k l4 y; . ' V 4.k. x -' 'y - q. m f.( g. l ggf. j:. - +. -, 4r \\ .- a ;.., T s '~ + 4

1. s. 6 4.;

~ , f. ' - ..t .4y.3 3 1. &,. ,3 y4 A j- ..3... J. ;.. - ~., y:. ...v y [_

.)..

( ~?y} .,f ' "[. (' s ' W.

L,

.l.".,~.. 7'. : w l - * [:A y + . M, -,, _ j.

• • rsa,

.q%.. ( s:. ,, 1 .,,.,L Q 'vr jzh.{,5 l.!;(' p.. ' g 8.. Q?, .r m- - 7}_ . ] e j., -, ' ' ( *- % .k@.3.. ( "3; b .f -*e ~ -.w v. .....;'. ; y. g o 3,,. fr..8 eq'f *y +. 3.,... <,: O ( ~ (4 '. ' Qs, R.

gg, i.w y

? ', ;",, g,, - [z f..; .-.4 Q p,] y. w 4 .' r y.;_ 9 ;;,..., 8.84 y y j. 4. ',,...

4 4 g,g

. ;c ;O :~. :..,' ' [L .., f. ';,f_ pl Y. 'q d , 3..r- ~ , g..

.< g.

,x ? [',,'.h.ekkV- -. M : 3 .T. I ' [L,.N l f;.

• '.h, ll e ?:

q j.. y f;; @jff '. &. l.,: .f ..f : ? t Y ' h4c p f R;

• e-y '. s f.Q8.g,+qyL, 4.tj

. f, ? $.j. p. @-c [6

• f' k

- 4e 1j.. 4, #[. $s:% q e,.

v. -

- ' '+1 ' - A,,.h.'. +, 3 q spc w./s " f ,5, 6 -,. Kh >. au% ' c..s.,.. ;. ' c. 4 19 f,/ f,, %.-*t. Q.f.;: . (y Q g .:, Q,.c .. ps & (qffy \\ ~ .lf, h? l 5'b l 3- , J: '? q iY M M '" % c M " . _Q 3MASW m. s, D ORIGINAL DOES NOT MEET o Fig. 3 - Tension pullout testing frame with slip instrumentation MICHOFILM STANDAROS. HARD COPY ON FILE '7 " 4 4 ~' l. Enn

^ - ~~ Slip of the expansion anchor in the concrete slab was measured by attaching arms to the test fixture. Slip displacements were measured with LVDT's that were attached to the measurement arms as shown in Fig. 4. The measurement reference point for the LVDT's were at least one W i embedment depth away from the axis of the test anchor. s The load cell output signal and the output signals from the slip 1 j measuring LVDT's were connected to the X-Y plotter. The X-Y plotter made n e an inanediate permanent record of 'he test anchor load-slip behavior. E 3 The instrumentation output signals were also connected to a digital 1 S t data scanner capable of reading the output signals at specific intervals n O dictated by the speed of the controlling computer and data scanner. The i j digital data read by the scanner was therefore not a continuous record .O d but a record of conditions at discrete points on the behavioral curve. A A S (.. - s schematic diagram of the entire data acquisition system is shown in 0 C i Fig. 5. a. e A computer was used.to control the operation of the digital data i s. y scanner and store the digital data on magnetic tape for further plotting n ~~ C-and data analysis. Application of the Tension Pullout Load I I l After the test anchor was installed, the anchor tensioning part of the loading fixture (Fig. 4) was attached and the testing torque was applied. For.the 24 samples which were subjected to the torque-tension i testing, the average torque specified by C-305 was used. The remaining 48 samples, which had not received a torque tension test, were torqued to the minimum and maximum values specified by C-305, as a function of their grouping. t w Y 72.2.0- cNS-N-3 l~ 114#

~ -M > Embedment > Embe dmen t = =- = = Depth Depth a l _______.--_________.I N - Linear Variable Dif ferential Ttansducer (LVDT) [ p o . g I 7 F~~~~~~~~~~~~~1 Plexi-glass U LVDT lloider Section A-A f n t

• T I

i ' g ension Rod T i E y I I )f J -- b [ ~*~ ~ -~*] Anchor A A p Non-magnetic'LVDT --Tensioning L Support Arm Fixture 10l 4 LVDT - ] N [ [ [ ] Test Specimen Y th-t t o Embedment I

l. W est Anchor Depth T

i w s si v t m Fig. 4 - Slip measuring instrumentation for tension pullout test

m -~. i f Load Cell Load Cell Op-Amp P ""# 8"PPIYn 1)uring Experiment N Excitation I X-Y Plotter Supply Voltage O Load ^ g ji Operational j Amplifier Isolation 7 5 LVDT #1 Circuit r 3 -)- LVDT y stip Excitation 3 d N Supply Voltage y g v r LVDT #2 Average LVI)T ) ~ g g .o 9 9 : 4 u a Displacement g y t u q M i Digital Data M Acquisition M Sys tem N A Ob Q Data Read f ( Digital Data s Command j g to l{emory ttulti Test Digital

f p

Data Plotting llewlett-Packard /, k Control Computer, .o ff[' g Printed Digital and / g, Data d Data Storage L4 Slip d D ts.g Fig. 5 Data Acquisition Flou Diagram for Single Anchor Static Tension Test e

~ ~.. _ _ __ 1 _ ~" ^ ' ~ ~ ~ ~ ~ ~ ^ ' ~ ( After starting the hydraulic pump, the computer control was acti-vated. A data sampling period of 4 seconds was used. The load was applied to each sample uniformly by the flow-regulated hydraulic pump. Each test was continued until the applied load caused a failure of the W { test anchor or the concrete slab, or until the anchor slipped out of the s. embedding material at least two bolt diameters. The load at failure and a n mode of failure were recorded on the summary data log sheets found in n e y, Appendix H. Any additional information pertinent to the test was also E N g recorded. 5 N t 3 Test Results C r m a n The test data are given in tabular form in Appendix J. These tables d A include the e nerate ase at the time of the test, the estimated concrets S g strength, the maximum load and slip and the failure mode. An anchor load 7' c i versus anchor slip plot was made for each of the pullout tests. Since a a t t e test group consisted of three pullout tests investigating the same l S, f { combination of variables, all pullout tests in the same test group have l D' l been plotted on the same sheet. Typical test group plots are shown in ~ Figs. 6 and 7. Applied load versus slip graphs for each test group are included in Appendix K. Table 2 sununarizes the static load data from Appendix J. This table includes the lowest load, greatest load, mean (i.e., average) load, standard deviation of the average load and the coefficient of variation (standard deviation divided by mean times 100) for each group. The torque categories identified in Table 2 as "over", " minimum", and " maxi-mum" refer to the C-305 specification requirements. The " minimum" and " maximum" groups used the reconnended minimum and maximum installation i 114 7 3 3 0- C//S-N"3 Y

7 + 0 , 2 ~ ... - ~ ._.~. 1 /. ~ .~ . i ~ C P B

. : : ~

.. ~ 0 _-. _ .~-. . ~ 0... _~ 0 ._ ~ 4_ 5! ~ . ~ 5 n 1 ~ ~ ~ ~ 2 ~ ( /. .. - - ~ \\. I 1 - : ~ I ~5 - - ~ s._ . _ s g':

. i __

. : : ~

. i __

). ..~. ~ 8. _::~ _ b. .k. 3 ...~. /

_-.:~

-. _ 5 _ 9 g. ::.__ : _ ~

._ 5.

_~

._~

_ :.. 5 s w _.1 _~ ~ e ~ . ~._ / P \\\\_. i= ~ h r ) _ i. (' 0_. : c o ~ j

_ i..

_ :.. 5. i _:. ~ _

h_ g.

. _ i..

v

.. ~

1 /.: ... ~ a .... ~ 1 \\.g_: ~

_ i _.

. :. _ 3 . O [ h i.' - ~

_ i..

.. 5

g :

c_: .. ~ e .g 5 -.: _ :.. 3 1.. _ s-.: _.: B ~P\\i__

. 2.. _-. : _ - _

_ - - _ \\ _.N:.: : e s l :._ i._. .g:.N ~..: 1. _ ! - _:..: _ p R y .. i ~.. _-.:. _._ e\\. (. . s. - _- ;.~

_ _ i _.

T .g.: ~

. _ i.

s_.: y: - ~ -. _ i.. 3. L 5.:._ e i _. 5. S l _ :.. ~ .. :. 1s g (_

: ~

N. - : : ~

. _ i..

5 i (. . 2 t N . _.s

-. : ~

c 1 .:g.

- ~

u D .\\ _.

. ~. :

. _ 5. .s g 5._

: ~

. _ ss. O l -. : : ~ . _s . g. a H_

_: :~

s2.. c 5 _ .. ~ i ..:_[Ws: C p 1 . : _s1

. _ ~

i. s_.

- - :~

.. i

. D..N.- y _ :s ._1

_ : :~

i. T .c.3

_ : :~

i o 6

. - : :~

.. i 0__:

6 , 7 N: _. - _

.: :~

... i

. : :~

_:..i. f%.::_._.

. : :~

_... i g { -..:- ~ .. i. i F _.::~ :. _ 5. ) /: - -. :~ :. _ 5.. . __. 5 ./

- : :~

_ 5.. . :.. 5. / : /. :. -. - : _~ _ 5.. . gf../.

~

_ 3

. _. ~.. .j )

/

j ~ f f

~

l . i.. 1 ./ j - : ~ ~ .. ~ _. -. ~ l /Yt . 3._ _. i /.

~

5. jr - ~ . -. ^ _ / O .~ 0 0 0 0 0 0 _l0 0 0 5 0 3 ~ el5 1 ~ 2 4 p ,c

• p., a-l -9<cL acy 2q r

7 l n gg$ b a [ - t 4 % }j ) g 9 N 3, ' 3 1 's 7 96 U ( 66 a_ a_ 4 - 0

i;; .l 0 2 m - j ~ 5 [ [ 1 [

_ - ~

..i.

} :

)

.~:

.:i: C: :.

i:

s. P- .:i: e. . :_ 0 - B - i-h

.: : 0: : : :

i:

r c o 0:

i..

_ __: 5: :.: -

i.

n i . 0 i a .:~.: v 0: :.:

- i:

i:

[ 0: h

3: :.: - _::i-e 1

_.:i: B -...: 5. -. _- _.:i: 7 e

i.

R p 3

i-y 4 _ :...

i-

: _ q_

T . :. :. ~ C.: :::i-

: g:

t )4: e S_

i:

l 8 ( s ~ t /: %s.q::. [ t 7 i

[ :

( r W: :.-

- i-R.

B _. I. - 1.. /

7 i :

O P: :.: -

: }

g:.:_:_

. :.. ~

l

. -.. ~

- 5 a

H. 2 c /_. ::- p-C. i 1 _ - .. 7 ::_:: :_:: p D _ :.: : N.; y

t

. 5 A. .2 T S ih: -. :. ~

=

p::.

Di:

~.:::

. : _ : ~

7 ~.. :. : ~ i : .g gi: i : i .g

_ : _ : ~

F .. :. : ~

. _.: ~

.\\

. :. -.~

- i :

.:\\

.. : ~ .\\ i ~ . \\. .. ~. ~ ~ - i : ~..

T i -

\\

~ ^ t ,\\ : _ : : ~. ~ , q 2 ( } k \\i~ g . h ~ . -. ~ _ - ~ h - [ l0. O o 0 0 0 _o ~ 0 0 O o 0 0 0 O o 0 5 0 5 o 5 2 I q 2 ac{ z" _ _ly ) p C.c b .F 2 li T 4 ) 3 r- _4j y gn 9{yh nt aaA" c c llll l; 4

( N TABLE 2 - SIRO!ARY RESULTS OF EXPANSION ANCHOR TENSION TEETING BOLT DIAMETER = 5/8 in. Manufacturer Hilti Generic type Wedge Embedment depth, in. 2-3/4" Concrete category, psi 4000 Torque category, ft-lbs Over Minimum Maximum fhattestday*, psi 6810 6850 6855 Installation torque, ft-lbs 250 130 160 Low, lbs 6100 5000 7700 Mean, lbs 7500 5700 9100 ,f High, Ibs 8800 6300 10600 Std. deviation, lbs 1360 650 1460 N Coefficient of Variation, % 18.0 11.5 16.1 O S 7 i BOLT DIAMETER = 3/4 in. 7 Manufacturer Hilti ~ Generic tvoe Wedge Embedment deeth in. 3-1/4" ~ Concrete categorv psi 4000 5000 Torque category, ft-lbs Over Minimum Maximum Over Minimum Maximum f' at test day *, psi 6800 6850 6855 6850 6895 6910 Installation torque, ft-lbs 340 240 270 340 240 270 Low, lbs 9300 9600 10900 11400 10400 10100 Mean, lbs 12000 10100 11000 11700 11900 11200 High, 1bs 15600 10400 11000 12000 13000 12000 Std. deviation,1bs 3250 460 60 300 1330 970 Coefficient of Variation, % 27.0 4.6 0.5 2.6 11.2 8.7 Estimated 4443 7x:to-c//s-o'-3 g

TABLE 2 -

SUMMARY

RESULTS OF EXPANSION ANCHOR TENSION TESTING (continued) (' BOLT DIAMETER = 7/8 in. Manufacturer ph4115ns Generic type Wedee Embedment depth, in. 4" Concrete category, psi 4000 5000 Torque category, ft-lbs Over Minimum Maximum Over Minimum Maximum f at test day *, psi 6810 6830 6880 6840 6880 6890 Installation torque, ft-lbs 375 275 325 375 275 325 Low, lb : 14200 17900

• 12900 14300 17500 18400 Mean, lu; 14600 18100 15900 15300 18200 18700 High, lbs 15300 18500 18200 16900 19600.

18900 Std. deviation, lbs 590 350 2710 1400 1180 250 Coefficient of Variation, % 4.0 1.9 17.1 9.1 6.5 1.3 O BOLT DIAMETER = 7/8 in. (continued) Bad anchor O Manufacturer Phillies 7-Generic type Wedee Embedment depth, in. 5-1/2" Concrete category, psi annn 5000 Torque category, ft-lbs Over Minimum Maximum Over Minimum Maximum test day *, psi 6805 6830 6855 6840 6875 6890 f at e Installation torque, ft-lbs 375 275 325 375 275 325 Low, lbs 22100 25100 24100 21400 26100 22100 Mean, lbs 23600 25900 25200 24300 27000 23200 'T High, lbs 25200 26500 27000 25800 27700 24700 Std. deviation, lbs 1560 720 1570 2480 830 1350 Coefficient of Variation, % 6.6 2.8 6.2 10.2 3.1 5.8 BOLT DIAMETER = 1 in. Manufacturer Hilti Generic type Wedee Embedment depth, in. 4-1/2" Concrete category, psi 4000 Torque category, ft-lbs Over Minimum Maximum f at test day *, psi 6810., 6880 6880 e Installation torque ft-lbs 575 425 475 Low, lbs 19700 16100 17400 Mean, 1bs 20600 17500 17600 High, lbs 21800 19000 17700 Std. deviation, lbs 1080 1450 170 Coefficient of Variation, % 5.3 8.3 1.0

• Estimated

~ N1147 73aD-C // Sd/

torques given in C-305. The "over" group had exceeded the maximum specified installation torque during the earlier torque-tension testing. Load data were also captured on an X-Y plotter during the test. These data were reported in a letter report to Bechtel dated April 24, W i 1981. Table 3 gives the results of the c.1ysis of loads from individual load versus displacement plots. The peak loads taken from the X-Y plots J are from 0.9 percent to 6.1 percent higher than the peak loads indicated n f, by the computer recorded data. Both computer and X-Y plotter used the f same load cell reading. The computer recorded the load cell output on a S N t 4 second sampling interval. The X-Y plotter conditioned the load cell n E O r output voltage which represented the load and plotted the load as a continuous function of displacement. The X-Y plotte* data was used to 7 d check the computer recorded data. Because the X-Y plotter represented an ~ A s s additional source of error in calibration, the data recorded by the o C i computer were used to establish the maximum load sustained by the a anchors. e S. l The greatest average pullout load, as shown in Table 2, for a given n C-anchor size, embedment depth and concrete strength, was observed in four groups which used the minimum installation torque. The 3/4 in. diameter-4000 psi-3-1/4 in, embedment depth and 1 in, diameter-l 4000 psi-4-1/2 in. embedment depth overtorqued groups both showed I greater average pullout loads than the minimum and maximum groups. The 5/8 in. diameter and the 7/8 in, diameter, 5000 psi, 4 in. embedment t j depth maximum torque groups showed the highest average pullout load when compared to the over and minimum torque groups. Reviewing the Table 2 data for the lowest average pullout load, " relation to the installation torque for a given anchor size, embedment ( 7Mo-C N

1. '3

C3 i l N 6 O TABLE 3

SUMMARY

RESULTS OF STATIC PULLOUT (STRENGTil) TESTS (FROM X-Y RECORDER) I 4 I ~ MANUFACTURER llILTI PilILLIPS HILTI 7 i GENERAL TYPE WEDGE WEDGE WEDGE BOLT DIAMETER (in.) 5/8 3/4 3/4 7/8 7/8 7/8 7/8 1 eh EMBEDMENT DEPTH (in.) 2-3/4 3-1/4 3-1/4 4 4 5-1/2 5-1/2 4-1/2 CONCRETE TYPE (psi) 4,000 4,000 5,000 4,000 5,000 4,000 5,000 4,000e h LOW (1bs)

• 6,500 9,800 11,800 14,500 15,000 21,800 21,800 20,200

==L i MEAN (1bs)* 7,900 12,330 12,065 14,970 15,970 23,800 24,830 21,100 N 111011 (1bs)* 9,000 16,000 12,300 15,600 17,800 25,800 26,400 22.300 N b STANDARD DEVIATION (1bs) 1,275 3,250 250 570 1,590 2,000 2,625 1,080 h NUMBER OF TESTS 3 3 3 3 3 3 3 3 km I i W

• Values of maximum load carried by the anchor bolt. Low and high values give range of 3 bolt tests.

d4 u cg

t ( depth and concrete strength, indicates no apparent pattern: three were in the overtorque category, three were in the minimum torque category, and two were in the~ maximum torque category. The Table 2 coefficient of variations for most of the groups are W i generally 10 percent or less. Exceptions include the three 5/8 in. S s. diameter Hilti groups. J A review of the Table 2 test data for the Phillips 7/8 in. diameter n y, anchor test groups indicate that the embedment depth affects the ulti-f mate capacity.. The test groups with a 5-1/2 in. embedment depth have C' aq s t signficantly greater ultimate load capacities than the groups which used C3 $r a 4 in. embedment depth. Appendix J also contains information concern-ing the number of turns of the nut required to establich a given torque d level. N A global generalization to relate torque to turns of the nut is 5 g reasonable from the data obtained. At a given torque level, the turns t i required appear to be related to initiation of wedging action. The data a t e indicate that control of installation torque is a more accurate way to I' I insure acceptable anchor preload than counting the turns of the nut. n To describe the behavior of statically tested anchors, the plots presented in Figs. 6 and 7 will be discussed. Fig. 6 shows typical behavior of anchors which exhibit a ductile-type failure mode. Ductile i behavior is characterized by large relative displacement prior to reach-ing anximum load. The test results indicate that ductile behavior is characteristic of deeper embedments. Failure usually results by pulling the anchor from the hole without rupture of the concrete. Fig. 7 shows behavior that might be characterized as brittle. Brittle behavior results when the peak load is reached with little displacement. Once the peak is reached, the concrete of ten fails. Fig. 7 also shows a common bN g 7ne-c N 5-N-3 s

J J l slip behavior of anchors af ter peak load is reached. The jagged trace is the result of the wedge mechanism regripping the sides of the hole as the anchor is pulled out. A plot of anchor diameter versus slip at maximum load in Fig. 8 also W i illustrates the effect of embedment depth on anchor behavior. Only the 5* 7/8 in. diameter Phillips anchor was tested at different embedment J depths. The effect of going from 4 in. to 5-1/2 in. embedment is clear n f, in the more ductile behavior exhibited by the more deeply embedded f anchors. C S 7 t In wedge-type expansion anchors, the bolt stud slips relative to n C Er the wedge and develops increasing load by causing the expansion mechan-ism to wedge itself tighter against the embedment material. Each anchor S d g typically requires at least one bolt diameter slip to develop the maximum s s load. O t I Generally, the failure mode of the samples was characterized by a t 7 e anchor slippage only at the peak observed loading. Subsequent loading, s. ~ { beyond the peak loading, resulted in a cone-type concrete rupture fail-H C ure. The one exception to this failure mode was the 7/8 in diameter anchors, which had the greatest embedment depth (5-1/2 in.). Only one of deeply embedded anchors produced a concrete cone-type failure. The failure mode of these more deeply embedded anchors was characterized by slippage only. Table 4a gives the average ultimate tensile loads for the anchors torqued to the minimum and maximum levels specified in C-305. Table 4b gives the average ultimate tensile loads for the overtorqued anchors. The tables list the anchor diameter, embedment depth, average tensile load and number of anchors tested for each concrete strength category. 1141 fy . r a. no-c u s - "- 3 1 . -.. ~.

2 p ,3 i f, '0 r l O "3 7 I m 1 t ~, 1 al o no om 1 ~ i i 1 1 i 1 LHO L L o8 o 50o5 1 m 1 IM OM MILOO 1 mLL M M o 7/8 - \\ i ? q k .E A E E 1 1 O EooE 5 m l-1 o 11 g 374, l M g EMBEDhENT Q U Torque D j Condition Shallow Deep D A 1 l D ll o o 5 m 0 Om i g 5/8 o g g g,.,, N m M Maximum 1 k/ a e i 9 0 0.2 0.4 0.6 S' lip at maximum load '(in.) g Fig. 8 - Slip at Maximum Load Versus Anchor Diameter V4

l / TABLE 4a - AVERAGE ULTIMATE TENSILE LOADS FOR ANCHORS TORQUED TO SPECIFICATION C-305 VALUES Concrete strength 4000 psi (1) 5000 psi (2) Diameter Embedment Tension No. tested Tension No. tested (in.) (in.) (1bs) (lbs) 5/8 2 3/4 7350 6 3/4 3 1/4 10550 6 11500 6 7/8 4 17800 5 18450 6 N 5 1/2 25550 6 25100 6 70 1 4 1/2 17550 6 g O O O TABLE 4b - AVERAGE ULTIMATE TENSILE LOADS FOR OVERTORQUED ANCHORS T Concrete strength 4000 psi (1) 5000 psi (2) Diameter Embedment Tension No. tested Tension No. tested (in.) (in.) (1bs) (1bs) ~ j 5/8 2 3/4 7500 3 l 3/4 3 1/4 12000 3 11700 3 7/8 4 14600 3 15300 3 5 1/2 23550 3 24250 3 1 4 1/2 20600 3 i (1) Concrete strength at time of test estimated to be 6850 psi ( (2) Concrete strength at time of test estimated to be 6900 psi 1141 722o c ar-7/- 3 g;; l l ....1

It should be noted that the actual concrete strengths at the time the t, tests were performed are almost the same for each concrete category,

SUMMARY

AND CONCLUSIONS W i Static tension pullout tests were conducted on three groups of S S* twenty-four concrete expansion anchors. One group was initially torqued J n to levels higher than the maximum installation torque allowed by the B e y, Specifications. A secor.d group was installed with a torque equal to the E 3 g maximum allowed installation torque. The third group was installed with S 3 t the minimum specified installation torque. The small sample size pre-n e C r cludes global generalization concerning the effect of installation a n torque on the ultimate load-carrying ability of a given size expansion d -~ A anchor. The data scatter as evidenced by standard deviations of test Sj resulcs illustrates the nonuniformity of behavior typical of wedge-type t { concrete expansion anchors. t e 5. Respectfully submitted, I n c. WISS, JANNEY, ELSTN' AND AS.cOOIATES, INC. gO/AR F. D. Heidbrink Project Engineer s P O !! C' "C 3 ? . R. C. Miller h*"".':v$r..y Project Manager n rm .u c.a ,e 3!9/A2 DATE:

• $"Mn SY:

jnJL ~ 114; 72.;;t.o-C // 5-//- 3 /e B

W i s S. J a n n e L E 1 S 'E n e ? I a APPENDIX J n d Summary Static Tension Test Results ) S 0 c i a t 7 2 S. I n c. 1141 73zo-CHS- /j / b)V50

E 4 i V i STATIC TENSION TEST RESULTS: EXPANSION ANCHORS DI8EDDED IN CONCRETE 9 ANCHOR TYPEI Hitti Wedge 80LT DIAMETER: 5/8 in. ^ ^3 1 ? a A J p w 2 n a a n 3. y 9 42 2n j E g 8^ 2' o" te av y-ga fu CC 'L I-3

4 2p e

sa at 3 2 .a 3 3,3 a 4e se a-a v-e 9 s -3 -r a a 3 a a a a a a e a 1 a a e 01/9 2-3/4 250 145 4000 1 6810 81 0.60 5 7700 0.04 Sc 2 6810 81 0.46 4-1/4 8800 0.06 S.C L. 3 6820 82 1.33 5-1/2 6100 0.00 Sc I Cl/10 2-3/4 130 4000 1 6850 88 0.69 5-1/2 5000 0.02 S.C 2 6850 88 0.10 6-1/8 6300 0.03 Sc 3 6850 88 1.89 6-1/4 5700 0.02 S.C f Cl/10 2-3/4 160 4000 1 6855 89 1.15 5-1/8 7700 0.12 S.c 2 6855 89 0.86 3-3/8 8900 0.12 S.C 3 6855 89 0.48 4 10600 0.29 SC C = Concrete ruptures in tension producing cone type failure S = Anclear s!!ppage from embedding material s a ? Z. LO-C //S'- \\ \\~ \\ ba

i sO v STATIC TENSION TEST RESULTS: EXPANSION ANCHORS EMBEDDED IN CONCRETE t ANCHOR TYPE: It!!ti Wedge O BOLT DIAMETER: 3/4 in. i 8-a a a 5 y s.* n r 1 o u 8 23 le. ] u 3 a 8: 3 3t gu -n = ca av a as 3-le 3 u-i ji SS is i

1. 1 55

))5 2i 5 55 i r i a e g a a a a 3 I u a Cl/10 3-l/4 340 255 4000 1 6810 81 0.10 4-5/8 9300 0.01 C ~~L4 2 6770 75 1.26 6-5/8 11100 0.03 S.C h C1/9 3 6820 82 1.69 3-3/4 15600 0.19 S.C C1/9 3-1/4 240 4000 1 6850 88 1.69 4-1/4 9600 0.32 S.C 2 6850 88 3.37 4-1/4 10400 0.16 S.C 3 6850 88 1.31 4-3/8 10400 0.13 S.C C1/9 3-l/4 270 4000 1 6855 89 1.42 6-7/8 11000 0.04 S.C 2 6855 89 0.76 5-5/8 10900 0.01 S.C 3 6855 89 0.40 5-1/2 11000 0.08 ' S.C Dl/2 3-1/4 340 255 5000 1 6845 78 0.43 4-3/4 11700 0.02 S.C 2 6845 78 1.15 6-3/4 11400 0.01 C Dl/1 3 6850 79 1.90 4 12000 0.05 S.C Dl/2 3-1/4 240 5000 1 6895 88 1.90 4-3/4 10400 0.13 S.C 2 6895 38 0.74 4-3/4 12200 0.16 S.C 3 6895 88 0.28 13000 0.30 C Dl/1 3-l/4 270 5000 1 6910 91 2.32 6 11400 0.01 C 2 6910 91 0.31 6-1/4 10100 0.05 C 3 6910 91 1.02 4-1/2 12000 0.04 S.C C = Concrete ruptures in tension producing cone type failure S = Anchor slippage from embedding material x a U\\ L 7 E- & O# C //S~. \\ \\ ~ \\ N .b-a

+ ~ t Of O i V 3 STATIC TENSION TE37 RESULTS: EIFANSION ANCHORS Df8EDDED IN CONCRETE 7 ANCHOR TYPE: Phillips Wedge i 801.T DIAMETER: 7/8 in. 4 i a 3 u 3 E D S a a j s . 'a ja W 2 ?? a-a A 0 "5 " A =# 5 22 UR o 22 3 3n 3 e a na

== 3= a

y

., j n u, "t na e 44 2 e

s. 3

=

_3 i

3 T* 3" 4 4" g ll 2 Zu = 2 ~ 3 I s0 a 0 a w H d sU a w ac 2 m. m at C1/10 4 375 300 4000 1 6770 75 1.05 6-1/2 14200 0.04 S ~ Cl/9 2 6820 82 2.00 6 15300 0.17 S.C i y Ct/10 3 6820 82 0.57

• 5-1/2 14400 0.07 S.C u

i C1/10 4 275 4000 1 6830 84 1.27 4-3/4 17900 0.43 S.C a 2 6830 84 2.23 4 17900 0.37 S.C 3 6830 84 0.32 5-1/2 - 18500 0.26 S.C C1/9 4 325 4000 1 6880 94 1.15 6 16500 0.14 C j 2 6880 94 1.32 5-1/4 18200 0.24 S.C 6 3 6880 94 2.07 5-3/4 12900 0.18 S.C b D1/2 4 375 300 5000 1 6840 77 0.77 5 16900 0.01 8 b 2 6840 77 1.77 6 14700 0.02 S.C 3 6840 77 1.06 6-1/4 14300 0.02 S b D1/2 4 275 5000 1 6875 84 0.61 4-1/2 17500 0.18 S.C b 2 6880 85 0.61 3-3/4 17600 0.21 S.C 3 6880 85 0.45 5-1/2 19600 0.28 S.C Dl/2 4 325 5000 1 6890 87 0.45 5-7/8 18400 ' O.14 C 2 6890 87 0.62 6 .18900 ' O.45 S.C 3 6890 87 0.78 6-1/4 0.18 S.C a - LVDT moved into a non-linear range while tensioning rod straightened out b - Anchor slipped and caught repeatedly during loading g g () C = Concrete ruptures in tension producing cone type fatture 5 - Anchor slippage f rom embedding material f z '>- ? z zo-etts.\\ d w-ua

.1[2.=.1 t u 'P*R *2"tT'2 .J / / (**T)

• 23 :2 2"A %AS ::

Ma 7 P.ot an=T m 2. dits ddd ddd ddd ddd ddd ddd ih N ("Il N 888 888 g:88 838 ::-]=== O 88 888 m-

=~ ::

am: --- --- ~~~ ~~~ -~~ ~~~ g g td 8 c"c ete cce ett ec N ' 'T2 'd 2..a.'-1 s. 'T3 .?.. 777 777 " T.T ??? ?..? Y = 2., a ~~ a B1 R <=> 1e T2 11 2 1

a ::= :::=== ::= :::

z..,m,.--

ada aad da aad daa add

s

=. -=2 ue iyi 2.1 2..!"1., 888 ::= :::=== ::= :: a.! ~ c a-o t $]8s a 8< = J !. gss saa=== sss cce as8 8 a

= :::====== ::

e: a

=.

m - m >= =T2.dn -- a : = a 1 g u....

2. m a

a g g a a

I.

e e, o "~%V. .". II (est-21) g g 't .nb2ol suT2..I e s s e q gq m

== 1 .E. o G q (.91-2 3) a :a .nb201 ROT 3.IT.2sul em e= e. w w s. m m m N m as a as

o. %".

4 E '. N N N N N N Q (.aT) 2 2 2 2 2 2

{ }y 42deq suamp. qui E

A E E A A 4 v< B B k = N N n

==.s a==Tsads 2 2 2 2 2 2 i s u u v a a o 1141 3-4 &VSU ?. 9>r0 r i I l

m y-O i T

• ?

STATIC TENSION TEST RESULTS: EXPANSION ANCHORS EMBEDDED IN CONCRETE ANCHOR TTPE: Hilt! Wedge q BOLT DIAMETER: 1 in. 1 e a a a 3 i y u ~ E n o o 1 u 3-T 9 eu g f. j ga]u 3 a s2 3 ^ -n ~ = av u y-

• n s

3-le 3 u-i is SS 35 i "1 is 135 ji 5

S i

i! 8 -3 = a a a a 8 = a i 1 a a .l a u C1/9 4-1/2 575 450 4000 1 6810 81 2.53 4-7/8 20300 0.18 S,C c 2 6810 81 0.53 4 19700 0.21 5,C 4 3 6810 81 1.56 4-3/4 21800 0.13 S.C CI/9 4-l/2 425 4000 1 6860 70 1.77 5-7/8 16100 0.10 S.C 2 6880 94 2.21 5-7/8 17400 0.06 C j 3 6880 94 1.87 4-1/8 19000 0.26 S j C1/9 4-l/2 475 4000 1 6880 94 0.93 5-1/2 17700 0.09 S.C 2 6880 94 1.74 4-3/8 17700 0.17 5 3 6880 94 1.75 4-1/8 17400 0.22 SC C = Concrete ruptures in tension producing cone type fatture S = Anchor slippage from embedding material 7 2.&O-6 //P D-\\ t . 4, hn A - a.

Enclosure h s Serial 206SL CONCRETE EXPANSION ANCHOR RELAXATION TESTS

1. U" W

i EECHTEL POWER CORPORATION s s. BECHTEL NO. 7220 J a B I~ WJE NO. 80804Q 9 s( e y. 'May 78, 1982 E I s t n N f a n d Enclosure L contains the Test Report j A only. /. s The Appendices to this report are available s o c at the Consumers Power Offices in Jackson, Michigan. e S. I n c. WISS, JANNEY, ELSTNER AND ASSOCIATES, INC. 330 Pfingsten Road Northbrook, Illinois 60062 (312) 272-7400

=_ d a 1 CONCRETE EXPANSION ANCHOR RELAXATION TESTS FOR BECHTEL POWER CORPORATION W BECHTEL NO. 7220 1 5 WJE NO. 80804Q 5. Jj May 28, 1982 n eI' INTRODUCTION E I S Wiss, Janney, Elstner &nd Associates, Inc. (WJE) was retained by n e Bechtel Power Corporation to perform a series of static tests on concrete j expansion anchors. The anchors are intended for use at the Consumers d Power Company Midland Nuclear Power Station Units 1 and 2. The test A series was performed in order to quantify tha structural behavior char-O f acteristics of the anchors. All tests were performed at the WJE labora-aj tory in Northbrook, Illinois during the period March 9,1981 to March 3, S. 1982. g n c. This report contains the results of preload relaxation tests con-ducted on concrete expansion anchors. Preload is established in expan-sion anchors by tightening a nut on the anchor bolt to a specified torque level. Preload is a tensile force imposed on the anchor bolt by the installathe torque. The decay of the installation preload wich time is termed relaxation. In these tests, after the initial tension level in the bolt was recorded, the load was monitored periodically over a twelve month interval to record the decrease in load.. - --. -.--

REFERENCE DOCUMENTS The tests on the concrete expansion anchors were performed in accordance with the following specifications and procedures: i S s, Consumers Power Company Specification 7220-C-115(Q), Revi-sion 1, " Technical Specification for Relaxation Tests of y a Expansion (Stud-Type) Concrete Anchors" n D Consumers Power Company Specification 7220-C-305(Q), Revi-y, sion 13, " Technical Specification for Design, Furnishing, E Installati n and Testing f Expansion Type Concrete Anchors j for the Consumers Power Company Midland Plant, Units 1 and 2, 5 Midland, Michigan t Consumers Power Company Specification 7220-C-23, Revision 8, r " General Requirements for Supplier Quality Assurance Programs a for the Midland Plant Units 1 and 2 for Consumers Power Com-n pany d A Wiss, Janney, Elstner and Associates, Inc. Procedures Manual OP-29, Revision 1, " Procedures Manual for Static, Torque-Ten-s l sion, and Relaxation Testing of Expansion Anchors Embedded in c Concrete for Bechtel Power Corporation" I a t e The Consumers Power Company Specifications 7220-C-Il5(Q)-Revi-s. I sion 1, 7220-C-305(Q)-Revision 13, and 7220-G-23-Revision 8 are enclosed n c. in Appendices A, B and C, respectively. These specifications are here ~ af ter referred to as Specifications C-115, C-305 and G-23, respectively, or simply as the Specifications. The WJE Procedures Manual OP-29 is enclosed in Appendix D. This operating procedure provides an outline of the steps to be taken in accumulating the physical test data for the test program. SCOPE The expansion anchors which were tested are generically identified as drilled-in expansion anchors. All anchors used in the test program --

B were manufactured by Hilti, Inc. of Tulsa, Oklahoma, except the 7/8 in. diameter anchors which were manufactured by ITT-Phillips Drill Division of Michigan City, Indiana. Expansion anchors are sold under a variety of trade names, each being associated with a different manufacturer. The W i anchors achieve load carrying capacity through a wedge mechanism located S S, at the bottom of the anchor. The wedge is expanded against the side wall 1 of the embedmant hole by tightening a nut which tends to pull the anchor n f, from the hole. When the wedge jams against the side of the hole, further f tightening of the nut establishes a tensile load on the anchor bolt. s t The Hilti wedge anchor system uses different steels in the bolt n e I shank and in the wedge. The bolt shank is made of AISI llL41 or 1144 steel meeting the chemical requirements of ASTM A108. The two inde-d g pendent expansion wedges near the becom of the anchor are made of AISI s g 1050 spring steel. The nuts supplied with each anchor meet the require-C i ments of ASTM A307 Grade A. The bolt shank material has an approximate a t e minimum yield and tensile strength of 95,000 and 105,000 psi, respec-S. l tively. n C The wedge anchor system manufactured by ITT-Phillips is a two-part' anchor system. The bolt shank is an AISI 1213 steel having an approxi-mate yield and tensile strength of 63000 and 90000 psi, respectively. The one-piece wrap-around expander clip (wedge) is made of Type 302 stainless steel. All testing was performed in accordance with the Consumers Power Company Specification C-ll5. This specification was provided by Bechtel Power Corporation. All anchors were installed in accordance with the anchor manufac-l turer's recommended procedures and Specification C-305. All testing conditions were in accordance with the Specifications...

A total of 36 expansion anchors were tested in the program. Table 1 gives the specific groups for each test configuration, which included-different manufacturers, anchor diameters, embedment depths and install-ation torques. A total of three samples were used for each test config-W i uration. S S, .I IDENTIFICATION SCHEME FOR INDITIDifAL TESTS a n a f, All test samples were coded by a unique test mark which identified f all the characteristics indicative of the test. The test mark was s t composed of an eight part alphanumeric name. The name parts were separ-n e I ated by dashes and identify: (a) type of experiment, (b) specimen name, (c) expansion anchor type and bolc diameter, (d) depth of anchor embed-d g ment, (e) installation torque, (f) test number in the test series, and s s (g) concrete strength and client identification. O C i The WJE manual OP-29, which is enclosed in Appendix D, gives addi-a t e tional details regarding the identification scheme. An example of the s. I identification code is given below. ne Example: R -C1/5-H/W(1/2)-2 1/4-55-1-4000(BPC). 1 2 3 4 56 7 8 1. Relaxation test 2. Concrete Specimen No. 5 3. Hilti: Wedge anchor with 1/2 in. bolt diameter 4. An embedment depth of 2-1/4 in. bethre torquing 5. An installation torque of 55 ft-lbs 6. Test No. 1 of the test series 7. Nominal concrete strength of 4000 psi. 8. Client is Bechtel Power Corporation (BPC).. -.. - ~.

4 TABLE 1 - RELAXATION TESTS Depth of Anchor Embedment Installation Diameter Before Torquing Torque Manufacturer (in.) (in.) (ft-lb) Hilti 1/4 1 1/8 10 Hilti 3/8 1 5/8 35 Hilti 1/2 2 1/4 55 Hilti 5/8 2 3/4 160 5/8 4 160 Hilti 3/4 3 1/4 270 3/4 4 270 3/4 5 270 Phillips 7/8 4 325 7/8 5 1/2 325 Hilti 1 4 1/2 475 1 6 475 v 4. _. -.

s a TEST SPECIMEN MATERIALS Concrete test specimens and expansion anchors were supplied by W Bechtel Power Corporation and delivered to the WJE Lab in Northbrook, I s s. Illinois. Certificates of Conformance with the material requirements of f the Specifications were furnished by Bechtel to WJE. Concrete slabs were n D stored inside the laboratory under ambient laboratory conditions. Y, g Expansion anchors were segregated and stored in the laboratory to pre-I s clude unintentional use of the anchors or substitution of other anchors. t n Structural plates and shapes used as parts of testing assemblies r met the requirements of ASTM A 36, Specification for Structural Steel. a n d Bechtel Power Corporation provided 3 ft x 3 ft x 1 ft concrete A s slabs with nominal 4000 psi compressive strength. Six anchors were S 0 e installed in each of six slabs. i a Compressive strength data, obtained from 6 in. x 12 in. cylinders, e s, were also received from Bechtel Corp. An evaluation of the data indi-I H cates that the concrete had a 90 day compressive strength of approxi-mately 7000 psi. The concrete test data and compressive strength versus age graphs are enclosed in Appendix E. SPECIMEN TRANSPORTATION AND LOCATIONS OF EXPERIMENTS Criteria for Transportation Expansion anchors were to be installed in concrete slabs containing no significant defects such as cracks, honeycombs or internal voids. Specimen stresses during transport were to be checked to ensure that maximum stresses were below the point at which cracking could occur. The.-.. -.. -

slabs were received in good condition with no cracking observed. No inspection was performed by WJE to determine if the slabs were free from internal defects, but the surface appearance of the slabs indicated the concrete was dense and well consolidated. W i 5 s. Lifting Inserts for Specimen Handling J a n Lifting inserts were cast into the specimen to facilitate trans-n e y, portation and handling. The lif ting inserts were positioned in each test Ej specimen so that they caused miniatan restraint on the testing of anchors s t embedded into the test specimen face. n e I L cation of Anchors a n d g Single anchors were embedded inte reinforced concrete specimens. sj Minimum center-to-center and edge spacings listed in the Specifications C i were followed in locating the anchors. a t e In all test specimens, the anchors were located to avoid the effects s, I of embedded lifting inserts in the concrete specimens. n l C EXPERIMENT PREPARATION i All anchors were installed using the manufacturer's recoussended l installation procedure. These procedures are included in Appendix F. i Expansion Anchor Installation Drilling of embedment holes - Test expansion anchor embedment hole locations were marked on the test specimen prior to drilling each hole.

~ i Holes were spaced to meet the requirements of ASTM E 488 and the Specifi-cations. The embedment holes were located and drilled to minimize the effects of restraint of the lif ting inserts. Holes for the test expansion anchors were drilled af ter the con-W i crete had cured for a minimum of 45 days. The diameter of the expansion s S' anchor hole was in accordance with the manufacturer's recommended pro-J cedure. Each hole was drilled with a carbide tipped bit, meeting the n y, requirements of ANSI B94.12 (1968). The recommended diameter drill bit f was installed in a roto hammer drill corresponding to the recommended s t drill for the size and diameter of the expansion anchor to be installed. n er Each time the drill bit was used, its use was recorded. When the bit had j been used 100 times, the bit was discarded. The depth of the anchor hole d was drilled deeper than the specified embedment depth. s S Anchor installation - Before each anchor was placed into the embed-i a ment hole, loose dust resulting from the drilling operation was removed e 5. by air pressure. Each anchor was inserted into the cleaned hole and I n driven into the hole to the specified embedment depth. c The anchors require rotation of a nut (turned element) on the anchor stud to expand the wedging mechanism against the concrete. A torque in accordance with ti.e Specifications was applied. The application of this i torque was termed the installation torque. Measured installation characteristics - During anchor installation, measurements relating to the conditions of the installation were I recorded on data sheets. The data sheets are found in Appendix G. Recorded information regarding installation included verification that the manufacturer's equipment and recommended installation procedure were followed.,

1 Anchor Condition Prior to Test All anchors were inserted in their holes with an out-of plumbness of less than three degrees. A sketch of the instrument to measure the { out-of-plumbness angularity is shown in Fig. 1. The out-of-plumbness S. y was measured before the anchor was torqued to the installation torque. a The objective of the instrument is to measure the angularity of the n n e y, expansion anchor with respect to a flat surface representing the surface E [ of the embedding material. The flat surface representing the surface of s { the concrete slab is represented by a smooth steel plate and is referred i e I to as the reference plate. A perpendicular plate was attached to the a threaded portion of the anchor and established the position with respect n d g to angularity of the installed anchor. A turning plate, as shown in sj Fig.1, was rotated through 360 degrees while bearing rgainst the per-C i pendicular plate. The. dial gage attached to the edge of the turning a t e plate measures the distance changes between the reference plate and 5. I Perpendicular plate as the turning plate is rotated and the stem of the n dial gage slides on the reference plate. The out-of-plumbness of the anchor can be determined from trigon-ometry knowing the minimum and maximum dial gage reading and the distance from the center of the anchor to the stem of the dial gage. This instrument provides a relative measurement of the out-of-plumbness. Possible sources of error are (1) the dial gage is not ] mounted perpendicular to the turning plate, (2) there is a gap between the anchor stud and the hole in the turning plate where the anchor stud goes through it, (3) the perpendicular plate may not be exactly per-t pendicular to the bolt axis because of machining inaccuracies, and (4)

-Plumbness Angle (c) Dial Gage NQ Expansion Anchor Stud N N N Turning Plate N Znstru nt ConstanC Perpendicular Plate ( Reference Plate _ 1 I i 1 1 I Embedding Materiai l ~ l _1 Maximum Dial Gage Reading - Minimum Dial Gage Reading a = tan 2 (Instrument Constant) l Fig. 1 Instrument to Measure Out-of-Plumbness 10 .. ~

high and low spots on the embedding material caused the reference plate to be unrepresentative of the reference surface. Considering these possible sources of error and that maximum and minimum dial gage readings are used in calculating the out-of-plumbness f angle, it is reasonable to assume that if the calculated out-of-plumb-5, y ness is less than 3 degrees, the anche,r itse'f probably has a true out-a g of-Plumbness with respect to the embedding surface of 'less than e y, 3 degrees. E } The test setup precluded making out-of-plumbness checks af ter the S [ anchors were initially torqued. These measurements were made after the er load cells were removed. a n d Embedment Materials and Inst 11ation Requiremeets A Sj Each anchor was embedded in accordance with the manufacturer's c i instructions and according to the procedures outlined earlier. Each a t e test anchor was an unused new anchor. After initial torquing, no addi-5. I tional applied load was imposed for the duration of the relaxation test. n c. A total of three samples were used for each group. Variations between groups included the anchor type, bolt size, embedment depth and installation torque. e pstrumentation The instrumentation used in performing the relaxation test included torque wrenches (2), load cells (36) and a bridge type strain indicator. Calibration certificates for these instruments and the Baldwin hydraulic test machine used to calibrate the load cells are enclosed in Appendix H. 4 _._

Direct reading calibrated torque wrenches manufactured by Van F. Belknap of Detroit, Michigan, were used to measure the installation torque. The torque wrench for the anchor sizes 5/8 in. diameter or less had a range of 0-175 f t Ib. The torque wrench for the 3/4 in., 7/8 in. W i 'and 1 in. diameter anchors had a range of 0-600 ft Ib. S s* Load cells were used to measure the tension in the anchor bolt J induced by tightening the nut on the bolt. The load cells were manufac-n f, tured by Sensotec, Inc. of Columbus, Ohio. The load cells for the f 1/4 in. diameter bolts were Model D, 2,500 lb capacity. For the 3/8 in, s t and 1/2 in. diameter bolts Model D, 10,000 lb capacity load cells were n used. All other bolts were tested with Model D, 30,000 lb capacity load j cells. The Model D load cells are constructed with foil type strain d gages wired into four active arms in a Wheatstone Bridge configuration. S g The strain gages are bonded to the sensing element of the load cell. The C i strain gages are provided with temperature compensating circuits which a h offset the change in resistance of the bridge due to temperature, and s. } compensate for the change in modulus of elasticity of the strain gages. n C-The load cells are rated as follows: Output: 2 mv/v Bridge resistance: 350 ohms Linearity and hysteresis: .5% to.6% Repeatability: .1% Temperature effect on: bridge resistance .005% full scale / degrees F modulus of elasticity .01% reading / degrees F Compensated temperature range: 60 degrees F to 160 degrees F _ _ _ _ - - - - ~- - -- - -- l

Prior to using the load cells, each one was calibrated in a Baldwin hydraulic test machine following a standard calibration procedure. The results of these calibrations are provided in Appendix H. The calibra-tion procedure is in Appendix I. W A bridge type strain indicator, Model P-350A, manufactured by i Vishay Instruments of Raleigh, North Carolina, was used to read load cell f output. This instrument has an accuracy of.01 percent or 5 micro-n D strain, whichever is greater. The analog display can be read to the Y. nearest 1 microstrain. g I s { TESTING e I Test Setup a n d I g A photograph of the test setup is shown in Fig. 2. All relaxation Sj tests were conducted using a plate beneath the anchor nut to provide 1 C { uniform bearing for the load cell. Plates on top of the load cell t e distributed the load from the out and washer and also simulated an 5. I attachment plate. The plate (s) were at least as thick as the diameter of D the fastener. The bolt hole size in the plate was drilled nomina 11y' 1/16 in. greater than the anchor diameter. The plate on which the washer rested was restrained from rotation as the nut was tightened. Each anchor was torqued initially during installation and then lef t undisturbed during the test. Measurement of the load condition existing in each anchor was made on installation, and at 2 hours, 4 hours, 8 hours, 1 day, 7 days, 14 days, 28 days and 91 days. Subsequent measurements were made approximately at monthly intervals. All readings are referenced to the time the installation torque was applied. Figure 3 is a photograph of the Vishay strain indicating instrument used to read % j... 1 .s .j - c. r . L.,.',>. ' h.., -

j. *>,

g. 7 l

,., '.,,. Ry, A'*., ,,u

_. N

.,,., ~,, -., ^, -,- n '.7..-_

• y ' -, -s

,. ' ;g z. -.5,.' y'_ y i +,

;...i, '

,..,u,, , ; sy, '... .z,,YjIU - [- 7, 9 ,,,* '. 'V s, p - y ',[ f ,V- 't [. ' ' [ [J-".. ,...~. ,%..,.',,_.:.',.,', : g.s., .. :[ ;- ' '. - ;a. .5;"- et, ;.

-s./ '

-%.(.,,,, - { g'.1, ,y,- i ^ ._,.,.,' '} ' N'. i ; '. '..

• 9.,r

s....

4 +* .r., .. s t i...., - ' 1 c. ' ' s ..,' j _ ',* '-[, l ' .,y .- -, W 5 ,' - Y ll. "w 7, . g.' g s. ~ J:

y..

t, N';. 7 s,_-e'.,, .. ',..' ? w l ;'r .,' q ) y. ,..'v_ + f y.,. ~% .A a ..., 4 _ a. .4 .., jy,, - 4 v. ,r;,.. .,-l.[ p }.. '. * L;.. _. m. - - - +..-- 7 :.[_ :[ ;.. 3,. s,.:.._' e ^ j,- j',. ', y l,] ; l..,

p..

.u,. _ ;.. ; -L ' f : 3'

• * ;_, -_ s ', :. ;

Sr. w.. .,y .,.~ p..< ,a n,,... : a.. ... ?., .. u-k;. t....e,

,-. w...j a

w., ;'.,

/,2 - i .9

• .'4 8

g 3 -? i. < %,;... ' e

9. :* : v..v

.:... + w.,r..'_ L x. ... :9;. ;... n e.- -z n., ,.,;M},?J. Vg,; A sp.+ ;&_*; r ;.

• e c. #., ->.,...,. p,

?,,,. t'?y,r,u 6 w' : - 7..... ; + yY.- r .~ u. g j ' ' f. ~ ' x

,%.l*4

" m s.v. .g a-g , 'i ' f;;.Q,,, !.i < '- .;f. Si JTyt 't ;-, ~A_ li. ' d Gi. - } }; . H 'ik 17

. g;.4fg$ ] ~ Q i.

^ j* . ;,j, ' El 4 ,} .g ? :.....,; ...,.g _ ;, 7 ;,,. T7.g : g 3,4 ,_ g ,.4 _ g. 7. 4.,.g y 7,. 4M.%,... s. y,. a, + r, ,g w.s-p av .; a ,yy.f. s 2-

1. }'+O-t..e ', ?.

34 l 7p p p., ...? ;lY Q.QjfQk _ W, -Q Qf ', ~ ~ <= ~ a . 3,< 9 : s . f. l M / +. r~-- --. -* 7._ a i ( l s l p ~ s e.,,... 7 s. 'y' % + +, %.p' ~ + 7R . a C v; }&&%94+4.%QQm ~ -).l.: o ~ ?e WMySAqd -yv. hk,ON'k,,M *wwn .gpf g ,5E M['. a [ '. dN;.y . u. hQ k h p-h*[ Q Q, y y ,.f f b " 9.J t J. n-Am v.4 W + I l Fig. 2 - Load cell in place on expansion anchor.-

e f 4 e 1 l l l i l _..q-U LW ~. ~,..' - ' - a W -' \\ . m m $$}, n.h $ g. h.. n. w $,$ %,n ?l i ~ i w g: u_m.n. a m ~ ,J.M fp2 '+i.cr w,%. X.9 fpV ? -.2,%.x u g YNINQS sht: >%N'hp% gg e &>1. p a &n Q.c. n?5ff Y g c. I .m \\ ge, rMis". 2 d' -yf p hg + "' Q Mr.G g+ - u n

. -%.,8 m

e. -n _= - n r , im g W = q sp.pg. m

• ...@@@g * *
• y

-r i _ _y= L % n_-3.;,. ,c%>;;g A u g ak-k - x s g i rl.Af&., w.:y?. ~ %9W 3 i k? t a >1ssv g r y;,0n. w 2 i o 2 ? ~

l o

o M,d8 v...; ,,.,Ng':m t -.i. E, ff.- O p' 2%%%m+ ;f,, _ ;f.dG} 4 ' C*.r.- J@. <. ( 4? i1 u ? [k3 b :.a..u 4.e'fg+i{>f .j if, -N i:Li'.T a ,h,:t {,}.b;g.,w..; ewa p u t.* ?. y, s w:.-; - = ~ e w f; g.y a +c - ~ .,.y. A.c+, t ?!b 2 5 a ~ u w. ~ ~ s.yir:,;u k. m, T,, s ~ - % *e ~ +4 -4.w 4 9.. j ;y 4,5 g,. m.m 4 v or,a eo ,[N' h . gf3E h ',.'2; Phfh ~ nmdk i@_WEM?%ji %3 = r e w A, $ ;dj. Q S

1. • f 6 Mnh M

'y :@g $!g hNhYYh* w e ~ m. % %j $ $ hb%GitM &i w ? 751W e - :nwm;v.-n,-,..n.s.1%w 9- ?sv.c. u.e.m. -g p. g .g Mpf, g R.,l. QG. y 1,-.&.f,th 's.Q:7 u .%. C . g.R '4,+,It .=$1% '.~f Q,.!. f ' Y.*.* ( s., x=a%bN /Mi/Mv-w 9. ~v%, iA'[ a,;h M.$ 'A J W~ 29. *.-e u cy..h e. Am

m. j,.

m ,r.. r p 4; i y . g. ' e _ m e+, s.*. s n.5,f s 'Ss. m'.w s.,. t 4

)

the load cells. No retorquing or loosening of the anchor nut was allowed during the test period. All data were manually recorded on sample data sheets found in Appendix G (three months of data) and Appendix K (all data). An overview of the test setup is shown in Fig. 4. W i Test Results J a n Results After Three Months. The test data are given in tabular form n e y, in Table 2. This table includes the anchor size, installation torque, Ej embedment depth, initial' load and percent load remaining in each anchor s { at the specified time intervals. The measured compressive strength over e the initial period of the relaxation testing was 5670 at 28 days to 6860 I a n at 90 days. d j g The average percen_ load remaining in each anchor group and 5j standard deviation (relative to the average) af ter 91 days is given in C i Table 3. This test data was evaluated using a one-way analysis of a t e variance test. This statistical technique is diseassed in many s. I statistical textbooks. The "one-way" indicates one variable is under n study. For the specific objectives of this analysis, the test groups are the variables. The,bjective was to determine if there are significant differences in the average percent load remaining at 91 days. The I analysis of variance tests examine the variance within groups, as well as l the variance of the means. The variance within groups quantifies the random error of the tests. The variance of the means quantifies the variation of the average ratings of each group relative to the other groups. If there are relatively large variations within the means, as compared to the variance within groups, it can be concluded that a significant difference exists between the groups. However, if the f

A _a .a sg S e 4 17.- z,e ,-s l . g., .y - nqf 1 k.":t1.t- '~ ~ 2 };T-[- 's 'f ~ ?' , ? ~T n u?.9 - $.1~ ~ .t a fy ~h ~$ -r.. ' ~ g,- s " ' s ..c y g ? ., t% h .'r7p. ; i f[;i \\ ,.;;f: ,f.j k r p "Q $)k lik : I-7 E t-4 c w w ew:..., =:::;[,u[g.e,j, ,I: ~~~'~-?' o r l's ~[*A sf ' ' '. 5 ./ t j - a ~ ..f h[g ".{ 2- .,. - ( .i [. _ J,-j g.g h~0/dfj i [. j 1 -; ~ T ( .h[ Ik y ~ - -

i o l j TABLE 2 - PERCENT LOAD REMAINING IN ANCHOR i Anchor Installation Embedment Initial Time Diameter Torque Depth Load 2 4 8 24 7 14 28 91 (in.) (ft-lbs) (in.) (1bs) hrs.

hrs, hrs.

hrs. days days days days 1/4 10 1-1/8 948 73 70 67 62 53 51 49 46 1003 70 67 64 59 51 49 45 41 i 891 72 70 67 62 54 53 50 48 ~ 3/8 35 1-5/8 1652 70 67 64 58 49 46 40 39 1774 69 65 61 54 45 43 41 42 1842 58 55 52 46 38 34 30 22 1 i 1 i 1/2 55 2-1/4 4177 73 71 68 64 57 54 48 43 l 2796 68 64 61 55 45 42 29 34 4 2870 67 64 61 56 49 46 32 39 ? S/8 160 2-3/4 3953 80 77 74 68 56 53 49 43 4145 73 70 67 61 52 49 45 41 i f 3073 72 69 65 59 48 45 42 38 4 3226 81 77 74 74 65 63 60 53 1 5200 81 78 74 69 59 56 53 47 4092 81 79 76 73 64 63 61 58 i 3/4 270 3-1/4 6514 80 77 74 70 62 60 57 52 7517 71 67 64 58 'e9 45 41 35 j 5686 86 83 80 76 66 63 59 53 4 6124 75 72 69 64 56 53 50 43 l 3746 86 83 81 77 71 68 65 60 j 11227 72 69 65 60 51 48 44 37 9 4 5 7037 72 69 66 61 54 51 48 42 1 7103 74 71 68 64 57 54 52 45 { 11333 69 65 62 57 50 48 45 40 i 1

TABLE 2 - PERCENT LOAD REMAINING IN ANC110R (Continued) Anchor Installation Embedment Initial Time Diameter Torque Depth Load 2 4 8 24 7 14 28 91 l (in.) (ft-lbs) (in.) (1hs) hrs. hrs. hrs. hrs. days days days days 7/8 325 4 11553 78 74 71 66 57 54 51 44 8637 75 72 68 63 54 50 46 39 i 6739 78 74 ,70 64 54 51 47 40 l 5-1/2 7431 88 86 84 80 72 70 67 64 10262 77 74 71 65 57 54 50 38 4496 95 93 91 88 79 75 71 63 l 1 475 4-1/2 6288 85 81 78 74 68 66 63 62 J. 8231 76 73 70 65 55 51 47 38 t y 6919 86 84 82 80 74 72 71 70 l ' 6 10569 73 70 67 62 56 54 52 48 8766 67 63 59 54 47 46 44 40 7902 69 65 61 55 48 47 45 43 i NOTE: All anchors embedded in nominal 4000 psi concrete. Concrete strength at time of test initiation was approximately 6500 psi. i i j j

TABLE 3 - RELAXATION DATA - PERCENT LOAD REMAIVING . Percent preload remaining at 91 days Sample number Standard Diameter Embedment 1 2 3 Average error 1/4 1-1/8 46 41 48 45 3.61 3/8 1-5/8 39 42 22 34 10.79 1/2 2-1/4 43 34 39 39 4.51 5/8 2-3/4 43 41' 38 41 2.52 4 53 47 58 53 5.51 3/4 3-1/4 52 35 53 47 10.12 4 43 60 37 47 11.93 5 42 45 40 42 2.52 7/8 4 44 39 40 41 2.65 5-1/2 64 38 63 66 14.73 1 4-1/2 62 38 70 57 16.65 6 48 40 43 44 4.04 overall Avg. 45% F Distribution Analysis Degrees of Sum of the squares freedom Variance Means = 1491.22 11 135.57 Within groups = 1910.00 24 79.58 Total = 3401.22 35 F ratio = 1.70 F distribution at 95% confidence level = 2.21 Result: Accept the hypothesis that there is no significant difference between groups.. - -.

variation of the means is small relative to the variation within groups, the statistical inference would be that the groups for the test series represent a consistent population. The statistical analysis indicated there was no significant difference among the groups at the S' 95 percent confidence level. Thus, the overall average load remaining 1 S

s. of 45 perceo'. at 91 days appears to be representative of all groups Ja regardless of the test parameter variations.

n A typical plot of the anchor tension developed during installation Y. E and the relaxation of anchor tension immediately following the instal-I s lation is shown in Fig. 5. The figure is for a wedge anchor embedded t ne in concrete. Figure 5 shows that upon removal of the installation I a torque, the anchor tension drops rapidly during the first minute after n d installation. As much as 30 percent of the installation preload is A j seen to be lost during the first 10 minutes af ter the installation. 0 9 A graphical presentation of the data is also provided in Appen-I at dix J for each test group. The test data (load versus time) was e plotted on a semi-log scale. A linear relationship is suggested I U between the anchor load and time relative to the application of the installation torque. Thus, a further reduction of the anchor preload can be expected with additional time, but the rate of reduction in anchor preload should obviously decrease. The data obtained in a three month relaxation test does not permit extrapolation to the design life of a nuclear power plant. Three long-term behaviors are possible. The three possibilities are repre-sented on Fig. 6 as Lines A, B and C. Line A represents linear behavior from which an empirical equation can be derived. Line B suggests that at some point in time the relaxation will stop at an -

.i 1 I e i e i E 9 8 E I E I I 9000-Application of Installation Anchor Type: Wedge ^ Bolt Size: 1/2 in. 8000-Torque (70 ft-lbs) ] Enbedment Depth: 4 in. (8D) l Embedment Material: 3500 psi Concrete 1 3 I o 7000-O. W I 6000-C i O l 1 Minute after Installation i m 5000-10 Minutes after t C, Installation l 4000-uo C 3000-o C 2000-1000-i e e O I 2 3 4 5 6 7 8 9 10 ll 12 Time fro m StOrt Of installation (minutes) i f Fig. 5 - Anchor tension behavior during and inanediately af ter installation

~' ll / / / f: // j ~ / e 3 / o r / E o %so E o o a' / 3 3 / E / s 5 8 A ) 3 m / / / / / i PE01 TET37uI go suacsed - - - _ - _ _

asymptotic preload level. If this data were fitted with a best fit straight line, the predicted future anchor preload would be lower than the actual preload, and therefore the prediction equation would be conservative. T The third type of behavior is that depicted as Line C. Assuming I Ss, that the data is fitted with a linear line, the predicted future Ja Preload would be higher than the actual value. Also, if the trend n a shown in Line C continued, the entire anchor preload would be lost over Y. E 8o** Period of time. 1 5 The experirental data after three months of testing suggested that t ne all three forms of behavior exist. Review of the plots in Appendix J, I a in most cases, supports a hypothesis that high preload leads to a more n d rapid relaxation. Table 4 provides an interpretation of the semi-log A S relaxation plots. Approximately 22 percent of the anchors show Line A s 0 9 behavior. The remainder are equally split between Line B and C 1 at behavior. After three months of relaxation testing, no clear trends e s, are evident which could allow prediction of typical long-term relaxa-I n tion values. In consideration of the value of additional long-term relaxation information, testing was continued for approximately twelve months. e Results After Twelve Months. The final test readings were taken between 352 and 359 days af ter installation. The load cells were then removed from the anchors and recalibrated (see Appendix L). The anchors were checked for plumbness. All anchors met the plumbness criteria except three. Two anchors twisted out of the concrete when the nut was removed. The other was 3.33 degrees out of plumb. These conditions did not appear to affect the other data taken in the test. __

TABLE 4 - CHARACTERIZATION OF SEMI-LOG RELAXATION DATA PLOTS Observed behavior type Diameter Embedment Sample number (in.) (in.) 1 2 3 1/4 1-1/8 B A B 3/8 1-5/8 A B C 1/2 2-1/4 C B B 5/8 2-3/4 A A B 5/8 4 C A B 3/4 3-1/4 A B C 3/4 4 C A A 3/4 5 C C C 7/8 4 C C C f 7/8 5-1/2 B C C 1 4-1/2 B B C 1 6 B B B Legend: A is linear - decreasing B is asymptotically decreasing to load greater than zero C is accelerated decrease in load on semi-log plot (Please refer to Fig. 5 and text for additional discussion.) 4 l. __..._--,.._._._.-_-___.m._

9 Corrections were made to the recorded data to account for the specific Vishay strain measuring instrument used, for zero shift of the load cells, and for differences between initial and final load cell calibration data. These corrections are described in Appendix K, which 97 also contains copies of the laboratory data sheets. 1[ The loads measured in each anchor on specific days after initial 3 teraioning are reported in Appendix M. The averages obtained for each a nn bolt diameter and embedment combination are presented in Table 5 below. ( 2 % Plots showing the percent of preload remaining as a function of time E are in Appendix N. The curves are plotted on a semi-logrithmetic s t scale, showing in general, an exponential relaxation behavior. H r As seen in Table 5, the 3/8 in, diameter anchors lost the greatest od amount of preload (77 percent loss) and the 1 in. diameter anchors at f4-1/2in.embedmentlost the least (49 percent loss). Examination of sj the plats of percent load remaining does not indicate that an ia asymptotic load level has been achieved during the year of testing. es. The relaxation mechanism continues to operate on the anchors. I n We also analyzed each set of data by exponential regression t techniques in order to characterize the data obtained from each group of three anchors. The results are plotted in Fig. 7. The lines plotted in Fig. 7 represent the best fit of the measured data to an exponential equation and might be used for projecting relaxation values c in the absence of additional long-term experimental data. The slopes of the lines are seen to generally increase with anchor diameter and initial load. Higher initial stress appears generally to lead to more rapid relaxation for anchors of a given size. After a year there was no apparent end to the relaxation process, but we cannot._. _ _.

' ^ a I TABLE 5 - AVERAGE RESULTS OF RELAXATION TESTS Relaxation Initial Final % Preload Diameter Embedment Period load load remaining (in.) (in.) (days) (1bs) (1bs) 1/4 1-1/8 352 963 327 34 3/8 1-5/8 350 1753 408 23 1/2 2-1/4 351 3327 1080 32 5/S 7-3/4 359 3689 1198 32 5/8 4 358 4165 2089 50 3/4 3-1/4 355 6532 2440 37 3/4 4 355 6963 2448 35 3/4 5 352 8346 2630 32 7/8 4 356 8954 3051 34 7/8 5-1/2 356 7337 3253 44 1 4-1/2 357 7201 3702 51 1 6 357 8979 3555 40 pr /.. -/ I j. i e 6.

" " * * *'"*~ I ....I i I .A { g P @ b,I s j l hh 4 I l I k k h' h j llhl-I l l . l 'j d l ll' [ lll3 i f 7,{ 4%ll% l b id.ll: !:,t l Ii Jl'p. i, e g; j -y i'T ) ; ' f 'I h p D! f l j f I %j ? i I ,i y, y s.- q ~ h.3 l e b 1 il..a q'l b6 . };i[jUM : i Sooo t. 7lp.t} I i e l t - lj,, p l l,! l [ N t LEGEND l ~ g i ll C .. O.;!!M M L y jM b i N (..g, '. g h, h Id * *** * " * *"d"*"* ) { l

t d

>'l,,i t 3 3 . l t < b t l' J .I -,. l i id h il! l p t A - 1 in. x 4 1/2 in.

f. Ih,,h,.;

p s q*'i ?f j ll l I i l ] ',}l:- b )a 'r I i .l $p a - 1 in. x 6 in. v i g flt Y 1 '};d } 5 J I W'I@' Y . N I M C - 3/4 1"

• 3 II' 8"*

4000 'l 4

\\ )'

I h. -{ 'l .:';;:-D i i i l l: 1 h h(I 'l D - 7/s in. x 5 1/2 in. i i' E D i l. l {J N ! a;'kll 'l l J 'E.I l c "i - 'k

i~ '.

i O

{l!.3 S a l'l E - 3/4 in. x 5 in. k_:.;l;l.;:!..i 11 j l ! d' E I I I'l t \\ t i :

-i;

- ~ r - 7/s in. x 4 in.

s .+

p lR v I bl ! ; ;

.,, p ~ y !ll, ;;,j , c - 5/s in. x 4 in. "l r - :Ir t 4 E 5 :i I.! !,..si;.[ L y0 , yh ), : u - 5/s in. x 2 3/4 in, l t; : f,:S i N t d - i:::M.'.:l,'$ ; J 3 Dh I];:l'yp 5 . N'j$..;'4:gJ oo

l N / 't."4.I - 3/4 in. x 4 in.

]k r N 300o 9;h 5 1; MN j. i T ;li l 'll! I - 1/2 in. x 2 1/4 in. .lIl 5 I?l. !!; i:. I' ~ kl. f.K-3/ sin.x15/ sin. l i i 1 Tf' l:.y 4 It . g'::g 5 s i ,4 N i ';:J . I'.{; i 1.,l..;.,7 Q i h i.: .i.;. $fl l it.. Q'li,:.n ! . 5.L .t i.e (c p L - 1/4 in. x 1 1/s in. s j } F ih.s.p-c I i:j !. o [I' ~ j

i p g iil,.~

.f M, i r }g l ll 1 ;; Ili il i ill. NI h a r

• l

,ll -t r7 M p-l - r - I-ij !y4 f d

l'l;-t:.

R f i i;,i j j,Q:. .:,i - ] l l.;j

y

.;l sp ) j; r -i f; f f li [';i ' -1 2000 i-f l l

11 llif

' l I T [di [i i J [ } ~~ d I h II l I" I 1000 ' 1 't i ll l l - Q i: b ) !h,i'a f l $ i 'CG ,,li f - {'4' h

'i i

Md-- I 1 g tl h I T 51 2 a I N' T b

l 2

l' f l l l,$;;.

) ',l.

1 dli illI ii:ill C d E I L. a N l ib. .i, 12 -ll.1I lN; 1 E S 0 .1 10 100 1000 10,000 Time af ter initial tensioning (Jays) i l l Fig. 7 - Result of exponential regression analysis l 1

conceive of a physical mechanism under constant temperature conditions which is likely to cause the rate of relaxation to increase with time af ter a year, nor the load on the anchor to reach zero as a limit. In service, where vibrations may be superimposed on the static relaxation W l 1 conditions, a zero load condition might be achieved. This hypothesis s 5, was not tested in these experiments. n Statistical analysis of variance showed no signifAcant difference n y, in the percent of preload remaining which could be attributable to f anchor diameter or embedment depth. Thus, the data for all anchor s t tests may be combined to characterize the behavior of all anchors in n e I general. A best fit regression line was computed to relate the percent a preload remaining to the time since the initial loading of the anchor. n d g Fig. 8 shows the result. On average, the anchors retained approxi-s j mately 40 percant'of their preload after one year. Extrapolating the C i behavior line to 40 years indicates that, on average, about 20 percent a te of preload would be remaining. However, the results of the testing s. I were highly variable. The standard deviation (c) was 9.23 percent load n ' remaining. Assuming normal probability distribution, all but a tiny fraction of the anchors would fall within limits set by three standard deviations from the mean. The three o limits are plotted on Fig. 8. From this plot we see that a small fraction of the anchors could be expected to lose their preload in the period between 10 and 40 years. As mentioned in the preceding paragraph, we suspect there may be a lower limit to the loss of preload, but the one year testing period did not provide an indication of what that limit might be. I am-m" --e -w .-,y y- -,+--yv --em--,-w y- -ww----v --wywm--v,* ww,- w w

A .m.lm. e / / / / / k / / / / / / / ... / g = = =...

• =.....

o ~ ~. sw., Sd

• ** *.. *. m* g C

.e .c GB U C / m g 4 4 en w l . g >- O e a g N 10 Z A e H e S LL.l LD c .st Cw g p 00 0 CO wg I f 8 8 a e s =* 9NINIVH38 OVO1 IN3383d i 1 e 1 I i 1 0 1 I I I - --,----,.. -,,.,.,-- ----,~-,,.

t

SUMMARY

AND CONCLUSIONS Thirty-six wedge type expansion anchors were tested in 12 groups of three to measure loss of preload during a one year testing period. When all anchors are taken together, an average loss of 63 percent of g7 f initial preload was found. The individual anchor losses ranged from J 45 percant to 88 percent. Statistical analysis of variance indicates a $ no significant difference in the behavior of the groups with respect to e Y' percentage load lost through relaxation. E 1 When data from all anchors was lumped together, extrapolation of S tn the line obtained by regression analysis indicates that, on average, er 20 percent of the preload will remain after 40 years. a l The shape of curves obtained by plotting preload remaining versus A time indicates that relaxation will cont.nue beyond the first year. S So Extrapolation of the test data to longer periods is difficult consider-C ia ing the scatter of the data. Thus, accurate prediction of the ultimate t !,lossinpreloadforindividual anchors, and when it will occur cannot fbemadebasedonthedataavailable. C Respectfully submitted, WISS, JANNEY, ELSTNER AND ASSOCt_'ES, INC. o l i F. D. Heidbrink l Project Engineer i J. R. C. Miller l Project Manager Reviewed by: [ ( Richard G. Lindstrom Quality Assurance Manager..

Serdal 2068h i i CONCRETE EXPANSION ANCHOR RELAXATION TESTS I I w f BECHTEL POWER CORPORATION s. BECHTEL NO. 7220 J aj WJE NO. 80804Q e Y' ADDENDUM 'E [ June 11, 1982 s 1 D \\ d \\ f contains the Test Report 5 only. O f Appendices O through T to this Test a Report are available at the Consumers [ Power Offices in Jackson, Michigan. S. (NOTE: Page renumbering at the bottom f of this report was done by CPCo t. microfilming dept. The first two pages of the report were merely identification pages used for microfilming purposes.) ^ WISS, JANNEY, ELSTNER AND ASSOCIATES, IFC. 330 Pfingsten Road Northbrook, Illinois 60062 (312) 272-7400 1141 7NO'0//5 N' A

r CONCRETE EXPANSION ANCHOR RELAXATION TESTS FOR ) BECHTEL POWER CORPORATION E BECHTEL NO. 7220 1 5 5. WJE NO. 80804Q J a ADDENDUM n n e June 11, 1982 Y. O E 1 INTRODUCTION 7 s t 7 n e Wiss, Janney, Elstner and Associates, Inc. (WJE) was retained by I a Bechtel Power Corporation to perform a series of static tests on concrete n d expansion anchors. The anchors are intended for use at the Consumers A 5 Power Company Midland Nuclear Power Station Units 1 and 2. The test 0 C series was performed in order to quantify the structural behavior 1 a t characteristics of the anchors. All tests were performed at the WJE e S* 1aboratory in Northbrook, Illinois. I The results of a twelve month re ?.axation test on 36 expansion anchors using load cell instrumentation were reported in " Concrete Expansion Anchor Relaxation Tests for Bechtel Power Corporation, Bechtel No.,7220, WJE No. 80804Q", dated May 28, 1982. This report is an ' addendum to the May 28 report. It contains the results of additional preload relaxation tests conducted on concrete expansion anchors. The anchors in this test were instrumented using precision strain gages. The test period was August 26, 1981 to March 26, 1982. 1141 924#- d //0 3W j 6 */

"'W mv= >e n, o o_,, 't Preload is established in expansion anchors by tightening a nut on the anchor bolt to a specified torque level. Preload is a tensile force imposed on the anchor bolt by the installation torque. The decay of the j installation preload with time is termed relaxation. In these tests, f af ter the initial tension level in the bolt was recorded, the load was s. monitored periodically over a seven month interval to record the y a n decrease in load. n e Yo REFERENCE DOCUMENTS E 1 S f The appendices associated with the May 28 report are referenced in e I this addendum. The appendices are contained in two volumes. Appendices a n A-J are in one volume dated July 2, 1981. Appendices K-N are in the d A second, dated April 9, 1982. The appendices associated with this s addendum are lettered 0 - T. These appendices are contained in a third Cj volume dated April 15, 1982. ~ t { The tests on the concrete expansion anchors were performed in I accordance with the following specifications and procedures: n l C. Consumers Power Company Specification 7220-C-115(Q), Revi-sion 1, " Technical Specification for Relaxatien Tests of Expansion (Stud-Type) Concrete Anchors" Consumers Power Company Specification 7220-C-305(Q), Revi-Technical Specification for Design, Furnishing, sion 13, Installation and Testing of Expansion Type Cencrete Anchors for the Consumers Power Company Midland Plant,-Units 1 and 2, Midland, Michigan." Consumers Power Company Specification 7220-G-23, Revision 8, " General Requirements for Supplier Quality Assurance Programs for the Midland Plant Units 1 and 2 for Consumers Power Com-pany." l Wiss, Janney, Elstner and Associates, Inc. Procedures Manual OP-29, Revision 1, " Procedures Manual for Static, Torque-Ten-sion, and Relaxation Testing of Expansion Anchors Embedded in Concrete for Bechtel Power Corporation"

1. ypp p c yr-33*1 Y

i P35

The Consumers Power Company Specifications 7220-C-115(Q)-Revi-sion 1, 7220-C-305(Q)-Revision 13, and 7220-G-23-Revision 8 are enclosed in Appendices A, B and C, respectively. These specifications are here-af ter referred to as Specifications C-115, C-305 and G-23, respectively, or simply as the Specifications. 3 S. 3 The WJE Procedures Manual OP-29 is enclosed in Appendix D. This a D operating procedure provides an outline of the steps to be taken in e Y. accumulating the physical test data for the test program. E I A subsequent Procedures Manaal, OP-32, was written to cover the S [ testing of strain gaged expansion anchors. This manual is provided in e I Appendix 0. a n d SCOPE A S S 0 t The expansion anchors which were tested are generically identified I a t as drilled-in expansion anchors. All anchors used in this phase of the e s* overall test program were manufactured by Hilti, Inc. of Tulsa, I E Oklahoma. Expansion anchors are sold under a variety of trade names, each being associated with a different manufacturer. The anchors achieve load carrying capacity through a wedge mechanism located at the bottom of the anchor. The wedge is expanded against the side wall of the embedment hole by tightening a nut which tends to pull the anchor from the hole. When th'e wedge jams against the side of the hole, further tightening of the nut establishes a tensile load on the anchor bolt. The Hilti wedge anchor system uses different steels in the bolt shank and in the wedge. The bolt shank is made of AISI 11L41 or 1144 Y =f p 3.0

• C,//S* 33-2=

6'

steel meeting the chemical requirements of ASTM A108. The two inde-pendent expansion wedges near the bottom of the anchor are made of AISI 1050 spring steel. The nuts supplied with each anchor meet the require-ments of ASTM A307 Grade A. The bolt shank material has an approximate i minimum yield and tensile strength of 95,000 and 105,000 pai, respec-S S, j tively. aj All testing was performed in accordance with the Consumers Power e Y. Company Specification C-115. This specification was provided by Bechtel E 1 Power Corporation. S [ All anchors were installed in accordance with the anchor manufac-e 7 turer's recommended procedures and Specification C-305. All testing I a } conditions were in accordance with the Specifications. A A total of 16 expansion anchors were tested in this program. S Table 1 gives the specific groups for each test configuration, which Cj included dif ferent anchor diameters, embedment depths and installation i t[ torques. T I D c. IDENTIFICATION SCHEME FOR INDIVIDUAL TESTS ~~ All test samples were coded by a unique test mark which identified all the characteristics indicative of the test. The test mark was l ' composed of an eight part alphanumeric name. The name parts were separ-ated by dashes and identify: (a) type of experiment, (b) specimen name, (c) expansion anchor type and bolt diameter, (d) depth of anchor embed-ment, (e) installation torque, (f) test number in the test series, and l (g) special consnents pertaining to the test. / 1141g y y g jf fa & Q 67 l

=

TABLE 1 - HILTI WEDGE ANCHOR RELAXATION TESTS Depth of Anchor Instrument Embedment Installation Diameter Channel Before Torque Torque (in.) No. (in.) ft-lb Remarks S 1/2 1 2-1/4 55 Long-term test 5. j 1/2 2 2-1/4 55 Long-term test a g 1/2 3 2-1/4 55 Long-term test e Y. 1/2 14 2-1/4 55 Long-term test - E 1 1/2 15 2-1/4 55 Long-term test S [ 1/2 16 2-1/4 55 Long-term test 1/2 17 2-1/4 55 Load cell (1) aj 5/8 7 2-3/4 160 Long-term test A 5/8 8 2-3/4 160 Long-term test S 5/8 9 2-3/4 160 Long-term test cj 5/8 10 2-3/4 160 control ((} Zero sh ~ t i* control ((} 5/8 11 2-3/4 160 Zero sh I n c. 5/8 12 2-3/4 160 Zerosh((} control 3/4 4 3-1/4 280 Long-term test i 3/4 5 3-1/4 270 Long-term test J/4 6 3-1/4 270 Long-term test i Notes: (1) This anchor was also tested with a load cell to obtain a comparison of strain gage versus load cell data. (2).These anchors were used in a control experiment to attempt to isolate strain gage drift. They were loaded and unloaded periodically and did not contribute long-term relaxation data. g_ 1141 7 p p,g,f/f-J3-J b w ---,---y-w---w-wwr y-g=pq -e r-.--w---w- --g--,-m-,

---

ww

The WJE manual OP-32, which is enclosed in Appendix 0, gives addi-tional details regarding the identification scheme. An example of the identification code is given below. W Example: R -C1/5-u/WC1/2)-2 1/4-55-1-SG f 1 2 3 4 56 7 S. J a 1. Relaxation test. n E 2. Concrete Specimen No. 5. Y. E 3. Hilti: Wedge anchor with 1/2 in. bolt diameter. 1 5 4. An embedment depth of 2-1/4 in before torquing. t n e 5. An installation torque of 55 ft-lbs. I 6. Test No. 1 of the test series. a n d 7. The test procedure used strain gages to measure load. A S S 0 C TEST SPECIMEN MATERIALS i a I Concrete test specimens and expansion anchors were supplied by l Bechtel Power Corporation and delivered to, the WJE Lab in Northbrook, ~ n c, Illinois. Certificates of Conformance with the material requirements of the Specifications were furnished by Bechtel to WJE. Concrete slabs were stored inside the laboratory under ambient laboratory conditions. Expansion anchors were segregated and stored in the laboratory to pre-clude unintentional use of the anchors or substitution of other anchors. Structural plates and shapes used as parts of testing assemblies met the requirements of ASTM A 36, Specification for Structural Steel. j } 69

== il Bechtel Power Corporation provided 3 ft x 3 ft x 1 ft concrete slabs with nominal 4000 psi compressive strength. Test anchors were installed in three slabs in accordance with spacing criteria of the i specifications. i Compressive strength' data, obtained from 6 in. x 12 in. cylinders, s S, j were also received from Bechtel Corp. An evaluation of the data indi-a D cates that the concrete had a 90 day compressive strength of approxi-E Y. mately 7000 psi. The concrete test data and compressive strength versus E I age graphs are enclosed in Appendix E. S 1 n er SPECIMEN TRANSPORTATION AND LOCATIONS OF EXPERIMENTS a n d criteria for Transportation A S S o Expansion anchors were to be installed in concrete slabs containing C f no significant defects such as cracks, honeycombs or internal coids. t[ Specimen stresses during transport were to be checked to ensure that-f maximum stresses were below the point at which cracking could occ'ur. The t. slabs were received in good condition with no cracking observed. No '~ inspection was performed by WJE to determine if the slabs were free from internal defects, but the surface appearance of the slabs indicated the concrete was dense and well consolidated. Lifting Inserts for Specimen Handling Lifting inserts were cast into the specimen to facilitate trans-portation and handling. The lif ting inserts were positioned in each test specimen so that they caused minimum restraint on the testing of anchors embedded into the test specimen face. 1141 yl2.2 0- C //S' bb'S bb0 -ar----- r

-e. ...m i Location of Anchors Single anchors were embedded into concrete specimens. Minimum center-to-center and edge spacings listed in the Specifications were W followed in locating the anchors. i s. In all test specimens, the anchors were located to avoid the effects f of embedded lif ting inserts in the concrete specimens. n n f, EXPERIMENT PREPARATION E 1 s All anchors were installed using the manufacturer's recommended t E installation procedure. These procedures are included in Appendix F. r n Expansion Anchor Installation d A s Drilling of embedment holes - Test expansion anchor embedment hole s 0 locations were marked on the test specimen prior to drilling each hole. 1 a tioles were spaced to meet the requirements of ASTM E 488 and the Specifi-e s-cations. The embedment holes were located and drilled to minimize the I n effects of restraint of the lif ting inserts. c. Holes for the test expansion anchors were drilled af ter the con-crete had cured for a minimum of 45 days. The diameter of the expansion anchor hole was in accordance with the manufacturer's recommended pro-cedure. Each hole was drilled with a carbide tipred 5it, meeting the requirements of ANSI B94.12 (1968). The reconsnended diameter drill bit was installed in a roto han:ier drill corresponding to the recommended drill for the size and diameter of the expansion anchor to be installed. Each time the drill bit was used, its use was recorded. When the bit had been used 100 times, the bit was discarded. The depth of the anchor hole was drilled deeper than the specified embedment depth. 1141 7.220-f//5-33-3 ?j l/

Anchor installation - Before each anchor was placed into the embed-ment hole, loose dust resulting from the drilling operation was removed by air pressure. Each anchor was inserted into the cleaned hole and driven into the hole to the specified embedment depth. f The anchors require rotation of a nut (turned element) on the anchor S. stud to expand the wedging mechanism against the concrete. A torque in y a n accordance with the Specifications was applied. The application of this n e Y. torque was termed the installation torque. E I s Measured installation characteristics - During anchor installation, i n e measurements relating to the conditions of the installation were I recorded on data sheets. The data sheets are found in Appendix P. a n d Recorded information regarding installation included verification A that the manufacturer's equipment and recommended installation procedure s s 0 t were followed. i a t e 3' Anchor Plumbness Check I n t. All anchors were inserted in their holes with an out-of-plumbness l l of less than three degrees. A sketch of the instrument to measure the out-of-plumbness angularity is shown in Fig. 1. The out-of-plumbness was measured af ter the tests were completed to avoid damage to the strain i i gages. The objective of the instrument is to measure the angularity of the i l expansion anchor with respect to a flat surface representing the surface l 1 of the embedding material. A flat surface representing the surface of the cone' rete slab is obtained by a smooth steel plate and is referred to f 1141 f l3,?O-C//5'.33 *O /3 /.2

• O 8

-Plu6 ness Angle (c) Dial Cage Q Expansion Anchor Scud N N N N N ~ N p ' Turning Plate y N N Znstru=ene j Constant Perpendicular Plate l ( Reference Plate I I l I Embedding Material l _1 !!a::isum Dial Cage Reading - Minimum Dial Gage Reading 'I a = can 2 (Ins tru=ent Cons tant) i Fig. 1 Ins trurent to : easure Out-of-Plu=bness 1141 YM MM M eM!f' hlS . - _ -. _ =, _.._ L

as the reference plate. A perpendicular plate was attached to the threaded portion of the anchor and established the position with respect to angularity of the installed. anchor. A turning plate, as shown in Fig.1, was rotated through 360 degrees while bearing against the per-f pendicular plate. The dial gage attached to the edge of the turning 5. plate measures the distance changes between the reference plate and y a g perpendicular plate as the turning plate is rotated and the stem of the e y, dial gage slides on the reference, plate. E ~) 1 The out-of-plumbness of the anchor can be determined from trigon-5 [ ometry knowing the minimum and maximum dial gage reading and the distance e ~ r from the center of the anchor to the stem of the dial gage. a n This instrument provides a relative measurement of the out-of-d A plumbness. Possible sources of error are (1) the dial gage is not s mounted perpendicular to the turning plate, (2) there is a gap between Cj the anchor stud and the hole in the turning plate where the anchor stud t { goes through it, (3) the perpendicular plate may not be exactly per-I pendicular to the bolt axis because of machining inaccuracies, and (4) n C. high and low spots on the embedding material caused the reference plate to be unrepresentative of the reference surface. Considering these possible sources of error and that maximum and minimum dial gage readings are used in calculating the out-of-plumbness angle, it is reasonable to assume that if the calculated out-of-plumb-ness is less than 3 degrees, the anchor itself probably has a true out-i of-plumbness with respect to the embedding surface of less than l 3 degrees. l l l eM" 1141 7yp g//g> 3 3 - S l i .-...__.__.,_.m._.

Embedment Materials and Installation Requirements Each anchor was embedded in accordance with the manufacturer's instructions and according to the procedures outlined earlier. Each Y test anchor was an unused new anchor, instrumented with two strain gages. I Af ter initial torquing, no additional applied load was imposed for the s s, J duration of the relaxation test. A total of thirteen anchors were tested a a for long-term relaxation. Variations between groups included the bolt n* e Y. E size, embedment depth and installation torque. 1 f Instrumentation e r The instrumentation used in performing the relaxation test included a D ~ d torque wrenches (2), a 10,000 lb load cell, strain gages and a bridge A s type strain indicator. Calibration certificates for these instruments 0 and the Baldwin hydfaulic test machine used to calibrate the load cell t I a t are enclosed in Appe' dix Q. e Direct reading :Olibrated torque wrenches manufactured by Van F. I Belknap of Detroit, Michigan, were used to measure the installation H torque. The torque wrench for the anchor sizes 5/8 in, diameter or less had a range of 0-175 ft lb. The torque wrench for the 3/4 in. diameter anchors had a range of 0-600 ft Ib. A load cell was used to measure the tension in one of the 1/2 in. diameter strain gag'ed anchor bolts. The load cell was manufactured by Sensotec, Inc. of Columbus, Ohio. A Model D, 10,000 lb capacity load cell was used. The Model D load cells are cons tructed with foil type s tr ain gages wired into four active arms in a Wheatstone Bridg* configuration. The strain gages are bonded to the sensing element of the y 30 -C /W 1141 6/5 _,,---..,y.,, .,..,-..,,9

• w,

-.__g 9 y., e-.,

load cell. The strain gages are provided with temperature compensating circuits which offset the change in resistance of the bridge due to temperature, and compensate for the change in modulus of elasticity of the strain gages. The load cells are rated as follows: i S s. Output: 2 av/v J a Bridge resistance: 350 ohms n E Linearity and hysteresis: .5% to.6% y. E Repeatability: .1% 1 N S Temperature effect on: t n e bridge resistance .005% full scale / degrees F a modulus of elasticity .01% reading / degrees F n d compensated temperature range: 60 degrees F to 160 degrees F A s Prior to using the load cell it was calibrated in a Baldwin C [ hydraulic test machine following a standard calibration procedure. The t ( results of these calibrations are provided in Appendix Q. The calibra-I tion procedure is in Appendix I. n C. Strain gages were used to measure load on all test anchors. Two gages were installed on opposite sides of the anchor shaf t to compensate for the effects of bending on the anchor. As shown in Fig. 2, a rigid 1 plate support was installed over the test anchor. The bottom plate was 11 in. long by 4 in. wide and either 1/2 in. or 3/4 in. thick. An aversized hole was drilled through the bottom plate to prevent damaging the strain gages during installation and testing. A slot was cut out at the other end of the plate for installation of another anchor to prevent rotation of the plate assembly. As shown in Fig. 2, two filler plates 1141 7.2J a cNP M 'dh/6

.'s. 4 z ..g ...,...g. ....e.-s. .y ) 73 s p. .a,;.... g;f . x.y M.,..

,y y _.,.

, a..... '..v,,..... .. -.g. y

y ;,.y p.

' ~ i;g -. Jv. j,<- . g }.,,,. .3 - e ; .c. t '..r.<, c%1,.' ' '. '.. ' ' f*$, 'x.y't ? ' '.. :., f.s. ~ y,, [,] - N*> ; '. lk., ,M+kp!g ;;)._Q;.n..s \\ 1 m. y-a.

.u_ :

r,,,3

.v 4- . ^ _g' s. ' s. - [, _ l:'} ,.. 3 _ .3....:- . y.. :.a.c ~. - .. + w e.v;.s 3..p.

c..

a..p, 4 s f. p

1. .g'.J

• P.,,,, .. ~'.. W. );,. j.

e .<3 w:xct. ~,. ' '.. :.,>s.. ).. ' ' % _. f %., -. y c ,y, g).; ;?; -( .u:t. ._e. -.h&c- '. '..,';y.,' y. p. .,...'..,e,

a. -

• 9

,. b sa '. ' ,'j! '. '"A- / .,,..y ; l gi e . f. '.

v.,,.g..'. c.,,.. <..,

,.: 7., s, .*A c. s. 7 ,k..[ E}. h-i. ..NE 's. )% eM [ ., 'n! %'it ?. ' * :. ', ~. ?: ~- .+4yj.- ...; j;.Q ~ :.- ?>..%,.y, f fat - AGLg

.. y ;

g a5.:- ,,# +.. j ' '. 'l.I .v.f.cf.f +hk&.\\Ym'

1. ?^.: Wl.Wl '. L.l _ -' l. l.C.

....L*' [.Q,A ~ ^~ . N nk . e' I.

g

,

. TJ.QfiMQlt.1.N:QM y>y .I,

. '-?

'y 1l: #g.,:, : s [ _4:.f. O X 6 241 ' %;ffji,y ~

. :. :. 3. L, V., _

.L M '. O r . D d d h M W Q 8).9. %e A. wak

w; 3 wm F ~

h.2$. gn% ig[&v.)p.h.. . ?."9; % Q w;.hy ;;%a ;m;.~ f n. - L. Wy[yh.h NN Nk '.J . f.. - E :.[ 'N.E.d. E l ~q T.

Q:n%m%.: m4 pw$w%y j Q [;% l,.i;;'. m.. : ; 2..,.

..... % m :..y-J ', y wM%g#@h_: m m n L. y a

A

m lq.y.::9} }.g; %. ;i

. s.

9 T fh i AaMQ&}..N..g%.f$$4.. w, Q, y q. '. j9ex%mw .f. ;.> %%.y .... q.i e. 4 f. __' ~.s \\~ g-Qg .Y p..... ..ghsM. i. .,. ~ khqb igth963h.N [MM Ak l

• j as.gg wu y,

.= l -2 ! -I [g - .MNNhg%%4%;}p!f.kNUh j)[.[ [, b}*' h '.- i h

e.,

. *s -f, D .. ~* u n.', I,p . w4 \\g ,T.. ;:.'_ y .,..<,.7..,, -.y l N .3 5' g* l\\.'7 l \\. i .c g 'ty. ; ~ e y ? e, w.,. - '..r., w > : - ~ ... I.lW.j e. 1 5.', 'h','. M,' ' '. <". ". Y '.'?- ' ' a.. s, $[c, l N .$g. 3.g [$8' .y. ~. T ' .f .7 .u~ j Q

.; s

.: ;s .s . : :M.1. r k Fig. 2 - Attachment Plate Simulator i 9 L

__._.._ _._ ~ 9 were welded to the bottom plate to allow room for the strain gages. The top plate distributed the load from the nut and washer and also simulated an attachment plate. The plate was at least as thick as the diameter of the bolt. The hole size in the plate was drilled nominally 1/16 in. greater than the anchor diameter. Strain gages were installed in f accordance with WJE Procedures Manual OP-10s, " Procedures Manual for s. Installation of Strain Cages on Metallic Surfaces", provided in Appendix y a g I, and ?P-32. Calibration data were obtained by applying strain gages to e y, 1/2 in., 5/8 in. and 3/4 in, diameter anchor bolts and testing the E I anchors in a Baldwin hydraulic test machine. The average of these s [ calibration tests was used. The results cf these calibrations are e I contained in Appendix Q. a n A bridge type strain indicator, Model P-350A, manufactured by d A Vishay Instruments of.Raleigh, North Carolina, was used to read load cell s and strain gage output.. This instrument has an accuracy of.01 percent Cj or 5 microstrain, whichever is greater. The analog display can bm read t { to the nearest 1 microstrain. I n c. TESTING Test Setup A photograph of the test setup was shown in Fig. 2. In all sixteen I anchors were tested. Three anchors were used in an experiment to check strain gage drift. These anchors will be discussed separately at the end of this section. The remaining thirteen anchors were torqued initially during installation and then left undisturbed during the test. Measurement of the load condition existing in each anchor was made on

• /3 3.0*C N V N 'S

=

installation, and at 1 minute, 2 minutes,10 minutes, 30 minutes, I hour, 2 hours, 4 hours, 8 hours, 1 day, 7 days, 14 days, 28 days and approximately two months, three months, and seven months. All readings are referenced to the time the installation torque was applied. No ret rquing or loosening of the anchor nut was allowed during the test W i s Period. All data were manually recorded on sample data sheets found in s. J Appendix P. a The three 1/2 in diameter anchors used to check strain gage drift, e Y' were periedically unloaded by unserewing the anchor nut. The load was o E I measured before the nut was loosened. Then a zero load reading was made. s N l, f The nut was tightened to reestablish the load. The anchors were restored e r to the load level measured before the nut was loosened,'or were reloaded l a j by applying a 160 f t Ib torque (the installation torque). A 8 Test Results O C f Strain Gage Drift Experiments t e s. The primary purpose of tests on the three 1/2 in. diameter anchors ~ I was to determine if the strain gages would exhibit stable behavior over o c, the test period. A shift in the zero reading might occur through I electrical deterioration of the gage or mechanical change in gage length, perhaps caused by the characteristics of the adhesive holding the strain gages to the anchor. The experiment was conducted by periodically loosening the nut to restore the anchor to zero load. Strain readings were taken before and after this operation. Then the load was reapplied. The results of these operations on the anchor are provided in tables in Appendix R. Figure 3 anW-YA/ 2 2 0-C,// [" d 3 0 f 19 3

5 8 . u.e 2 Sa*T oc tant s:s R Eda! "~" . :,ca 8l2. s W,TV: :: Q b2f' g

ee-i a=

-.. t r. - 4 5 c? s t, - g .x .x g . x j. j .. + .x Oo s = en u 0, <Q w x+ Z 8 ~ 4 tu w x + M x. 21 0 4 x + x. . g z x + ,g 2 - x 8 8 e DNIOV38 1VNM 30 5 SV.idlHS OB3Z 72.2 o C //f" 33.L 11M l EOff IEI I

Presents the results showing the zero strain readings expressed as a Percent of the final reading taken some 205 days af ter the initiation of the experiment. Note that two of the anchors showed zero shifts of no greater than eight percent. The third anchor was quite unstable, with l the final zero reading some 2.25 times greater than the initial reading. 5. The average zero shif t of the two stable systems was.99 of the initial y a g value in one case and 1.05 in the other. We conclude that if the results e Y. of the strain gage relaxation measurements appear to be stable that zero E I shift should cause an error of no grezter than eight percent on any 5 [ individual reading. e ~ A secondary purpose of the experiment was to examine the effect of I a g restoring load to anchor bolts by reapplying torque to original levels. A Discarding the data from the anchor exhibiting unstable zero shif t we S find that the loads imposed on the remaining two anchor bolts by C { reimposing 160 f t 1bs of torque varied widely from the initial load. In t { the anchor which was restressed eleven times at the 160 ft Ib torque I level, the reinstated load varied from 1282 lbs greater than the initial n C. Ioad, to 430 lbs less. In the case of the other anchor which was restressed three times at 160 ft 1bs of torque, the load varied from 575 lbs greater to 2112 lbs less. Clearly measurement of torque is not a highly accurate means of measuring the load in these 1/2 in, diameter anchors. Relaxation Experiment The final test readings were taken approximately seven months after i installation. The anchors were unloaded by removing the nuts and zero W NO-C /W' Q J P9/

load readings were made. Results of the control experiment indicated problem with strain gaged anchors. To that zero shift could be a evaluate the effects of zero shift corrections on the data recorded during the experiment, we examined a series of data plots. The zero W shift correction consisted of applying the measured zero shift in i s s, proportion to time with the maximum correction made to the last data J In the case of runs 1-9 we also took a second reading some 2 hours a point. n D af ter the anchors were unloaded. In some cases further zero shift was Y. E ree rded. This phenomenon is attributed to deviate strain gage 1 s behavior. The several data plots are contained in Appendix S. The y n e values used to plot the curves are coatained in Appendix T. On review of I this data we concluded that none of the anchors suffered from gross a n d strain gage failure and that the results could be averaged to approximate A typical behavior of concrete anchors undergoing static relaxation. The 5 s O plots based on the original data with immediate zero shift correction are c 1 f considered most representative of the physical behavior of the anchor e s' bolt system. These plots are presented in Fig. 4 through Fig. 8 on the I D following pages. Figure 4 for 1/2 in, diameter anchors shows excellent correlation of results. Af ter twenty-eight days all anchors seem to have stabilized with about 38% preload remaining. This is the only group of anchors we

tested, including the 36 load cell instrinnented anchors, which approached an asymptotic level of load during the test.

The average initial load on this group of three anchors was only 2934 lbs compared to 3317 lbs for the anchors whose results are plotted in Fig. 7. The average initial load of the 1/2 in. diameter anchors used in the load cell relaxation tests was 3327 lbs This might indicate that low 1141 MN &M

,F ~ 0004 1 05 . 03 e 5 m 03 iT in su 1 s 1+ . 9 r f e V 0 g 1+ . 2 n n i o 4 n i 1 i t l ce" " w 7 i+ r r e o S R c Y ^ t d A a f i " D o h" h t s l N e g g n o . 1 r e"" I P e r r et z us q h E t rl t n +u oa i" " G e Tn n w A c i u nm R 123 r +I oo e i n t P a - + 'x l l e d R e at a U g t e o O 4 +ex dt sr L) . H ennc 977 s WeI n l 549 b 385 1 g m o a ,,X idb c f 1 322 t el t( i tb i i F i mt o n N HEf p I I . M + 8x n .i 0 n l 1 i4 o / 0 b 21 0 m x*+ / 50 y 1254 S + ax _a X o D 0 m 0 0 4 0 0 1 DZHZHtEJC OCOg gzjJ.CJ - l g. t e 1 (t LC L L De f k%% Ww sD df Nh Gb(y yMC^ ~ T i.i' 4

r m 8 E E . = .3 + ux s + x + M x tic + ax h C e w + .x a - 1 e e != = _ s

• =

a m e e o Q g~ a. x b x +E &g =

• E

8. -

tu u 3" de : C.D x += "S 5 = <a= m gg x

=

s' x sm ? 5 .4" 3 - o scht 2 $2R 3 x m mdoB 3$ N x t = s 3et a = x x a h b M m4 O

Tal 1

"+ mena m x 8., 8 8 2 9NINIVH38 OV01 IN3383d 1141 yp p o c // r-33 .2-5Z000: IP I I DO -4 -n w

$2 S . 8 8 g s x. 0 G s + x = + x a g C + x u y C g + x a + x u ga s M 5 = m 2 e s== a g-m s 5

g..

+ x = u E-5 LIJ u O T** AE = 3

j

.3' e + x a b 5 34 3 - E +

• =

Tune ..= s ;0

s

= =t-e =

• =-

+ x = tigu C'5L2 1 x = =- + x m . x O _5 " e?s@ l " * + en-, e +x M I e g g e 8 8 8 4 2 9NINIVW3B DV01 IN3DB3d 1 M 1141 y g p o -C//S~# 3}.3 025 9 E O (; f r r i p l l = =..

00 '0 4 1 05 ' 03 58 e 3 m i E T s 1 + ' 9 usr eV 8 + ' 2 g no n 4 + E i ' 1 i tc n e"" i r E ' 7 a r m o S e c Y R t f A i " d h" D a h s t o g o N l iE n r ' 1 r '- e z" " I e e r et P us h E q t t rl i" " G x* a oa w n Tn n A e i u 455 x+* nm R 111 c oo r i n e ta-P

• A l

R l e d e at a U g t e o O

• t dt sr L)

' H 7 ennc 317 s WeI n l 987 b m o a 061 i A idbc i (1 333 g t el t lb i i i at o n N F i HFf p I I n ' M x a .i 0 n l 1 i4 o / 0 b 21 0 m x*+ / 50 y 1254 S x 7 m* 0 0 0 0 0 0 0 8 8 4 2 1 yaC O eZHZHtZL" OJ J 1 hb D 9f\\\\kdw yb ag. y, b@ N N C m~ m 9 8 1 l I Jj j

I. 008 I 4 1 0 a 583 58 e e 3 m iT g no i n s t o e 1 u ci 8 s N et r rc ex re e or V cro 8 R t c a 2 g f n i t 8 hf a 4 i x sl 1 n h i oa a r ,x eo e 7 m zre S e hz R Y tih A d wt D a i h ew r o t g o g al h l N e I n sl c a 1 e e n I r t nc a et i P ua ad e q ra m E t rl t o a I oa sl s G n Tn n A e i u 77 n c nm m 11 o oo r i n l e t l P a-e 8 l c l e d R e at a d U g t e o a e H O 8 o M dt sr l) l ennc WeI n l s m o ab d i u tdbc i n g t t el t( a i ib i i F t mt a n s N t Ef p I e I g I n g s M a .i 0 n l n 1 i45 o i /50 b a 21 0 m xa r /- 0 y t 12 4 s S xB xs 0 0 0 0 0 Q 0 8 8 4 2 1 OC 3J_< C. HZCya eZHZs4b C s W NgN I tam OnN ~ N O,, cc i: I iL j o

I i

l\\

initial load reduces the time required to reach a stable level of load through the relaxation mechanism. The data, however, are not sufficient to prove this theory. Figure 5 for 3/4 in. diameter anchors shows some data scatter, W { which we attribute to the strain gages. S. Figure 6 for 5/8 in. diameter anchors shows two anchors with a generally parallel relaxation curves and the third with a flatter slope. n D e This illustrates the nonuniform behavior which is typical of the drilled y, E g in expansion anchors we have tested. N s t Figure 7 for 1/2 in. diameter anchors shows generally good m n er correlation except for the last data point on Run 15. n Figure 8 is interesting from the fact that this anchor was d instrumented with both strain gages and a load cell. Both types of g S g instrumentation appear to track well together. C i The five data plots were made af ter making the zero shif t correction a t e to the recorded strain values. The plots show the percent preload S, I remaining. Table 2 gives the initial and final loads, average loads and n' percent preload remaining. Although each group of three anchors was installed at equal torque levels, the load imposed varies considerably. There are insufficient test data to provide an accurate correlation between the stress in these anchors and the rate of relaxation. We also analyzed each set of data by exponential regression techniques in order to characterize the data obtained from each group of three anchors. The results are plotted in Fig. 9. The lines plotted in Fig. 9 represent the best fit of the measured data to an exponential [ equation and might be used for projecting relaxation values in the absence of additional long-term experimental data. 7.2.20-C//P334 l Pf8

....g 7 . uwlmn eA n,lpJ wI ,j -x -~ l .l U I-l A 3/4 I"*

• 3"II' I"

'l Q ll i l k' f ! l l ((I[ Q 3 5/8 in. x 2-3/4 in. ]I l ll'i'l h l' l':G I ; [k' l II Y l 11. l L: it f li C 1/2 la. x 2-1/4 in. ddN 'i O "' * ~ l N') 14 fEi$p WE @!S$ p$ ^~ o l y g we m y ,.t dmiy ,u k ng g u ggw' L4 gj p ag'

NIN, hh [!

lfj,! M hjMsi M+DML i I 8 W hp py w yig,e n .3,..w i m :< y p r w m e a g 2m c Uh n t %p.o.a JIN 3 w . 'l a ) et l i gp -i i W tw .I J -.F :6 ~.3~ a dg pp i I9 q brly I.. w e m m 1 l!qR hi f &. j$ [i m [. m+ n m I s . 1. : - i I r__~.m . n. n --i u i ,7 l l 4.. i qd ,_- i, ~$ 1 i sp! Mim eq, c mMr= a I-m y d m ak . o xm.t i MHV '(lij M l' al @1 j' l; W p ) N ai ~ l l Ji l i$w n : n a . bF!!i G ] M 6 -- i 4 I-P

o !-l D

.1 1 10 100 1000 10,000 n m Ar m m n Ai. n as m inc (o m ) s (n h Fig. 9 - Results of Exponential Regression Analysis m

TABLE 2 - RESULTS AFTER ZERO SHIFT CORRECTION Anchor Relaxation Initial Final Preload Diameter Period Load Load Remaining (in.) Run (days) (1bs) (1bs) (%) 1/2(I) 1 212 3859 1196 35.6 1/2 2 211 2847 950 33.4 1/2 3 211 2597 953 36.7 Average 211.3 2934 1033 35.2 ~ 1/2(1) 14 187 3093 1262 40.8 1/2 15 167 3681 1144 31.1 1/2 16 186 3177 1519 47.8 7 Average 186.4 3317 1308 39.4 5/8(2) 7 206 6885 2556 37.1 5/8 8 206 8989 2995 33.3 5/8 9 206 10059 6003 59.7 Average 206 8644 3851 44.6 0 3/4(3) 4 206 7356 3179 43.2 3/4 5 206 9441 4252 45.0 3/4 6 206 8850 4340 "e 9. 0 Average 206 8549 3924 45.9 Notes: (1) 2-1/4 in, embedment (2) 3-3/4 in. embedm'ent l (3) 3-1/4 in. embedment l l 1141' 722o-C.// C.5 34 w b 36 u -

The slopes of the lines are seen to generally increase with anchor diameter and initial load. There was no apparent end to the relaxation process except for the case of the group of three 1/2 in. disaeter auch r8 (Run 1, 2, and 3) which had a 1 w initial 1 ad imposed. We think W i this demonstrates that the relaxation of expansion anchors will s 5, ,y eventually reach a limit which is established by the initial stress in i a' g the system. The lower the initial stress level, the sooner this limit e will be reached. E 1 The same load data used to plot Fig. 9 was used to plot Fig. 10 5 f which is an exponential regression analysis of the data in terms of I e I percent of initial load remaining versus time.' The slopes of the four aj lines are nearly parallel. The data for the 1/2 in. diameter anchors A shews a more rapid initial loss of tension, but we cannot reach a general s conclusion that this size anchor will always exhibit such behavior C because the amount of data is limited. a t ( The data from all strain gaged anchor tests were grouped to obtain f the plot shown in Fig. 11. A similar plot of data from the load cell C. instrumented anchor is presented in Fig. 12. In Fig. 13 the data from all relaxation tests, including the load cell tests given in our May 28, 1982 report, are plotted and a regression line is shown. This plot best i suusnarizes the experimentally documented behavior of the drilled-in anchor bolts we tested. Extrapolating the one year test program data to j 40 years indicates that on average, approximately 25 percent of preload will remain in the anchors. However, the standard deviation (a) of data about this mean value is

large, equaling 9.62 percent of load mp.22o-c H S-334 ere r3 9

=. -

l i // / / m // / / // / / // / / ll l / // / I // / / / t/ i ' h / / ;l i k lii g C - m .S l - 3 e u cn 1 t o e c z g, H xe w LD 0 ~ i i ?iiii Iceee i g???? o e~~~ ~ Eiisi C } 5ee_ce 3 3- <.o. I I t i f

• g g

8 8 ? 2 9NINIVH38 OV01 IN3DB3d 51417 220- C //T' 3.5-3 EEOOOt L 1PI l -e --.y- .-,.--y. ,,.-,y _.g

I, / / / .a / E / / / / I 3 / a o 5 g .c g R a .a a ao g ~ e v4 r.n a y cm wo H .e ,t Lu R e 8 a .O .4. u 00 G AC l ~ sH .s.e.o Gne I i f SNINIYH38 OV01.lN3383d

t 1141'-j s.2 o-u/r-. 33. 2.

1 8J33 >EOOO! i l' 1 I --- --~~.--_._ _ _..._ _ w y v w ,---,e.-.-- y v.-- -,w-- --e- ,.-g.p w

00 ' 8 4 N \\ 05

• 8 N

3 N a t N a d N 5 r 8 o N 3 hc N na S N S de t n n 8 e o 2 m i t u 4 c = r e 1 t rr s o - 7 n y c i t f l i h S l e s Y c o A r - 1 D d N ez a r o e N l t f I M a fo e E ta G s d i A N l s l y e l c N a d n a a m I n o 7 i 1 ss .o e N r g ts e e R T 2 1 .g iF -0 0 0 o 0 0 0 0 9 s 4 2 1 DZH2g<1 E O:43 WZJOEJLQ 1 1 L L $AVP $.Og 7 wWIh gg* I nMOOm. I I 1 l .!l! I l!ii i ;'i, l:

000 s 4 N 1 s s 0 x 5 . 83 s a s t 5 a ,,;. + I . 8 d 3 , ;. + ..q.!.:: .[ t s .i... e 1 t s . 9 ,. e \\.,' :,g; n o i 8 t g 1, . 2 a 4 x 4 N *.:.: k. . 7 r k . 1 a l = e N .1 .X,,:. i 4,f l l S a f Y f A o .:,i . 1 D s g-i N s

-i,;. =

i I

y l E a .. = , i n G a A ii no iss er a g taa e dt R a ed g al gl 3 e nc 1 d i l n ad e ra g t o g e SL i g :!l L F

x+ 0 0 o o 0 o 0 a d 2 1 U o 5 z s-4gmc a<om zJJCE tLL = A* Mb $ 4 oh%s mwtb

• mc a O;

- v~ hah i 4 ji ^

m--+.-wiss ~P E-w.w eaw.. a w's re m aing. Figure 13 also shows the three o limits, indicating that a small fraction of the anchors may completely lose their preload before 40 years of service is obtained. We continue to believe there may be a lower limit to loss of g Preload, but cannot predict its value based on the data obtained in the S S, y short-term test program. a n n e

SUMMARY

AND CONCL11SIONS E I S f Twelve wedge type expansion anchors were tested in four groups of er three to measure loss of preload during a seven month testing period. a D When all anchors are taken together, an average loss of 57 percent of d A initial preload was found. The individual anchor losses ranged from i S 40 percent to 69 percent. C k The shape of curves obtained by plotting preload remaining versus t l [ time indicates that relaxation in most cases will continue beyond the i I first year. n C. Extrapolation of the test data from the strain gaged anchors to longer periods is difficult considering the scatter of the data. This series of tests confirms the conclusion made after reviewing the results of the earlier relaxation tests. Accurate prediction of the ultimate loss in preload in individual anchors, and when it will occur cannot be made based on the data available. For large numbers of anchors, the j extrapolation of data from all relaxation tests indicates that approximately 25 percent of the preload will remain on average after 40 M 1141'f 2 3gp g//7-jj J f 3d 3

O years of service. However, individual anchors may lose all preload in the 20 to 40-year period af ter installation. This conclusion is based on extrapolation of one year's testing to 40 year's parformance. Hence the Possibility that a relaxation lower limit exists cannot be ruled out. p I s S. J Respectfully submitted, a WISS, JANNEY, ELSTNER AND ASSOCIATES, INC. f' Ib. 1 F. D. Heidbrink 7 s Project Engineer n a J. R. C. Miller j Project Manager A 5 S 0 t i a Reviewed by: MM [ Richar'd G. Lind'strafm f Quality Assurance Manager c. i v. i f i . iW-1141 MAO t337 m. _w

---

,y 7.

-}}