DEPARTMENT OF THE ARMYU.S. Army Corps of Engineers

CECW-ED Washington, DC 20314-1000 ETL 1110-2-355

Technical LetterNo. 1110-2-355 31 December 1993

Engineering and DesignSTRUCTURAL ANALYSIS AND DESIGN

OF U-FRAME LOCK MONOLITHS

Distribution Restriction Statement

Approved for public release; distribution is unlimited.

DEPARTMENT OF THE ARMY ETL 1110-2-355U.S. Army Corps of Engineers

CECW-ED Washington, DC 20314-1000

Technical LetterNo. 1110-2-355 31 December 1993

Engineering and DesignSTRUCTURAL ANALYSIS AND DESIGN

OF U-FRAME LOCK MONOLITHS

1. Purpose

This engineer technical letter (ETL) provides guid-ance for performing structural analysis of U-framemonoliths for navigation locks.

2. Applicability

This ETL applies to all HQUSACE elements, majorsubordinate commands (MSC), districts, laboratories,and field operating activities (FOA) having responsi-bilities for the design of civil works projects.

3. Discussion

Design and analysis of U-frame lock monoliths is acomplex structural engineering task. It involvesmany assumptions and methods not required for

design of other types of structures. First-time design-ers of U-frames may overlook some of these keyassumptions and methods, possibly resulting in eitheran uneconomical or an inadequate design. Theappendixes to this ETL present a recommended pro-cedure for design of U-frame lock monoliths, basedon recent experience at several design districts.

4. Action

Structural engineers should review Appendix A priorto initiating design of a U-frame lock. It should beused as a guideline for developing the design processand required design resources. Modifications to therecommended procedure may be appropriate based onthe designer’s previous experience and on specificproject conditions. However, such modificationsshould be developed in consultation with the designteam, appropriate MSC personnel, and CECW-ED.

FOR THE DIRECTOR OF CIVIL WORKS:

2 Appendixes PAUL D. BARBER, P.E.APP A - Structural Design and Analysis Chief, Engineering Division

of U-Frame Lock Monoliths Directorate of Civil WorksAPP B - References

ETL 1110-2-35531 Dec 93

APPENDIX A: STRUCTURAL ANALYSIS AND DESIGNOF U-FRAME LOCK MONOLITHS

1. Introduction

This guidance is to be used by the structural engineerduring the design of a U-frame lock monolith. AU-frame lock is a structure in which the base slab ofthe lock and the walls of the lock are monolithic.Therefore, a U-frame lock as discussed hereinincludes a W-frame lock structure. The advantagesof the U-frame type of lock are the reduction in thevolume of concrete in the walls, better seismicresistance, a reduced number of monoliths to design,a structure that is more readily dewatered, a possiblereduction in pumping costs during dewatering due toless seepage, balancing of the loads applied to themonolith, and minimization of differential settlementand rotation of the walls with respect to the base.

1-1. Scope. This appendix includes technical guid-ance on structural analysis and design of U-framelock monoliths. Planning and layout of navigationlocks is covered in other guidance. Short excerpts ofother guidance documents are repeated herein. Otherguidance in this appendix includes definition of indi-vidual loads and load combinations; structural analy-sis methods and design assumptions; constructability;and serviceability. Additional guidance pertinent toU-frame lock design is contained in several otherengineer regulations, engineer manuals, and engineertechnical letters (as referenced in this document).Topics addressed by these other documents includestrength design for reinforced concrete; seismicdesign; pile foundation design; and thermal-mechanical analysis of concrete. Such topics areaddressed briefly herein; however, for details of thesetopics the engineer must see the referenceddocuments.

1-2. Applicability. This guidance should be usedon any civil works project that contains a U-framelock. The guidance provided herein can be usedbeginning with the reconnaissance phase of projectdesign and should continue to be referenced throughthe preconstruction engineering and design phase andpreparation of plans and specifications. The guidancemay also be used as needed for engineering duringconstruction.

1-3. References. References are included inAppendix B.

2. Design Planning

2-1. Coordination. Throughout the planning anddesign process, coordination within the design team isessential in achieving a quality design product. Thedesign team should consist of representatives fromconstruction/operations division, the cost sharingcustomer, planning, real estate, safety, cost engineer-ing, and a representative from each of the engineeringdisciplines. Changes made to the structural configu-ration will often impact geotechnical, hydraulic,mechanical, and electrical engineers. Therefore,frequent communication with counterparts on a regu-lar basis will facilitate identification of any problemsthat may have been created by the change. Coordina-tion with higher authority is also necessary asdescribed in ER 1110-2-1150.

2-2. Design sequence. Structural design of amonolith will be performed during various designphases. These are the reconnaissance phase, feasibil-ity phase, preconstruction engineering and design(PED), and engineering during construction (EDC).The engineering requirements for each phase aredefined in ER 1110-2-1150. Specific structural engi-neering responsibilities are defined in ETL 1110-8-13(FR), and general navigation lock design require-ments can be found in other guidance. Each of thesedocuments should be reviewed during the reconnais-sance phase.

a. Reconnaissance report (RR). Analysis duringthe reconnaissance phase will usually be very limited;however, some analysis may be necessary to confirmthe feasibility of the proposed plan. The initial deci-sion to use a U-frame structure will be made duringthis phase. Since the plan presented in the RR isbased largely on engineering judgment, it is importantfor an experienced structural engineer to be involvedin this phase. The structural engineer must helpdevelop a reasonable project configuration and thedesign cost and schedule for the next phase. Depth

A-1

ETL 1110-2-35531 Dec 93

of the lock foundation should be carefully assessedduring this phase since the depth can have a signifi-cant impact on construction cost.

b. Feasibility report (FR).

(1) Lock design during the feasibility phase (FP)must be sufficient to fully define the project scopeand to develop an adequate baseline cost estimatewith reasonable contingencies. The design will bepresented in the engineering appendix to the FR.Also during the FP, the structural engineer mustdetermine schedule and cost requirements for theremaining phases of the design. This informationmust be included in the Project Management Plan.

(2) Structural content in the engineering appendixwill include a full definition of functional and tech-nical design criteria, and initial analyses used toestablish the basic geometry for the project. Thesuitability of a U-frame configuration must be firmlyestablished during the FP. Decisions should be madeon foundation type (soil or piles), chamber size, wallheight and thickness, sill height, slab thickness, andfoundation depth. The design team should also selecttypes and sizes of guidewalls, filling and emptyingsystems, and closure systems for maintenance andemergencies. The structural engineer must be directlyinvolved in this process, and sufficient analysisshould be performed to verify these decisions.

c. Design memorandum (DM). Detailed designand analysis of the lock structure are documented in aDM. The DM is prepared either during PED or EDCphases. While design details may remain incomplete,the DM should contain an essentially complete designof the U-frame lock. During this design effort thestructural engineer should accomplish the following:verify all design criteria; define all loads and loadcases; select controlling load cases; develop a finalpile layout if required; verify all member geometries;determine required concrete reinforcing steel; andcalculate quantities for use in developing the costestimate. This work should be thoroughly docu-mented in the text and plates of the DM.

d. Plans and specifications (PS). Contractrequirements are defined by the PS. The PS for thelock are usually prepared during the EDC phase sincethe PED phase ends with the initial constructioncontract for any other project feature. Structuralengineering work for preparing the PS includes thefollowing: lay out main reinforcement using

quantities calculated for the DM; detail all secondaryreinforcement; prepare and check contract drawings;develop and edit specifications; and calculate quanti-ties for use in developing the government costestimate.

2-3. Types of monoliths.

a. Background.

(1) As a result of the functional requirement fora lock to contain large tows, the chamber lengthparallel to flow can become very long. On a majorwaterway the chamber length can be up to 1,200 ftlong. Consequently, it becomes necessary to incorpo-rate monolith joints along the chamber length. Thelocations of these joints are used to define monolithswhich have unique requirements towards the operabil-ity of the overall lock. Additional guidance withrespect to locations, requirements, and lengths neededfor the various types of monoliths is available inother documents.

(2) Monoliths that compose a lock can be cate-gorized into five general groups. These five groupsare:

(a) Intake/discharge monoliths.

(b) Gate monoliths.

(c) Culvert valve monoliths.

(d) Chamber monoliths.

(e) Other monoliths (e.g., guardwall monolith,bulkhead monolith).

(3) The complexity of design of these monolithsvaries and the degree of complexity should be consid-ered when assigning the design tasks for these variousmonoliths. Gate monoliths are the most difficult todesign primarily due to the three-dimensional loadingwhich is applied to these monoliths. Design of a gatemonolith should be performed by a senior engineer.Intake/discharge monoliths and culvert valve mono-liths can be difficult to design due to the fact that thegeometries of these monoliths are difficult to evaluatein two dimensions. The chamber monolith is thesimplest of the monoliths listed to design since it canusually be designed in two dimensions without con-cern about the out of plane direction. Design of achamber monolith can be performed by a junior

A-2

ETL 1110-2-35531 Dec 93

engineer with the guidance of an experienced engi-neer. Other monoliths will generally be more diffi-cult to design than a chamber monolith but not asdifficult as a culvert valve monolith or intake/discharge monolith.

b. Intake/discharge monoliths. The intake anddischarge monoliths are located at each end of thelock. The intake ports must be located in the upperpool while the discharge ports are located in thelower pool. Generally for a U-frame the intake/discharge monoliths will be U-frame monoliths, but itis not necessary if the applied loads are balanced oneach wall. Placement of the manifolds towards theoutside of the monolith will be advantageous inreducing congestion caused by the intersection of walland base reinforcing if the manifolds are placedtoward the inside of the monolith. However, site orhydraulic conditions may require that the discharge beplaced on the inside of the monolith. Various portgeometries and face locations have been used in thepast and are generally established by the hydraulicsengineer. The selection of the type of port should bemade during the feasibility phase of a project, but theactual geometry will be determined for the designmemorandum.

c. Gate monoliths. The gate monoliths arelocated at each end of the lock and house the gatesused to let tows in and out of the lock chamber. Inaddition, these monoliths usually house the machineryused to actuate the gates. All anchor supports, bear-ings, and other embedded metals are contained withinthe walls and bases of these monoliths. Sills arefrequently provided along the chamber base to estab-lish draft requirements and provide a sealing surfaceto minimize flow between the gate and the concretebase. In the area of the sill, consideration should begiven to including a recess in the base for silt depos-its which would interfere with gate operation ifrecesses were not present. The gate monolith mustalso provide a recess to house the gate in its openposition. Bulkhead slots are usually located upstreamand downstream of the lock gate to allow for emer-gency and maintenance dewatering. These bulkheadslots may or may not be contained within the mitergate monolith.

d. Culvert valve monoliths. Culvert valve mon-oliths contain valves which control the filling andemptying of the lock chamber. A culvert valve mon-olith is located at the upstream and downstream endof the lock. The culvert valve is supported by

anchoring into the monolith wall. The operatingmachinery for the culvert valve is also housed in theculvert valve monolith and is generally located at thetop of the lock wall where steel frames are embeddedin the concrete to secure the machinery. Bulkheadslots are provided for dewatering of the culvert valverecess for maintenance and inspection of the culvertvalves. Since the culvert valve monolith generallylies between one of the massive gate monoliths andone of the more slender chamber monoliths, it isoften used for a transition to align the culvert in theother two monoliths (see Figure A-1).

e. Chamber monoliths. The chamber monolithsare basically included to provide continuity of thelock chamber and culvert between the upstream anddownstream culvert valve monoliths. Since thechamber monoliths are not required to support valves,gates, or operating equipment their walls are muchthinner than the other monoliths. To save concrete,the outside edge of a chamber monolith can betapered above the culvert. The chamber monolithscontain ports from the culvert to the lock chamberwhich are used for filling and emptying the lock.The spacing of these ports should be considered whendetermining the length of the chamber monolith.These monoliths will often contain a gallery near thetop of the monolith which extends the length of thelock and is used to carry electrical wires and mechan-ical piping.

f. Other monoliths. Additional monolithswhich may be included in a U-frame lock could beguardwall monoliths and bulkhead monoliths. Aguardwall monolith will not always necessarily be aU-frame structure since it may only be needed on oneside of the entrance to the lock. Typically, a guard-wall monolith will be placed at the entrance of thelock to transition the area into the lock. A bulkheadmonolith will become necessary if the location of thebulkheads is not contained within one of the othermonoliths. If it is necessary for a lock to have abulkhead monolith, it will often be of similar geom-etry to a chamber monolith. While a bulkhead mono-lith’s geometry may be similar to a chambermonolith, its loading will not be since it will need tocarry load in the upstream and downstream direction.Finally, any of the aforementioned monoliths can alsoact as bridge pier monoliths. When a bridge pier islocated on a monolith, it can have an effect on thedesign of the monolith due to the loads transmitted tothe monolith from the bridge pier. This is particu-larly true in active seismic areas.

A-3

ETL 1110-2-35531 Dec 93

Figure A-1. Culvert valve transitions monolith. Plan view and elevation

A-4

ETL 1110-2-35531 Dec 93

2-4. Foundation alternatives. The decisionwhether to have a pile-founded or soil-foundedU-frame lock must be based on numerous consider-ations: erodibility of foundation, potential for scour,factor of safety against flotation during dewatering ofthe lock, differential movements between monoliths,soil-bearing capacity, sliding stability for large unbal-anced loads, the level of seismic activity, projectlayout, and cost. The type of foundation will affectthe analysis methods and monolith geometry.

2-5. Two-dimensional versus three-dimensional behavior. U-frame lock monolithscan be categorized as two- or three-dimensionalbehavior. The behavior of a monolith is dependentupon both geometry and loading. Gate monoliths actthree dimensionally due to loads on the gates whichact in the longitudinal direction. Even though thebehavior of a gate monolith may be three dimens-ional, it is possible to analyze a gate monolith byusing a set of two-dimensional analyses to capture thethree-dimensional behavior. Culvert valve and intake/discharge monoliths can be considered three-dimensional monoliths due to their geometry butgenerally two-dimensional approximations of thesemonoliths can be made which adequately capturetheir behavior for design purposes. Seismic analysesand nonlinear, incremental structural analyses of theculvert valve and intake/discharge monoliths shouldbe three-dimensional analyses since the geometry ofthe structure has a much larger impact for theseanalyses than for static structural analyses. Chambermonoliths behave strictly as two-dimensional mono-liths unless a loading in the longitudinal direction isplaced on the monolith.

3. Design Criteria

3-1. General. The design criteria for navigationlocks must be established during the feasibility phaseof the design process. Criteria should cover stability,strength, serviceability, and foundation requirements.There are three categories of loadings defined inother guidance. These categories include usual,unusual, and extreme load conditions as described inparagraph 5-2. The separation of load conditions intothese categories implys the nature, frequency, andconsequence of the loading and also dictates therequired factors of safety. The decrease in factor ofsafety allowed in all three categories of loading main-tains limits that yield a linear elastic response of thestructural elements. In some situations, however,

nonlinear response could be acceptable dependingupon the extent and duration of the response, redun-dancy of the structure, potential damage implications,and probability of occurrence. Decisions on accept-able nonlinear behavior must be coordinated withCECW-ED.

3-2. Stability. Stability criteria are applicable tosoil-founded locks and include safety requirementsagainst sliding, flotation, and overturning.

a. Sliding. Sliding stability is defined and cal-culation methods are shown in ETL 1110-2-256, andEM 1110-2-2502.

b. Flotation. Floatation stability is defined andcalculation methods are shown in ETL 1110-2-307.Some provisions that are unique to U-frame locks canbe found in other guidance.

c. Overturning. Overturning is not usuallycritical for U-frame lock monoliths because of thebase width. However, bearing capacity at the outeredges of the structure is a concern. Stability withrespect to resultant location is defined in EM-1110-2-2200 and EM 1110-2-2502. More information onresultant location calculations is found in para-graph 6-3. When performing resultant calculations itis important that unfactored loads be used. The resul-tant should fall within the middle third of the struc-ture under usual loading conditions.

3-3. Strength. All components of a lock monolithmust be able to resist all load conditions, includingthe reinforced concrete framing members, structuraland miscellaneous steel, foundation piling, and foun-dation material.

a. Reinforced concrete. Detailed design guid-ance for reinforced concrete sections is covered inEM 1110-2-2104.

b. Steel structures. Design of structural steel,embedded metal, and miscellaneous steel should bebased on EM 1110-2-2105. Design of major lockappurtenances such as miter gates, tainter gate valves,and associated machinery is covered under variousother guidance publications including EM 1110-2-2703 and EM 1110-2-1610.

c. Foundation piling. Detailed design guidancefor pile foundations is contained in EM 1110-2-2906.Further discussion is found in paragraph 6 below.

A-5

ETL 1110-2-35531 Dec 93

d. Foundation material. This information isgenerally determined by the geotechnical disciplineand furnished to the structural engineer. Bearingstrength for soils and methods for determining bear-ing strength based on field and laboratory data aredescribed in EM 1110-1-1905.

3-4. Seismic design criteria. Seismic designcriteria are defined in other guidance. Seismic loadcases are generally considered unusual or extremeconditions and have reduced factors of safety associ-ated with the criteria. The design that results fromstatic analysis for usual and unusual conditions can beadequate for seismic response integrity. This is dueto the use of damping effects, more sophisticatedanalyses, and different load factors in seismic analysisand design of reinforced concrete monoliths.

3-5. Serviceability. Serviceability requirementsare unique among projects, and the designers areresponsible for establishing these requirements. Toestablish serviceability requirements the design teamshould consider a number of aspects in the lock struc-ture. Considerations should include minimization ofconcrete cracking, seepage and leaking, and reinforce-ment corrosion. Global deflections, settlement, andrelative deflections are other primary concerns, espec-ially those that affect mechanical interaction with thestructure such as near the miter gate sill or valvelocations. Additionally, maintenance, personnelaccess, and safety are important considerations forserviceability.

4. Loads

The most common loads on a U-frame lock are thosedue to the dead weight of the concrete structure, andthose loads imposed from the soil and water whichsurround the structure. Additional loads from servicegates, valves, bulkheads, emergency closure equip-ment, and operating machinery are also present. Thestresses imposed by temporary loads must also beconsidered. Such loads are barge impact, ice, earth-quake, wind, etc. Other loads such as surcharge anddebris and ice loads may exist, depending upon thesite-specific conditions. The following paragraphsprovide a brief discussion on each of the commonlyencountered loads and also provide guidance on howthese loads are determined. Further information onthese loads is contained in other guidance.

4-1. Dead loads. Dead loads consist of concreteand structural steel items such as miter gates, taintervalves, and emergency closure and maintenance bulk-heads. The weight of the concrete structure is com-monly the predominate force in the design of aU-frame lock. This load must be appropriately dis-tributed so that its centroid coincides with the geo-metric centroid of the concrete item being analyzed.Effects of buoyancy on the concrete are accounted forseparately as uplift forces described below.

4-2. Water. Water is either free standing or con-fined. Free-standing water refers to water above thesoil, which is unaffected by either seepage or headloss. For example, water contained within the lockchamber, lock culverts, or outside the lock wallsabove any backfill, silt, ground surface, or concretesurface is free-standing water. This water producespressures normal to any surface or plane which itcontacts. For convenience, water pressures are con-sidered such that the forces are either horizontalpressures or vertical weights. Confined water is thatwater which exists below the saturation line in anybackfill and foundation material. This water producespressure normal to any surface just as free-standingwater except that the effect of seepage and head lossmay need to be considered in determining the valueof the pressure at any point. Water forces on mono-lith expansion or contraction joints resulting frombroken waterstops must also be considered in analysisand design. The value of the horizontal pressuremust equal the value of the vertical uplift pressure atany given point.

4-3. Uplift. The pressure of water creates forcesacting upward on the bottom of the U-frame lockbase. These uplift pressures are determined by multi-plying the head of water above the bottom of the baseby the density of water. The value of head used mustinclude the effects of seepage from the upper pool tothe lower pool. The rate of head loss must be deter-mined for each project depending upon the permeabil-ity of the foundation, total head for the project, andpresence of pressure relief systems (foundationdrains). See Sherman (1968 and 1972) for additionaldetails.

4-4. Soil pressures.

a. Vertical. The vertical weight of any soilacting on the structure is determined by multiplying

A-6

ETL 1110-2-35531 Dec 93

the density of the soil by the volume of soil present.For soil above the water table, a moist density isnormally used. For soil below the water table, asubmerged (buoyant) density is used.

b. Lateral. The value of lateral soil pressure atany point is determined by multiplying the verticalweight of soil above the point (with the appropriatereduction in soil weight for when submerged) by thelateral pressure coefficientk. Observations frommeasured earth pressures on U-frame locks haveshown that the coefficient of lateral earth pressurevaries over a wide range along the height of the lockwall and also varies with time. Because of this varia-tion, it is common practice to bracket the earth pres-sure by using an upper and a lower bound fork.These maximum and minimum values are then com-bined with the appropriate load cases for analysis in aconservative manner. Selection ofk values for designshould include knowledge gained from past instru-mentation programs as well as results of classicalsolutions. Factors affectingk are flexibility of thewall, soil types, loading case, seismic condition,density of the backfill, and shear strength of the back-fill. Examples ofk values used vary from 0.2 to 1.0.

c. Silt. Since the load that the structure reactsto from earth pressures depends upon the depth ofbackfill, the possibility that silt could be depositedabove the backfill must be taken into account. Theamount of silt that may be deposited upon a structureis dependent upon the silt load the waterway is carry-ing, the water velocity at the location siltation isexpected to occur, and the geometry of the structure.These factors vary for each project and must be con-sidered when deciding upon the amount of depositedsilt. For example, the lower Red River in Louisianasees silt deposits of between 5 and 20 ft every timethe river rises and falls. Silt loads are incorporatedinto the vertical and horizontal earth pressures byconsidering the silt as an additional soil layer aboveother existing soil layers. Normally silt deposited inthe lock chamber is not considered as a load on thestructure.

4-5. Drag. The downdrag force is a shear forceacting downward along a vertical plane adjacent to ornear a structure-to-soil interface (Ebeling, Duncan,and Clough 1990; Ebeling et al. 1992). Downdragforces acting on the stem and culvert walls ofU-frame locks are more difficult to characterize thanare those for gravity walls founded on rock becauseof the interactions among the structure, the

foundation, and the backfill. For guidance regardingdowndrag, reference should be made to the resultsdiscussed in the report on the Port Allen and OldRiver Locks (Clough and Duncan 1969) as well asthe report on Red River Lock and Dam No. 1(Ebeling et al. 1993). These studies show theextremes with respect to downdrag which have beencomputed to date. Given the state of practice, acomplete soil-structure interaction (SSI) analysis isthe most reliable procedure available for estimatingdowndrag forces on soil-founded U-frame locksbecause the more compressible the foundation is, thegreater the need for SSI analyses to determine thevalues to be assigned for downdrag forces. The finiteelement method of analysis is used in this type ofanalysis to compute the stresses and displacementsfor both the structure and the backfill. The finiteelement program SOILSTRUCT has the capabilitiesfor performing a complete SSI analysis to obtaindowndrag forces and has been used successfully onnumerous projects, including those cited in this sec-tion (Ebeling, Peters, and Clough 1992).

4-6. Foundation pressure. Foundation pressureis the response of the soil to the force of the structureplaced upon the soil. To determine the foundationpressure, see paragraph 6, Foundation Analysis.

4-7. Impact. Impact loads from barges striking thelock walls should not be considered in the overalldesign of the monolith, but should be considered inlocalized areas. These loads are covered inETL 1110-2-338. The impact loads on the lock wallsare generally less severe than those on more exposedcomponents of a navigation project since the angle ofimpact is limited in the lock, and speeds are generallymuch slower in the lock. The magnitude of theseforces is dependent upon the size of the tow andbarge that the lock can accommodate and therefore isproject-dependent. Loads transferred through the lockgates due to impacts on the gates should be consid-ered. Guidance on values to be applied to gateimpacts is included in EM 1110-2-2703 andEM 1110-2-2105.

4-8. Hawser. Hawser loads should not be consid-ered in the overall design of the monolith, but shouldbe considered in the design of the upper part of thelock wall. Hawser loads are the forces generatedresisting the inertial forces of moving barges.Hawser loads act in the opposite direction to impactloads. These loads are covered in other guidance.For additional information, see ETL 1110-2-247.

A-7

ETL 1110-2-35531 Dec 93

4-9. Seismic loads. There are two types of seis-mic analyses, psuedostatic and dynamic. The psuedo-static analysis is used in low seismic zones and withlow peak ground accelerations (PGA). The dynamicanalysis is used in higher seismic zones and withhigher PGA’s. The extent of seismic analysisrequired depends upon the results of the seismologicalsite investigation. Detailed requirements on the seis-mic design of U-frame locks are covered in otherguidance.

4-10. Gate reactions. Any force resulting fromthe dead weight of lock gates or culvert valves andforces transferred from the gates due to hydrostaticpressure acting on the gates should be considered inthe design of the gatebay or valve monoliths.EM 1110-2-2703 describes how these forces shouldbe determined and applied to the monolith.

4-11. Thermal. Thermal loads are caused by volu-metric changes caused by changes in temperature,temperature gradients through sections of the concretestructure, geometric discontinuities in the structure(e.g. culverts), and external restraints (e.g. piles).Thermal loads must be evaluated in a U-frame lockstructure because large member thicknesses preventthe heat generated from hydration from being dissi-pated as quickly as it is generated and therefore thetemperatures within the structure rise. Effects ofthermal loads can be evaluated through a nonlinear,incremental structural analysis as described inparagraph 7-5.

4-12. Cofferdam tie-in. The forces exerted on thestructure by cofferdams should be incorporated intothe appropriate monoliths. The forces in sheetpileinterlocks as well as any horizontal loads and dragloads from the fill within the sheetpile cells should beadded to the appropriate monolith.

4-13. Sheetpile cutoff reaction. The sheetpilecutoff reactions on U-frame monoliths are generallysmall and neglected. For further information, seeother guidance.

4-14. Localized loads. Loads from equipmentand appurtenant items are discussed in other gui-dance. These loads must be carried by the lockmonolith. Localized loads normally do not controlthe structure design or overall stability, but may con-trol the design in localized areas. Examples of thistype of load include the horizontal thrust from themiter gate bull gear support frame, the emergency

bulkhead crane pedestal, support columns for controlhouses, tainter valve trunnion anchorages, miter gategudgeon pin anchorages, miter gate pintle bases,miter gate latches, emergency bulkheads and mainte-nance bulkheads, emergency bulkheads loweringcarriage machinery, jacking forces on gatebay mono-liths due to gate diagonals tensioning operations, arealighting towers, etc. In the case of jacking forces ongatebays due to gate diagonals tensioning operations,anchors and/or jacking points should be provided forin the design of the gatebays to ensure that sufficientmeans are available to tension miter gate diagonals.

4-15. Other loads. Wind loads are relativelysmall and should be neglected. Ice loads on lockwalls are not ordinarily included in the structuraldesign. However, approach walls and mooring facili-ties, particularly those items in the upper approach,are sometimes subjected to moving ice and the effectsshould be accounted for. For isolated installationswhere ice conditions are severe, and the ice sheet isshort and can be restrained or wedged between struc-tures, its magnitude should be estimated, with consid-eration given to availability of records of ice condi-tions. It is recommended that an impact pressure ofnot more than 5,000 lb/sq ft be applied to the contactsurface of the structure, based on the expected icethickness. In the United States the ice thicknessassumed for design normally will not exceed 2 ft.Ice pressure should be applied at the upper poolelevation. For further information, see EM 1110-2-1612, ETL 1110-2-295, and ETL 1110-2-320. Super-structure loads include the reactions to control housesand access bridge spans. Miscellaneous loads includelarge temporary surcharge loads and mobile equip-ment loads. Typical items may include cranes formaintenance and placement of stoplogs, constructionequipment used for concrete placement, etc.

5. Load Cases

5-1. General. Load cases are combinations of thevarious loads described in paragraph 4. The forces inthe load cases are factored or unfactored dependingupon the analysis being performed. The load factorsused depend upon the type of load, certainty of mag-nitude of load, and frequency of load being applied.

5-2. Categories of load cases. The structureduring its life will be subjected to many differingloads. The severity of these loads and the frequencyof their occurrence along with the consequences of

A-8

ETL 1110-2-35531 Dec 93

the structure being damaged lead one to grouping theload cases into three different categories: usual,unusual, and extreme. A usual load case is one thataffects the structure for extended periods on a recur-ring basis. Such load cases permit no reduced loadfactors to the structure components. An unusual loadcase is one that the structure sees occasionally and/orfor short periods. Such load cases have minor reduc-tions to load factors on the structure components. Anextreme load case is one that might happen only onceor twice to the structure. Such load cases have majorreductions to the load factors on the structurecomponents.

5-3. Load combinations. Load cases are formedby combining the individual loads together. Usuallythe controlling load cases will be those with thegreatest vertical or lateral loads. However, somecases may control that do not have the highest verti-cal or lateral load but for which the combination oflateral and vertical loads is more severe. In addition,for monoliths subject to dewatering, the dewateredcase may require that additional means be provided toresist uplift or that limits be placed upon the maxi-mum pool elevations at which the monolith may bedewatered. Both maximum and minimum coeffi-cients of lateral soil pressure can be used in the fac-tored stress analysis to bracket the actual lateral soilpressures. Also, both uniform and stepped bearingpressure distributions are used in the factored stressanalysis to bracket the actual base pressure distribu-tion. By combining the above loads with the varyingfoundation pressure distributions and limiting lateralsoil pressure distributions, the number of possibleloading combinations soon becomes astronomical.All of these combinations need not be analyzed.Sensitivity studies, engineering judgment, and otherrational methods should be used to select a reasonablenumber of cases to analyze. It is important to notethat for pile-founded structures, some load cases maynot be critical for the pile foundation design but couldcontrol the concrete monolith design. All load casesused on pile-founded structures should be analyzedafter the final pile layout has been developed so thatproper pile forces are included in the concrete mono-lith design.

5-4. Application of load factors. Analysis withunfactored (service) loads is used for foundationdesign and can be performed for the ultimate strengthdesign of reinforced concrete. For frame analysiswith service loads, internal member reaction resultsare factored for member design. Application of

appropriate load factors is simplified when internalmember results are categorized as dead and liveloads. Alternatively, analysis can be performed withfactored loads which directly yields internal memberreactions for member design. Analysis with factoredloads causes the points of inflection to shift and thefoundation pressure distribution to change when com-pared with an analysis with service loads.

6. Foundation Analysis

6-1. Determination of type of foundation. Theselection of the type of foundation is probably themost critical aspect of the design of a U-frame lockbecause of cost considerations and the overall behav-ior of the structure. Since this decision will have asignificant impact on the project cost, the determina-tion of the type of foundation should be made in thefeasibility phase of the project. A thorough sub-surface investigation and testing program should beundertaken to define the soil strengths and param-eters. A soil foundation is usually more economicalif special measures (deeper excavation, elaboratepressure relief system, etc.) are not required. Thesoil foundation has to be able to satisfy stabilityrequirements for sliding and overturning, as well asresisting uplift (flotation) and earthquake forces.Included in evaluating soil-founded versus a pile-founded lock should be the consideration of differ-ential settlements between monoliths. If a soilfoundation is not feasible because of site conditions,then a pile foundation is required. When consideringa pile foundation, all types of piles should be consid-ered and the most feasible and economical types ofpiles should be chosen based on strength, geotechni-cal conditions, availability of material, and construc-tion limitations. In order to make these comparisonsand the comparison to the soil foundation alternative,pile quantities should be computed based on assumedlateral and vertical pile capacities and the minimumpile spacing that is expected, taking into considerationthe fact that the density of piles may need to behigher in some areas of the structure than others.This quantity computation should be performed onone of the more massive monoliths and on one of thechamber monoliths, and the results should then beextrapolated out for the entire lock. If this initialcomparison shows the pile foundation to be moreeconomical or approximately the same as the soilfoundation, then the designer should proceed with amore detailed analysis of the layout using the mosteconomical type of pile. A rigid base analysis can be

A-9

ETL 1110-2-35531 Dec 93

performed using computer program CPGA (Case PileGroup Analysis) (Hartman et al. 1989) for this pur-pose, or if a two-dimensional monolith is being eval-uated, a flexible base analysis can be performed usingCWFRAM. The final decision, which will be basedon cost, can be made using the refined pile layoutcompared with a soil founded configuration.

6-2. Pile founded.

a. General. The pile foundation design devel-opment should follow a procedure that conforms tothe normal submittal phases for civil works designprojects. During the feasibility phase of the project,the determination for use of a pile foundation shouldbe made, the most economical type of pile to usemust be decided, and an approximate cost of thefoundation should be projected. The actual detaileddesign of the pile foundations should occur during thepreparation of design memorandums for the project.The final pile layouts for all monoliths should bedeveloped so that only minor refinements and addi-tion of details are required during the development ofplans and specifications.

b. Types of piles. There are many differenttypes of piling that can be used for a U-frame lockfoundation, each with its advantages and disadvan-tages. Common types of piling include: steel H-piles,steel pipe piles, precast concrete piles, cast in placeconcrete piles, mandrel driven piles, and timber piles.For a detailed discussion of the types of piles andhow to evaluate them, see EM 1110-2-2906.

c. Initial pile layout. Determination of theinitial pile layout should be made in the feasibilitystage. Preliminary layouts for costing purposes canbe accomplished by using conservative lateral andvertical capabilities for a single pile and applyingthese values to resist the total lateral and verticalloads for the worst load cases. This gives a veryrough idea of the total quantity of piling required.Piles should be located in grid patterns relative toconcentrations in foundation forces, geotechnicalconsiderations, and pile-driving tolerances. The gridshould be established so that no pile interferes withanother, and such that no interference occurs amongpilings under adjacent monoliths or with sheetpilecutoffs. Computer program CPGI (Pile Group Inter-ference Check, CASE computer program X0086) canbe used to determine interferences. The grid shouldalso consider the effects of close pile spacing ondesign criteria, particularly for friction piles.

Generally speaking, for U-frame locks, greaterconcentrations of piling should be located beneath theheavier portions of the monolith such as the lockwalls and less dense concentrations beneath lighterareas like the chamber floor. Also, tension piles maybe needed at the center of the chamber floor to pro-vide resistance against uplift during maintenance orother dewatered conditions. Tension anchors couldalso be used in this regard, and the use of drains canhelp reduce uplift forces. Preliminary pile analyses atthis stage could be performed using CPGA (Hartmanet al. 1989) which assumes an infinitely rigid baseand allows for two- or three-dimensional analysis.For U-frame locks, the rigid base assumption is notnecessarily correct but is satisfactory for determina-tion of preliminary pile quantities for costing pur-poses. Preliminary pile layouts should be developedfor the major types of monoliths that comprise thelock so that accurate costs can be obtained. Calcula-tion of pile loads and some refinement in the layoutcan be accomplished using CWFRAME (Jordan andDawkins 1990). This program analyzes two-dimensional plane sections through the lock andaccounts for the flexibility in the base of the lockstructure. This will give a more accurate distributionof pile forces for a U-frame lock than will CPGA(Hartman et al. 1989). However, three-dimensionaleffects must be accounted for independently andadded to the two-dimensional CWFRAME (Jordanand Dawkins 1990) results. This flexible base behav-ior verification is particularly important for monolithsthat may require tension piling under the lock floor toresist uplift during dewatering.

d. Final pile layout and analysis. If the pilelayouts were properly developed during the feasibilityphase of the project, they can be used as a goodstarting point for development of the final layouts.Lastly, after an acceptable pile layout is determinedbased on assumed critical cases, all load cases shouldbe checked for effects on the pile layout and the baseslab bending.

e. Rigid versus flexible base. The designershould determine the relative rigidity of the lock basewith respect to the pile foundation. This can beaccomplished by running some simple parametricstudies in which the pile forces for a simplified flex-ural model are compared with the rigid base results.For U-frame locks, most monoliths should be ana-lyzed as flexible base structures unless it can beshown that the rigid base results closely approximatethe flexible base. Initial design estimates could

A-10

ETL 1110-2-35531 Dec 93

consider the base rigid, and the preliminary analysiscan be performed using a rigid base analysis tool,such as the computer program CPGA (Hartman et al.1989). This program performs the pile analysis fortwo- and three-dimensional loading conditions. Forsubsequent foundation design and any structural con-crete design, the pile cap should be treated as aflexible base. Therefore, the pile foundation andstructural analysis must be performed using a pro-gram that will consider the internal stiffness relation-ship of the structure. The same flexible base analysiscan be used to analyze the piles and the concretestructure. This is possible using the computer pro-gram CWFRAME (Jordan and Dawkins 1990) orother frame analysis or finite element applicationswhich include pile elements. A complete SSI finiteelement model can be useful for this analysis, thoughit is usually much more complicated.

f. Pile stiffness coefficients. Before beginningany detailed pile analysis, the pile stiffness coeffi-cients should be determined by performing single pileanalyses based on available soil data or previous piletests with similar soils and conditions. These coeffi-cients are essentially linear springs that approximatethe nonlinear behavior of the soil-pile foundation.Normally, it is desirable to perform a parametricanalysis where the bounds in variability of the coeffi-cients can be captured. Once determined, thesecoefficients are input to the various programs chosenfor the pile group analysis. See EM 1110-2-2906 fora further discussion of stiffness coefficients.

g. Lateral load resistance. Lateral load resis-tance in pile foundations is dependent upon the piletype, strong axis orientation and batter angle, and onthe assumed or experimental lateral subgrade moduliused in design.

(1) Pile orientation. Lateral loads are most effi-ciently resisted by battered piles. However, batteredpiles are more difficult to drive and result in a morecomplicated layout to design and construct. Addi-tionally, battered piles tend to dramatically change thepile force distribution. If the lateral loads are notsignificant, the designer should consider using allvertical piles. If the lateral load is significant, pileswith unequal stiffnesses about the orthogonal axes(H-piles for example) can be turned to increase stiff-ness in the direction of the load or they can be bat-tered. The preliminary batter slope and number ofbattered piles can be determined by using force vec-tors or similar methods. Capacities, limitations, and

suggestions for use of battered piles are defined inEM 1110-2-2906. In using battered piles, consider-ation must be given to geometric constraints fromadjacent pile-founded monoliths and sheetpile cutoffs.These constraints can be assessed using the computerprogram CPGI (Pile Group Interference Check, CASEcomputer program X0086). Generally, piles from onemonolith should not extend into the area beneath anadjacent monolith because of the possibility ofinterference.

(2) Pile head fixity. If it is not practical to usebattered piles to resist the lateral loads because ofgeometric constraints, all vertical piles may still be apossible solution, but lateral deflections may becomecritical. If lateral deflections are too high using verti-cal piles with a pinned condition at the pile head, thepile may be embedded deeper and analyzed as fixedat the pile cap. Pile head fixity is discussed inEM 1110-2-2906. Refer to Castella (1984) for moreinformation on pile head fixity.

(3) Lateral subgrade moduli. When an accept-able initial layout is achieved based on pile forcesand stresses, a comparison of calculated pile headdeflections to those seen in test results or assumed inthe pile stiffness coefficient analysis must be made.Since the pile/soil stiffness degrades with deflection,the calculated deflections seen in analyses shouldcompare with the deflections assumed or generated inthe selection of the pile stiffness coefficients.

6-3. Soil founded. The analysis of the foundationincludes checking for resultant location, sliding, uplift(flotation), differential settlement, and bearing failure.One of the single most important elements in thedesign of a soil-founded U-frame lock is the assump-tion regarding the distribution of the effective basepressure.

a. Pressure distribution. There are two basicapproaches used in determining the distribution ofbase pressures. One is a soil spring approach and theother is an assumed pressure distribution. The springmethod is discussed in paragraph 7-3e. The assumedpressure distribution approaches are a uniform distri-bution and a stepped distribution with appropriatecorrections for eccentric loading. The methods pre-sented have been derived from analysis of instrumen-tation data where base pressures were measured andcompared with conventional calculations. A toolavailable to compute base pressures is the CASEcomputer program 3DSAD (Tracy and Kling 1982).

A-11

ETL 1110-2-35531 Dec 93

The uniform pressure distribution for concentricloading is based upon the assumption that the baseslab of the monolith behaves as a rigid base. Tocompute the pressure distribution, the sum of allvertical forces acting on the base of the monolith isdistributed equally across the monolith bottom. Anexample of this computation is given in Figure A-2.This uniform pressure is modified to account forloads eccentric to the centroid of the monolith base.An example of this computation is given in Fig-ure A-3. The stepped pressure distribution is anapproximation based upon observed data from instru-mented U-frame locks founded on soil. In this distri-bution, the pressure beneath the lock wall is increasedto a set percentage of the pressure beneath theremainder of the monolith. These two pressures aremodified proportionately until the total pressureequals the sum of the vertical forces. Observed datafrom Port Allen Lock suggest the amount of increaseshould be 75 percent. An example of this computa-tion is given in Figure A-2. This stepped base pres-sure is modified to account for loads eccentric to thecentroid of the monolith base. An example of thiscomputation is given in Figure A-3. For furtherinformation on the amount of increase to use, seeSherman (1968) and (1972).

b. Location of resultant analysis. A resultantlocation analysis using unfactored loads should beperformed on each two-dimensional and three-dimensional monolith. The analysis consists of deter-mining the location of the resultant of all loading inrelation to the kern of the monolith base. The resul-tant location for usual load cases should be the mid-dle third of the base. The resultant location for theunusual load cases should be the middle half of thebase. The resultant location for the extreme casesshould be within the base. Usually the location ofthe resultant is not a problem.

c. Sliding analysis. A sliding analysis usingunfactored loads should be performed in accordancewith ETL 1110-2-256. The CASE computer programCSLIDE can perform this analysis (Pace 1987).

d. Bearing analysis. A bearing pressure analy-sis should be performed using unfactored loads. Thefoundation capacity should be developed taking intoaccount such items as soil type and stratification.The computed bearing pressures must be less than thefoundation capacity.

e. Flotation. A flotation analysis using unfac-tored loads should be performed on each monoliththat can be dewatered. For further information, seeETL 1110-2-307. Drag loads will not be used toresist uplift (flotation) due to the varying nature ofdrag loads. If insufficient capacity exists to keep themonolith from floating, the monolith can be helddown with anchors, heels, or more concrete mass, orimproved foundation drainage systems can be addedto the monolith.

f. Differential settlement. Differential settle-ment occurs between adjacent monoliths due to thedifference in size and weight of the monoliths as wellas differing foundation conditions beneath each mon-olith. Differential settlement should be held to thepractical minimum possible. There are several waysto handle this problem, including use of keys, dowels,and construction sequencing. Keys can be formedbetween adjacent monoliths. Dowels can be addedbetween the base slabs of adjacent monoliths. Theconstruction sequence for adjacent monoliths can bespecified such that the heavier monolith is partiallyplaced prior to placement of the lighter monolith.The magnitude of the forces being carried by dowelsor keys is difficult to predict, but the designer musttry to account for these forces by some rationalmethod.

7. Structural Analysis

Once the design criteria have been established, allreasonable load cases have been identified, and theinitial foundation parameters have been established,the analysis of the structure may be performed. Thestructural analysis is necessary for ensuring that thewall and slab thicknesses are sufficient and for deter-mining the reinforcement requirements of the struc-ture. Before performing an analysis of a U-framelock, the designer must decide whether each monolithbehaves in a two-dimensional or a three-dimensionalmanner. The method of analysis must also beselected, which can be a frame analysis, a finiteelement analysis, hand calculations, or a combinationof these. These decisions are based on experienceand good engineering judgment. A parametric studywhich bounds the extremes of behavior of a structurecan also be used as a tool to ensure adequacy of astructure. As a result of these analyses, the designercan then determine final member sizes andreinforcement.

A-12

ETL 1110-2-35531 Dec 93

Figure A-2. Bearing pressure - uniform

A-13

ETL 1110-2-35531 Dec 93

Figure A-3. Bearing pressure - stepped

A-14

ETL 1110-2-35531 Dec 93

7-1. Two-dimensional monoliths.

a. Lock monoliths that meet the followingrequirements can be considered two dimensional foranalysis purposes:

(1) The cross section of the monolith, transverseto the lock centerline, is constant or nearly constant.

(2) Loads acting on the monolith do not causesignificant overturning of the monolith in the direc-tion parallel to the lock centerline.

(3) Loads acting on the monolith do not causetorsion of the monolith. Torsion is considered to berotation about a vertical axis through the center ofgravity of the monolith.

Monoliths which meet these requirements can beanalyzed using a typical strip. Generally, chambermonoliths, and for some cases culvert valve mono-liths and intake/discharge, can be considered to acttwo dimensionally. In certain cases the loading andgeometry may be such that some of the aboverequirements are not completely satisfied but a two-dimensional analysis may still be used to accuratelymodel portions of the monolith.

b. Much of the analysis of U-frame locks canbe performed using frame analysis methods asdescribed below. For two-dimensional frame analysisno additional loads from adjacent monoliths/stripsshould be applied. Should investigation of a mono-lith indicate that a frame analysis is not adequate foranalysis of the structure (e.g., a monolith which haswalls with a low member length-to-depth ratio), afinite element analysis should be performed. Typi-cally, for a finite element analysis of a two-dimensional structure, a plane strain analysis shouldbe performed. Another application of finite elementanalysis for two-dimensional analysis is to calibrateand verify the results from frame analyses.

7-2. Three-dimensional monoliths.

a. Lock monoliths which do not meet therequirements for two-dimensional monoliths must beanalyzed as three-dimensional monoliths. Gate mon-oliths are usually considered to act three dimension-ally and in some cases intake/discharge monoliths arealso considered to act three dimensionally. Sinceactual three-dimensional modeling is not a commonpractice for most designers, analysis of

three-dimensional monoliths may be done byperforming several two-dimensional analyses whichcapture and envelop the three-dimensional behavior.

b. Typically, a three-dimensional analysis willbe performed using finite elements. A three-dimensional finite element analysis of any structure isa complicated technique. Caution is required whenperforming a three-dimensional analysis, and it shouldbe performed only by an engineer who is familiarwith finite elements and with the behavior of thestructure being analyzed. In most cases a three-dimensional finite element analysis is not requiredsince reasonable results can be obtained throughseveral two-dimensional approximations. However,these approximations also require structural designexperience, judgment, and insight.

c. If a frame analysis is used for analysis of athree-dimensional monolith, strips are modeled usinga plane frame with in-plane loads and shear loadstransferred from adjacent monoliths/strips. Strips in amiter gate monolith exhibit three-dimensional behav-ior due to increasing bearing pressure towards thedownstream end of the monolith. The increase is dueto vertical shears transferred between strips (seeparagraph 7-3b). Accounting for the shear transferredbetween the strips in the two-dimensional model isessential to obtain stresses which can be comparedwith three-dimensional results. If the shear transfer isnot properly accounted for within the two-dimensional model, then it is likely that the loads willbe unbalanced, particularly if the foundation pressuresor pile loads applied were obtained from a three-dimensional analysis.

7-3. Frame analysis.

a. General. Frame analysis is the most widelyused engineering tool for analyzing U-frame locksdue to its ease and speed of use. Most lock mono-liths have complicated geometry, but can be modeledas a linear elastic plane frame with the use of simpli-fying assumptions. The frames are analyzed usingCWFRAME (Jordan and Dawkins 1990), CFRAME(Hartman and Jobst 1983), or other programs. Typi-cally a representative strip is determined for analysis.

b. Strip selection.

(1) For pile-founded U-frame locks, strip selec-tion should consider pile spacing, layout pattern,stiffness, and batter. For soil-founded locks, a

A-15

ETL 1110-2-35531 Dec 93

1-ft-wide strip is usually sufficient. Strip selectionshould also consider blockouts that cause discontinu-ity or member property reduction in the structuralframing. Some strips will have tributary load. Anexample of these strips/sections occurs at the culvertvalve well. The walls of the well can be assumed toact as a plate fixed against rotation on three sideswith reactions on upstream and downstream sectionsand the base slab. Wall reactions are distributedthroughout the width of the upstream and downstreamsections. Vertical forces are assumed to be resistedby the foundation below the wall with no transfer toadjacent sections. The wall plate must be designed totransfer the load in the assumed direction (seeFigure A-4).

(2) All strips must be in static equilibrium.Strips in three-dimensional monoliths will have unbal-anced vertical loads due to the foundation pressuregradient. Strip equilibrium is achieved by usingvertical shears between strips with magnitudes asrequired to satisfy external equilibrium. This isreferred to as shear transfer. Shears should beapplied such that moments are not introduced into theexternal stability of the monolith. Shear transfer alsoprovides redundancy in the monolith which isrequired to distribute the effects of small discontinu-ities (from blockouts/voids) in the structural framing.This permits the designer to ignore small voids in theframe analysis. Generally within a monolith, thickreinforced concrete members establish shear transferwithout special details.

(3) A monolith may require several strip analysesand parametric studies in order for the designer tounderstand its behavior. Selecting a strip and inter-pretation of analysis results are challenging tasks andshould be assigned to more experienced engineers.

c. Frame member. Generally, framing ismodeled along member centerlines except for thevery deep member that forms the culvert roof inmonoliths such as the gate monoliths. This memberis modeled near the top of the culvert and intersectsthe culvert walls at their centerlines (see Figure A-4).The block above the culvert can be modeled as arigid body if its span-to-depth ratio is 1 to 1 or lower.If the member above the culvert is relatively thin, itbehaves like a typical frame member. Member sec-tion properties are computed using member grossconcrete dimensions.

d. Rigid links. Rigid links are short members atjoints that are stiffened to represent the behavior ofwide supports. They are used to approximate realbehavior at the intersections of thick concrete mem-bers. The length of the rigid link is generally half thedistance between the joint and the face of the sup-porting concrete. The length can be extended by halfthe length of a fillet if present. A link should have astiffness of at least ten times greater than that of theintersecting flexible member. In regions of complexgeometry, finite element runs can be used to calibratethe length of rigid links. For application of rigidlinks, see Figure A-5.

e. Foundation modeling.

(1) The foundation can usually be modeled byelastic springs (both vertical and horizontal) for soilor piles. Pile springs are attached to the base slabcenterline by rigid links which model the eccentricityto the foundation. The length of the link is generallyhalf the thickness of the base slab.

(2) Hydrostatic uplift is modeled as a load.Hydrostatic uplift reduces bearing pressures, whichaffects frame response.

(3) For soil-founded locks, assumed shapes ofbearing pressures can be modeled as a load in lieu ofusing foundation springs (see paragraph 6-3a). Sup-ports are still required to provide stable boundaryconditions, but each support reaction should be zero.Note that displacements of the soil must be compati-ble with the deflections of the structure in order toaccurately model the soil-structure interaction (seeparagraph 6-3b). The use of foundation springsaccommodates this requirement and is the preferredmethod of analysis.

(4) Pile foundation analyses should include hori-zontal base shears, particularly if battered piles arepresent. Horizontal base shear on battered pilescreates a vertical component of force that will loadthe U-frame. Torsional moments on a monolithcreate horizontal base shears and should be evaluated.

(5) Due to uneven distributions of foundationbearing pressures, differential settlement betweenmonoliths and within a monolith should be consid-ered. Within a monolith, usually the base slab isconstructed first, the subsequent load from buildup of

A-16

ETL 1110-2-35531 Dec 93

Figure A-4. Culvert valve monolith strip section (Continued)

A-17

ETL 1110-2-35531 Dec 93

Figure A-4. (Concluded)

the walls may locally compress the foundation whichwill induce moments into the base slab. If the wallsare constructed before the slab, generally only long-term differential settlement between the walls andslab needs to be considered. The second case createsa vertical construction joint at the intersection of thewall and base slab and should be used with discre-tion. This is a region of large shear and moment;therefore, the construction joint could be moved to aregion of lower shear and moment as determined byanalysis. Reinforcement splices must be coordinatedwith the location of the construction joint. Structural

performance (shear-friction and diagonal-tensionshear) and related detailing of the vertical joint shouldbe carefully considered. In all cases, designassumptions should be consistent with the method ofconstruction.

f. Variable thickness slabs.

(1) Base slabs may vary in thickness. The addedconcrete thickness can be designed and detailed to actcompositely with the rest of the base slab. Items toensure composite action are good construction joint

A-18

ETL 1110-2-35531 Dec 93

Figure A-5. Rigid link application

preparation, design for shear flow at the joint, and anadequate length of thickened section along the basemember (see Figure A-6). Some thickened sectionswill not have sufficient length to stiffen the base slab(see related discussion on cover plated steel beams inthe AISCManual of Steel Construction). Alterna-tively, the added concrete thickness can be detailed to

act independently of the rest of the base slab bysegmenting it with watertight joints parallel to flow.Joint spacing should be such that composite action isnot developed.

(2) Thicker slabs can stiffen adjacent thinnerslabs similar to the stiffening effect that a T-beam

A-19

ETL 1110-2-35531 Dec 93

Figure A-6. Variable thickness slab (Continued)

A-20

ETL 1110-2-35531 Dec 93

Figure A-6. (Concluded)

A-21

ETL 1110-2-35531 Dec 93

stem has on its flange. The stiffened width of thethinner slab depends on the specific geometry of thebase slab and the designer’s assumptions.

g. Location of design moments and shears.Joint flexural design moments at the face of support-ing concrete or toe of a fillet should be used. This isalso where reinforcement development length starts.Moments at these points can be found easily bymodeling a joint at the desired location. Maximummidspan moments are available from computer runsor can be found using free-body diagrams. Shearshould be checked, initially, at the face of the sup-porting concrete. Alternative locations for checkingshear, in compliance with American Concrete Insti-tute (ACI) 318, can be used if required.

7-4. Finite element analysis.

a. General. Finite element analysis is a numeri-cal method which can be used to determine thestresses, strains, and displacements. Finite elementanalysis can be performed in two or three dimensions.While the finite element method is a powerful anduseful tool, it must be used with care. Results fromfinite element analysis often appear accurate evenwhen the input is incorrect. The apparent accuracy offinite element results stems from the fact that resultsare often given to the fourth or fifth decimal place.To ensure results that are accurate, it is imperativethat finite element input data be thoroughly reviewedprior to proceeding with a design. Numerous textsare available on the subject and guidance for model-ing with finite elements is provided in ETL 1110-2-332. ETL 1110-2-332 should be reviewed by anydesigner who will be performing finite element analy-sis. In addition, for the designer not familiar withfinite elements, review of Will et al. (1987) is anexcellent example of how a novice should approachfinite element modeling. Finally, ETL 1110-2-254should be referenced for the purpose of documentingfinite element results.

b. Strip selection. Information on strip selectionfor two-dimensional monoliths can be found inparagraph 7-3b on strip selection for frames.

c. Boundary conditions. Boundary conditionsbecome very important when using finite elements.In the case of symmetrical structures, boundary con-ditions can be used to reduce the amount of input andoutput produced by an analysis. This is shown inFigure A-7 where the structure can be modeled with

half as many elements by taking advantage ofsymmetry. Boundary conditions can be used so thatonly a portion of the structure, such as a single wallor a portion of a wall, needs to be modeled asopposed to the entire structure, once again reducingthe input and output required (see Figure A-8). Thisaspect becomes very important when performing two-dimensional analyses on portions of a three-dimensional monolith. The designer should use carein the selection of the applied boundary conditions sothat behavior of the model represents similar behaviorof the real structure.

d. Foundation modeling. Modeling of the foun-dation can typically be accomplished through the useof elastic springs which may be computed from thepile stiffness coefficients. Some finite element pro-grams contain pile elements which may be used or itmay be possible to employ a soil-structure interaction(SSI) model. Close coordination with the geotechni-cal engineer is required when selecting the springconstants for the foundation, whether it be a pile-founded or a soil-founded structure. A pile-foundedstructure should be analyzed so that the piles carrythe entire load. If an SSI analysis is being per-formed, then provisions must be made so that thepiles carry the load. Finite element modeling of thefoundation is an option. In many cases it is not usedbecause the increased accuracy of the results is notimproved enough to justify the increased cost of theanalysis. If the foundation is modeled, guidelines fordeveloping the foundation mesh can be found inJones and Foster (in preparation).

e. Variable thickness slabs. Typically, variablethickness slabs occur only in gate monoliths, and theshapes that result in these monoliths can be verydiverse as seen by the cross sections shown inFigure A-9. Because of the shape of the base andbecause the loading on a miter gate monolith is threedimensional, a three-dimensional finite element analy-sis will likely provide the best available solution for avariable thickness slab. Due to the fact that a three-dimensional analysis is a difficult procedure, even forexperienced finite element users, simplified two-dimensional finite element analyses may be used todesign a base slab. Two-dimensional models whichmay be used to model a base slab using shell ele-ments are the sloped plate model, the stepped platemodel, or the offset beams model. These models aredescribed in the paragraphs below. Any of thesemodels may be used by the designer. Selection ofthe best method may depend on the specific geometry

A-22

ETL 1110-2-35531 Dec 93

Figure A-7. Use of symmetry in finite element modeling

A-23

ETL 1110-2-35531 Dec 93

A-24

ETL 1110-2-35531 Dec 93

Figure A-9. Possible cross-section shapes of variable thickness slabs

A-25

ETL 1110-2-35531 Dec 93

of the slab. In using any of the methodologies, alongwith engineering judgment, caution should be exer-cised when evaluating the results. In addition, thedesigner may wish to evaluate a variable thicknessslab using other methods, including multiple two-dimensional strips through both transverse and longi-tudinal sections.

(1) Sloped plate model. The sloped plate modeluses shell elements located at the centroids of thebase slab. Since the centroid of the thick portion ofthe slab is at a different location than the thin portionof the slab, a transition between the two planes ofelements is needed. This is accomplished by con-necting the two planes of elements with the first rowof elements in the thick portion of the base as seen in

Figure A-10. A disadvantage to using this model isthat the transition element is oriented unrealistically;therefore, results near this row of elements will beunreliable.

(2) Stepped plate model. Again, elements areplaced at the centroids of the base slab as in thesloped plate model and again require a transitionbetween the two planes of elements. As seen inFigure A-11, a set of vertical beams connects theplanes of elements which must transmit the forcesbetween the slab sections without introducing anyunrealistic stiffness to the model. The beams may beassigned arbitrary large values for the required sec-tion properties.

Figure A-10. Sloped plate model

A-26

ETL 1110-2-35531 Dec 93

Figure A-11. Stepped plate model

(3) Offset beams model. The elements for thismodel are all placed at the centroid of the thin sectionof the base slab. At the location where the slab isthicker, a grid of beams is added. The locations ofthe elements and beams are shown in Figure A-12.The elevation of these beams should be located at thecentroid of the additional thickness being modeled,and the section properties should also be computedbased on the tributary area of the additional thickness.In order to use this model, the finite code being usedmust be capable of offsetting the location of thebeams through a member eccentricity command sincethe beams are located at the same nodes as the plates.

(4) Single centroid model. The base slab maybe modeled using shell elements and assigning theelements within the model with different thicknesses.This will require some approximation since the

location of the centroids of the various portions of theslab will need to be placed at the same elevation,when in fact they are at different elevations.

f. Shear, moments, and thrusts (CSMT). Sinceoutput supplied by finite element programs is often inthe form of stresses and displacements, steps must betaken to convert the resulting stresses into moments,axial thrusts, and shears which can used for design.To assist in obtaining the necessary shears, moments,and thrusts needed from a finite element analysis, theprogram CSMT was developed. The program isdocumented in Huff et al. (1988). The user inputsstresses along a given line from the finite elementanalysis into the CSMT program, and the programcomputes the thrust and moment from the axial stressblock as well as the resulting shear on the sectionfrom the shear stress block. If the designer chooses,

A-27

ETL 1110-2-35531 Dec 93

Figure A-12. Offset beams model

these types of calculations can also be performed byhand, although this can become very tedious due tothe volume of data available.

7-5. Nonlinear, incremental structuralanalysis (NISA). A NISA is a finite element analy-sis which models the construction sequence of aconcrete structure from the time when its first lift isplaced up through when service loads are applied.Within that time frame an analysis provides resultswhich consider the changes in concrete temperaturedue to heat of hydration and to ambient conditions,the placement and removal of forms, the aging modu-lus of elasticity, creep, and shrinkage. Properties aredefined as a function of time, and the structure isincrementally constructed in the finite element model,simulating actual construction of the monolith.

Guidance for performing a NISA is contained inETL 1110-2-324. ETL 1110-2-324 requires a NISAto be performed on new types of massive concretestructures, structures which have exhibited unsatisfac-tory past performance, and when cost savings can beachieved through the performance of a NISA. Forthe design of a U-frame lock the major objective forperforming a NISA would be to achieve savingsthrough more cost-effective construction proceduresand concrete mixtures.

7-6. Seismic analysis.

a. Guidance. ER 1110-2-1806 mandates seis-mic design considerations for all Corps of Engineerscivil works projects and provides general guidanceand direction for seismic design and evaluation.

A-28

ETL 1110-2-35531 Dec 93

Seismic analysis can be accomplished in static(pseudostatic) or dynamic terms. Dynamic analysismethods can be separated into two types: responsespectrum analysis (RSA) and time history analysis(THA). Generally, initial designs should be based onpseudostatic analyses. Depending on seismologicalrecommendations, a RSA may be required during thepreparation of the design memorandums. A THAmay not be required and should not be done on apreliminary basis, but may be chosen to be performedduring the preparation of the design memorandums.With computational capabilities improving constantly,THA has become less formidable and can yield moreefficient designs, especially when used in combina-tion with a complete soil structure interaction model.

b. Geological and seismological investigations.The first step in performing a dynamic analysis is toobtain potential ground motion response through ageological and seismological evaluation at the site.This may yield actual site-specific motions or syn-thetic motions generated analytically based on site-specific geological data. Specific results from theseinvestigations should yield definitions of the opera-tional basis earthquake (OBE) and the maximumcredible earthquake (MCE) in terms of the peakground acceleration (PGA). Additionally, actual orsynthetic (or both) time histories and correspondingresponse spectra should be obtained.

c. Miscellaneous considerations.

(1) Damping. Foundation and structural dampingcoefficients are described in other guidance. Thereare normally different damping values used for theOBE and MCE conditions.

(2) Backfill modeling. Modeling of backfill onlock walls is a complex issue. Generally, in pseudo-static simplified analyses, traditional lateral earthcoefficient methods are used to compute backfillforces. For finite element models, modeling is nor-mally accomplished with linear springs attached tothe structure with stiffnesses based on the calculationof dynamic or pseudostatic backfill pressure. How-ever, in actuality, during an earthquake motion theearth pressure coefficients are varying from passive toactive values. Analytical models do not normallyhave the capacity for nonlinear springs, or if they do,they are analytically complex and computationallyexpensive. Therefore, engineering judgment on thevalue of the spring coefficients is required and must

be critically evaluated throughout the dynamic analy-sis process.

(3) Water modeling. Water loads due to seismicforces are normally modeled in dynamic analysesrelative to the Westergaard formulation of hydrody-namic forces. In pseudostatic analyses the hydrody-namic distribution of pressure can be applied as adistributed static load. In dynamic analyses thisdistribution is applied usually through the use ofadded mass attached at node locations along theperimeter of the water location. Sloshing must betaken into account. Usually calculated usingHousner’s method, sloshing is applied as a static loadin pseudostatic analyses and as an added mass indynamic analyses. Hydrodynamic water loads alsoaffect the miter gate reaction loadings on the walls.

8. Special Considerations

8-1. Monolith joints.

a. Independent monoliths. Generally, U-framelock monoliths are designed to act independently.Isolation simplifies the analysis and is a reliable basisfor predicting performance.

b. Interacting monoliths.

(1) For pile-founded locks, it may be necessaryfor adjacent monoliths to act together to resist appliedlateral loads. For example, resistance to the thrust onmiter gate monoliths could be supplemented by adja-cent monoliths through the use of proper joint detail-ing (see paragraph 9-7).

(2) For soil-founded locks, it may be necessaryto key or dowel the monoliths together to minimizedifferential settlement.

c. Adjacent structures.

(1) Details of connections and transmitted loadsfrom adjacent structures must be thoroughly investi-gated. Poor detailing at these connections couldresult in localized failures and/or serviceability prob-lems. Some areas to look at are: cofferdam tie-ins,abutting dam piers and their joint treatment, andguidewall tie-ins. The load for designing sheet piletie-in connections should consider the interlock forceof the piles as referenced in EM 1110-2-2503.

A-29

ETL 1110-2-35531 Dec 93

Watertight joints to adjacent dam piers will changethe hydrostatic loads applied to the lock.

(2) Adequate joint thicknesses between structurespermit them to maintain separation when deflectingunder load. Guidewalls can have either a dead loadreaction or impact reaction on the lock.

8-2. Seismic effects.

a. In general, a U-frame lock is inherently seis-mic resistant due to the lock symmetry (no torsion),monolithic shear wall and diaphragm action in voidareas, integral foundation mat, wall aspect ratio, andoverall rigidity. All of the above are benefitsachieved by good geometrical layout and properdetailing of reinforcing. The main seismic weaknessof a U-frame lock is the large base shears transferredto the limited foundation lateral load resisting system.

b. Structures adjacent to the lock should havesufficient separation (joint width) from the lock suchthat the two structures will not strike each other dur-ing a seismic event. Deflections used to size jointthickness are computed using concrete gross sectionproperties since the members are expected to remainin the elastic range. Cracked section properties willbe used if seismic analysis of the frame indicates anonlinear response.

c. The service bridge seats must have sufficientbearing length (pinned ends) to accommodate thelargest lateral movement experienced by the bridgepiers combined with a reasonable assumption for thethermal contraction of the bridge. Proper detailingwill ensure that the bridge remains seated during aseismic event.

d. Lateral loads above large voids must betransferred to the mass below the void and foundationusing interior walls as shear walls or by frame actionif shear walls do not exist (i.e., culverts and galler-ies). To ensure ductile frame action during seismicloading, typical ACI reinforcement details (ACI 318)should be reviewed for applicability. Contiguousconcrete, perpendicular to analysis strips, providesjoint confinement which will reduce spalling concrete,and therefore generally eliminate the need for specialseismic detailing.

e. It is necessary to check equipment anchoragefor the OBE since the lock must remain in operationduring and after this event.

8-3. Effects of voids. The use of voids or block-outs is acceptable. Voids and blockouts reduce theweight of the structure as well as the loads on foun-dation piling and/or foundation pressures. Voids orblockouts also reduce the amount of concrete needed.Using voids will increase the effort required to designand build the monoliths, but should result in lessexpensive structures due to reduced concrete requiredin the base and wall sections. Reducing the amountof wall concrete will reduce the top base bendingmoments, and possibly reduce the required basethickness. A means to remove seepage water fromthe void must be provided or the seepage waterweight must be included in the monolith design.

8-4. Foundation drains. Foundation drainsbeneath the lock and along the landwall of the lockare used to reduce the piezometric head from seepagefrom the upper pool to the lower pool. The founda-tion drains beneath the lock monoliths may be frenchdrains, consisting of either select sand or select sandwith filter drains. These drains are usually connectedto the lower pool with no control on backflooding.Drains along the lock landwall are used to reduce thehorizontal hydrostatic load acting against the lock-wall. Such drains consist of horizontal runs of well-screen or perforated pipes connected to vertical clean-outs and manholes. Means of preventing backflood-ing through the drains should be incorporated into thedrain design. If the drain is relatively deep in rela-tion to the height of the lockwall, it is recommendedthat stainless steel wellscreen and pipe be used for thehorizontal drain pipe. Reducing uplift duringdewatering by exiting foundation drains into the lockchamber and then removing the drainage with pumpsshould also be considered. The effectiveness ofdrains should be considered in analysis.

8-5. Instrumentation. Instrumenting the lockstructure and its foundation can serve two basic func-tions. Site personnel can monitor performance whilethe lock is in service, and design assumptions andparameters can be verified. Some examples of datathat can be accumulated include uplift and pore pres-sures, monolith tilt and alignment, cofferdam-cellmovements, concrete crack widths, and internal con-crete temperatures. More information on instrumenta-tion for structures can be found in EM 1110-2-4300,and more information on foundation instrumentationcan be found in EM 1110-2-1908. Many types ofinstrumentation exist which serve different purposes,but all forms require planning for both design andconstruction. Consideration of instrument installation

A-30

ETL 1110-2-35531 Dec 93

during the construction process is crucial for thesuccessful long-term operation of the equipment andto minimize disturbances to construction.

8-6. Silt. Silt accumulations upstream of mitergates and in the lock chamber may become an opera-tional problem. Large accumulations of silt canrestrict the free movement of the miter gates. Thereare several methods available for removing silt fromaround miter gates. Blockouts may be used upstreamof the gates to provide more area for silt to accumu-late in before it must be removed. A silt flush sys-tem using water may be installed to resuspend the siltin the water and move it out of the gates’ path.Small-diameter pipes can be placed in the lower gatesill to provide for flushing action around the lowermiter gates.

9. Constructability Considerations

9-1. General. A U-frame lock design includesdetails that ensure structural performance and can beaccurately detailed, bid, and constructed. Gooddetailing will be rewarded by reducing contractmodifications and reducing engineering effort wheninterpreting the plans for field personnel and thecontractor. There are some areas that are trouble-some, and these will be briefly discussed in the fol-lowing paragraphs.

9-2. Construction sequence for monoliths.

a. Typically two monoliths will be constructedseparated by a space for a third monolith. The mid-dle monolith will use the end monoliths as formworkand as supports to which joint material can beattached. Expansion joint material must not be com-pressed by the fluid force of fresh concrete, but mustbe compressible to accommodate thermal expansionof the monoliths.

b. Construction sequencing of the monolithsshould be to place the deeper founded monoliths first.This eliminates potential undermining and loss offoundation confining soil during excavation if themore shallow monolith were placed first. Backfillingof overexcavation for the deeper founded monolithsshould be done prior to placing concrete for the adja-cent monolith.

c. The general shape of a monolith’s bearingpressure diagram is characteristic for differentialsettlement between the walls and the base slab. Con-crete placement sequence also affects the differentialsettlement of the monolith.

9-3. Reinforcement placement. Proper rein-forcement detailing will simplify congested areas,ensure ease of reinforcement placement, and facilitatethe placement of concrete.

a. Congestion. (See reference ACI 309.3R forrelated discussion.)

(1) U-frame locks generally have small amountsof reinforcement in relationship to the volume ofconcrete; however, reinforcing is concentrated at theconcrete faces. Reinforcement layering and bundlingof bars can reduce congestion; however, additionallayers of horizontal reinforcement can create difficul-ties in the placing of concrete and bundled bars havecharacteristics (ACI 318) that may make themundesirable.

(2) Possibly the most congested area of rein-forcement is at the intersection of the base slab andthe lock wall (particularly for culvert intake anddischarge manifolds). At this location, wall dowelsintersect and may conflict with the layers of base slabreinforcement. Generally, main structural reinforce-ment should be located first and less important rein-forcement spaced to eliminate interference. Forexample, the placement of vertical reinforcement inthe lock wall may be dependent upon missing voidsor embedded items encountered higher up in the wall.It follows that base slab reinforcement should bedetailed, to scale, to avoid the vertical lock wallreinforcement. Other items which may contribute tocongestion are heavy structural steel for support ofreinforcement, electrical conduits, instrumentation,embedded metals, heavy reinforcement requirementsat ports/manifolds, and formwork supports.

(3) The contractor may opt to set the ends ofwall vertical bars at the top of concrete lifts insteadof holding them up at elevations as specified on thedrawings. Although this is generally allowed, theadditional reinforcement may add to the congestion.

A-31

ETL 1110-2-35531 Dec 93

(4) Congestion and other construction problemsmay become evident if reinforcement splices anddevelopment lengths are drawn to scale on the con-tract plans.

b. Splicing.

(1) In general, the use of splices should be mini-mized, and all contractor-added splices should becarefully reviewed to ensure that they do notadversely affect structural performance. All splicelocations and splice staggers should be in compliancewith the latest ACI 318 recommendations and shownon the contract plans or indicated in the specifica-tions. In determining splice locations, the designercan use the normal maximum fabricated length of60 ft for horizontal bars. Forty feet is usually used indetailing the maximum length for vertical reinforce-ment because of constructability limitations.

(2) Numerous load cases cause inflection pointsand regions of high moments to vary. The designershould consider this phenomenon when locatingsplices.

(3) U-frame locks require #14 and #18 bars toresist large bending moments. Large reinforcing barsare difficult to fabricate, ship, handle, place, andsupport. Large reinforcing bars also require mechani-cal splices. Mechanical splices are difficult toinspect, the coupler adds to the congestion, and thepossibility of installation error is greater than for alap splice. Large reinforcement bars should be usedonly where required.

c. Bending.

(1) Bends in reinforcement are usually made forstandard hooks, corner bars, and bar terminations.Large bars have large bend radii that often interferewith the placement of intersecting steel (seeFigure A-13). It is suggested that all bar bends bedrawn to scale on the plan drawings to help identifyand correct reinforcing placement problems. Hori-zontal #14 and #18 bars with bends will cause prob-lems since they will be tied to vertical bars set inhardened concrete that may have been placed withoutconsidering the large bend radii of the horizontal bar.

As a result, concrete clearances will be sacrificed (seeFigure A-14). A possible solution is to detail verticalbars along the bend of the large horizontal bars.

(2) Hooked #14 and #18 horizontal bars havelong extension lengths at their free ends (3 ft-5 in. for#18 bars). If the free end protrudes a small distancefrom the upper lift into a lower lift, it is permissibleto rotate the end from the vertical position until thefree end is out of the lower lift. This allows thecontractor to place the bar on top of the lower liftafter it has hardened. Alternatively, the free end canremain vertical and the lower lift can be blocked outto receive it (see Figure A-15). This minimizes thenumber of reinforcing mats that the contractor has tosupport during a concrete placement. The option ofrotating the free end of the hook or blocking out canbe given to the contractor with reference to a note onthe contract plans.

d. Reinforcement and waterstops. Waterstopswill not be omitted or punctured so that the reinforce-ment can run its normal path. Adding concrete coverover the reinforcing bars or detailing reinforcementaround the waterstop in the initial design caneliminate this problem (see Figure A-16).

e. Geometric discontinuities. At blockouts (i.e.,recess in a wall), flexural reinforcement must beterminated, usually in a standard hook, and structuralcontinuity reestablished by placing additional rein-forcement under or to the sides of the blockout.Added reinforcement should be developed past eachend of the blockout. Added reinforcement under theblockout probably requires widely spaced (larger than6 in., ACI 318) noncontact lap splices. Additionalreinforcing, such as development length past thepotential crack zone (ACI 318) or stirrups to controlcracking, may be required (see Figure A-17).

9-4. Fillets.

a. Fillets are used in U-frame locks for manyreasons and in many locations. Fillets from the floorto the wall in the culvert have many advantages anddisadvantages.

(1) Advantages. Fillets can reduce honeycomb-ing in the main structural members, and permit theuse of the lower design moments and shears whichoccur at the toe of the fillet.

(2) Disadvantages. Disadvantages include:planes of weakness in the construction joints of thefillet, suspended forms for the fillet and culvert floor

A-32

ETL 1110-2-35531 Dec 93

Figure A-13. Reinforcement interferences in base slabs

A-33

ETL 1110-2-35531 Dec 93

Figure A-14. Reinforcement interferences in lock walls

A-34

ETL 1110-2-35531 Dec 93

Figure A-15. Solutions for reinforcement placement problems

A-35

ETL 1110-2-35531 Dec 93

Figure A-16. Reinforcement and waterstop interferences

are required to eliminate planes of weakness, suscep-tible to honeycombing in the toe of the fillet, filletsare discontinuous at lock filling and emptying ports.Due to the many disadvantages, the floor-to-wall filletis not recommended. Fillets at the roof of the culvertare less of a problem and can be advantageous.

b. The designer should detail the constructionjoint at the wall to the base slab at the level of theculvert invert. The joint should be cleaned, possiblyroughened, and reinforced to ensure an adequateconnection. This joint location also facilitates thefinishing of the culvert invert and base slab (seeFigure A-18).

c. In summary, the potential benefits from afillet and its probable method of construction shouldbe reviewed by the designer before using it in analy-sis and design.

9-5. Construction/lift joints. Selection of liftjoints should be closely coordinated with the mater-ials engineer and a representative of the constructiondivision. Lift heights in base slabs of U-frame locksusually do not exceed 5 ft while lift heights in thewalls are typically 5 to 10 ft high but can exceed15 ft in the culvert walls. Typically any changes tooptimize originally selected lift heights will be

A-36

ETL 1110-2-35531 Dec 93

Figure A-17. Reinforcement at geometric discontinuities

determined during the course of a NISA. Whenconsidering the lift heights in the wall, a constructionjoint should be placed at the top and bottom of allvoids. Other changes in geometry are also placeswhere lift joints should be located. The lift heightsselected should be made as consistent as possible onall monoliths to allow the contractor to use one set offorms in a number of different locations. For largeslab placements where vertical construction joints arenecessary, efforts should be made to locate thesejoints in the lowest stressed areas of the slab. In

addition, if tension is expected to occur across thesejoints, then appropriate measures should be taken toprepare these joints with reinforcing dowels beingplaced across them.

9-6. Monolith joints (monolith length). Thelength of monoliths is determined by evaluation ofconstructability, temperature effects, and cost. Lengthof monoliths will generally range from 50 ft to over100 ft.

A-37

ETL 1110-2-35531 Dec 93

Figure A-18. Varying elevations of construction joints and concrete finish floor elevations

a. Constructability. One of the primary factorswhen determining monolith length is the capacity ofthe concrete batch plant to be used at the project.Generally this factor is resolved by the materials andconstruction engineers. The location of the culvertports and their spacing must be accounted for whendetermining the length of a monolith. Monolith spac-ing should be arranged such that a monolith joint isapproximately half way between culvert ports. Simi-larly, accommodations for instrumentation recesses

and their spacing should be considered prior to final-izing the length of a monolith.

b. Temperature effects. The length of a mono-lith may be limited by the effects of temperature.Generally, the longer a monolith is, the higher thestresses in the longitudinal direction. Determinationof a suitable monolith length with respect to tempera-ture can typically be made from experience and theperformance of a NISA.

A-38

ETL 1110-2-35531 Dec 93

c. Cost. Increasing the lengths of monoliths ina U-frame lock is generally considered to be a cost-saving measure. Increasing the length of monolithswill reduce forming costs and the number of joints toreceive preparation. Another item which should beconsidered with respect to cost is the length of thereinforcing bars. Since reinforcing bars are generallyshipped in lengths of 60 ft, a monolith slightly longerthan 60 ft may not be cost effective due to splicingwhich would be required.

9-7. Joint treatment/detailing.

a. Independent monolith action is ensured byproper detailing at monolith joints. The joint thick-ness should consider the monolith deflections towardeach other and the compressibility of the jointmaterial.

b. The base slabs of adjacent monoliths may berequired to act together to resist shear, tension, and/orcompression (see paragraph 8-1b). This is done bydoweling the bases together to obtain a shear-frictionconnection or by the use of shear keys. Tensionbetween base slabs is resisted by dowels. Compres-sion is resisted by concrete bearing. Joint treatment

between monoliths should be compatible with thetype of force(s) transmitted across the joint. A water-stop at the joint between monoliths will be requiredto stop the transmission of foundation materialthrough the joint. Walls of monoliths can be detailedand analyzed to act independently if their base slabsare doweled together.

c. Further discussion on joint treatment/detailingis found in other guidance.

10. Specifications and Details

The items discussed herein address the analysis anddesign of U-frame lock monoliths without discussingthe specifications and details required to complete adesign. While not addressed specifically in this guid-ance, proper attention to the specifications and detailsis a vital part of the complete design process. Guidespecifications are available for use by the designerand should be studied carefully and modified as nec-essary for each project. Good details are an essentialelement in producing a quality product. Designersshould draw on past experience and review otheravailable guidance.

A-39

ETL 1110-2-35531 Dec 93

APPENDIX B: REFERENCES

ER 1110-2-1150Engineering and Design for Civil Works Projects

ER 1110-2-1806Earthquake Design and Analysis for Corps of Engi-neers Projects

EM 1110-1-1905Bearing Capacity of Soils

EM 1110-2-1610Hydraulic Design of Lock Culvert Valves

EM 1110-2-1612Ice Engineering

EM 1110-2-1908Instrumentation of Earth and Rock Fill Dams, Parts 1and 2

EM 1110-2-2102Waterstops and Other Joint Materials

EM 1110-2-2104Strength Design for Reinforced-Concrete HydraulicStructures

EM 1110-2-2105Design of Hydraulic Steel Structures

EM 1110-2-2200Gravity Dam Design

EM 1110-2-2502Retaining and Flood Walls

EM 1110-2-2503Design of Sheet Pile Cellular Structures Cofferdamsand Retaining Structures

EM 1110-2-2703Lock Gates and Operating Equipment

EM 1110-2-2906Design of Pile Foundations

EM 1110-2-4300Instrumentation for Concrete Structures

ETL 1110-2-247Lock Wall Accessories

ETL 1110-2-254Finite Element Analysis - Interpretation and Docu-mentation Guidelines

ETL 1110-2-256Sliding Stability for Concrete Structures

ETL 1110-2-295Ice Control on Miter Gate Arms

ETL 1110-2-307Flotation Stability Criteria for Concrete HydraulicStructures

ETL 1110-2-320Methods to Reduce Ice Accumulation on Miter GateRecess Walls

ETL 1110-2-324Special Design Provisions for Massive ConcreteStructures

ETL 1110-2-332Modeling of Structures for Linear Elastic FiniteElement Analysis

ETL 1110-2-338Barge Impact Analysis

ETL 1110-8-13(FR)Structural Engineering Responsibilities for CivilWorks Projects

American Concrete Institute 1987American Concrete Institute. 1987. “Building CodeRequirements for Reinforced Concrete,” ACI 318R-87, Detroit, MI.

American Institute of Steel Construction 1980American Institute of Steel Construction. 1980.Manual of Steel Construction, 8th ed., Chicago, IL.

Castilla 1984Castilla, F. 1984. “Fixity of Members Embedded in

B-1

ETL 1110-2-35531 Dec 93

Concrete,” Technical Report M-339, ConstructionEngineering Research Laboratory, Champaign, IL.

Clough and Duncan 1969Clough, G. W., and Duncan, J. M. 1969. “FiniteElement Analysis of Port Allen and Old RiverLocks,” Contract Report S-69-6, U.S. Army EngineerWaterways Experiment Station, Vicksburg, MS.

Ebeling, Clough, Duncan, and Brandon 1992Ebeling, R. M., Clough, G. W., Duncan, J. M., andBrandon, T. L. 1992. “Methods of Evaluating theStability and Safety of Gravity Earth Retaining Struc-tures Founded on Rock,” Technical Report REMRCS-29, U.S. Engineer Waterways Experiment Station,Vicksburg, MS.

Ebeling, Duncan, and Clough 1990Ebeling, R. M., Duncan, J. M., and Clough, G. W.1990. “Methods of Evaluating the Stability andSafety of Gravity Earth Retaining Structures Foundedon Rock-Phase 2 Study,” Technical Report ITL-90-7,U.S. Army Engineer Waterways Experiment Station,Vicksburg, MS.

Ebeling, Mosher, Abraham, and Peters 1993Ebeling, R. M., Mosher, R. M., Abraham, K., andPeters, J. F. 1993. “Soil-Structure Interaction Studyof Red River Lock and Dam No. 1 subjected to Sedi-ment Loading,” Technical Report ITL-93-xx,U.S. Army Engineer Waterways Experiment Station,Vicksburg, MS.

Ebeling, Peters, and Clough 1992Ebeling, R. M., Peters, J., and Clough, G. W. 1992.“Users Guide for the Incremental Construction, Soil-Structure Interaction Program SOILSTRUCT,”Technical Report ITL-90-6, U.S. Army EngineerWaterways Experiment Station.

Hartman and Jobst 1983Hartman, J. P., and Jobst, J. J. 1983. “User’s Guide:Computer Program with Interactive Graphics forAnalysis of Plane Frame Structures (CFRAME),”Instruction Report K-83-1, U.S. Army EngineerWaterways Experiment Station, Vicksburg, MS.

Hartman, Jaeger, Jobst, and Martin 1989Hartman, J. P., Jaeger, J. J., Jobst, J. J., and Martin,D. K. 1989. “User’s Guide: Pile Group Analysis(CPGA) Computer Program,” Technical Report ITL-89-3, U.S. Army Engineer Waterways ExperimentStation, Vicksburg, MS.

Huff et al. 1988Huff, D., et al. 1988. “User’s Guide for RevisedComputer Program to Calculate Shear, Moment, andThrust (CSMT),” Instruction Report ITL-88-4,U.S. Army Engineer Waterways Experiment Station,Vicksburg, MS.

Jones and Foster (in preparation)Jones, H. W., and Foster, J. (in preparation). “FiniteElement Method Guidance for the Analysis of Grav-ity Dam Structures, Foundation Effects - Phase Ib,”U.S. Army Engineer Waterways Experiment Station,Vicksburg, MS.

Jordan and Dawkins 1990Jordan, T. D., and Dawkins, W. P. 1990. “User’sGuide: Computer Program for Two-DimensionalAnalysis of U-Frame of W-Frame Structures(CWFRAM),” Instruction Report ITL-90-6,U.S. Army Engineer Waterways Experiment Station,Vicksburg, MS.

Pace 1987Pace, M. E. 1987. “Sliding Stability of ConcreteStructures (CSLIDE),” Instruction Report ITL-87-5,U.S. Army Engineer Waterways Experiment Station,Vicksburg, MS.

Sherman 1968Sherman, W. C. 1968. “Analysis of Data fromInstrumentation Program, Port Allen Lock,” TechnicalReport S-68-7, U.S. Army Engineer WaterwaysExperiment Station, Vicksburg, MS.

Sherman 1972Sherman, W. C. 1972. “Analysis of Data fromInstrumentation Program, Old River Lock,” TechnicalReport S-72-10, U.S. Army Engineer WaterwaysExperiment Station, Vicksburg, MS.

B-2

ETL 1110-2-35531 Dec 93

Tracy and Kling 1982Tracy, F. T., and Kling, C. W. 1982. “A Three-Dimensional Stability Analysis/Design Program(3DSAD), General Analysis Module,” InsructionReport K-80-4, Report 3, U.S. Army EngineerWaterways Experiment Station, Vicksburg, MS.

Will et al. 1987Will, K. M., et al. 1987. “Procedure for StaticAnalysis of Gravity Dams Using the Finite ElementMethod - Phase Ia,” Technical Report ITL-87-8,U.S. Army Engineer Waterways Experiment Station,Vicksburg, MS.

B-3