A publication of the James F. Lincoln Arc Welding · PDF file · 2011-02-17Since 1936, the James F. Lincoln Arc Welding Foundation ... Sales and Marketing, The Lincoln Electric Company

A publication of the James F. Lincoln Arc Welding Foundation

Volume XX, Number 1, 2003

Donut Consumption 101

Australia and New ZealandRaymond K. RyanPhone: 61-2-4862-3839Fax: 61-2-4862-3840

CroatiaProf. Dr. Slobodan KraljPhone: 385-1-61-68-222Fax: 385-1-61-56-940

RussiaDr. Vladimir P. YatsenkoPhone: 077-095-737-62-83Fax: 077-093-737-62-87

INTERNATIONAL SECRETARIES

The manufacturing, construction and miningindustries that use welding technologytogether account for about one-third of theU.S. Gross Domestic Product, yet most companies in these fields have never done a thorough study of their welding costs, norhave they ever evaluated the contribution of welding to their complete manufacturingprocess. The preceding statements aremade and supported in a comprehensivestudy* published jointly last year by theAmerican Welding Society and the EdisonWelding Institute.

So fundamentally, when it comes to welding, most manu-facturers don’t know how much they’re spending, or whatthey’re actually spending it on...or even why. That’s the badnews. Now for the good news: the same study found thatcompanies with a solid understanding of the value weldingcan add to the manufacturing process, as well as a firmgrasp of welding economics, are able to compete success-fully, both nationally and globally.

Probably most readers of this publication know the firstrule of welding economics—that in the U.S., laboraccounts for over 70 percent of total welding expenditures.If welding costs are compared to a donut, by attempting tochip away at equipment and consumable costs, manufac-turers are focusing on the hole. The only way to trulyreduce costs is to take a big bite of the donut itself—thelabor costs. This is done by raising productivity.

Three Ways to Dunk the DonutOpportunities for increasing welding productivity fall intothree broad categories: automation, design and education.

The AWS/EWI study cites automation as the first route tohigher productivity, and yet it states, “nearly 60% of allfirms reported no effort to actively pursue the automationof welding processes.” It observes that most companieswait for industry leaders to take the capital risks ofautomating, and then only gradually, after seeing provenresults, adopt automated methods.

Considering welding requirements should be fundamentalto every phase of designing a structure or product. AsWelding Innovation design consultant Omer Blodgett says,“If the engineer makes the mistake of considering weldingto be just another type of fastener, the item or structure asdesigned will fall far short of its potential capabilities.

Welding is not a fastener; it is method of designwhich, properly used, takes full advantage of theversatility of the material.”

Qualified welding personnel are in chronically shortsupply and almost half of the firms responding tothe study said their welding-related training needsare not being met. Manufacturers cited shortagesof qualified personnel and a lack of advanced weld-ing education programs at every level of the field,from apprentice welder to engineer.

Taking the First Few BitesSince 1936, the James F. Lincoln Arc Welding Foundationhas been dedicated to enlarging the market for welding byrewarding achievement and sharing technical knowledge.Therefore, we enthusiastically endorse the recommendationsin the AWS/EWI study, especially proposed efforts to:• Develop procedures to help companies understand the

economics of adding value by raising welding productivity.• Identify and pursue improved educational opportunities in

the field of welding at local, state, and national levels.• Coordinate efforts to share knowledge of productive weld-

ing practices between and among different industries.

Regular readers of Welding Innovation appreciate the factthat many of these objectives are addressed in the pages ofthis magazine, through such columns as “Lessons Learned inthe Field,” “Design File,” and feature stories profiling a broadrange of exceptional projects. So if we’re preaching to thechoir, please pass this magazine on to someone whom youthink might need to hear the message.

Now, Wake Up and Smell the CoffeeWhen times are good and business is booming, it seemswe’re always too busy to focus on innovation. And when busi-ness is bad, we lack cash to invest in anything that doesn’tpromise an immediate return. But when it comes to innova-tion, there is no time like the present. Understand the trueeconomic issues, grasp the fundamental design concepts,and begin a serious study of automation. Do it now.

Richard D. Seif, Vice President, Sales and Marketing, The Lincoln Electric Company

* Welding-Related Expenditures, Investments, and ProductivityMeasurement in US. Manufacturing, Construction, and MiningIndustries, May 2002.

1Welding Innovation Vol. XX, No. 1, 2003

Cover: The William Jefferson ClintonPresidential Center in Little Rock,Arkansas features two massive paralleltrusses that were fabricated in individualpieces and then assembled in the air, in the vertical position. Photo: ArkansasAerial Photography. Story on page 17.

The views and opinionsexpressed in WeldingInnovation do not neces-sarily represent those ofThe James F. Lincoln ArcWelding Foundation or TheLincoln Electric Company.

The serviceability of aproduct or structure utiliz-ing the type of informationpresented herein is, andmust be, the sole responsi-bility of the builder/user.Many variables beyond thecontrol of The James F.Lincoln Arc WeldingFoundation or The LincolnElectric Company affect theresults obtained in applyingthis type of information.These variables include,but are not limited to, weld-ing procedure, plate chem-istry and temperature,weldment design, fabrica-tion methods, and servicerequirements.

Volume XXNumber 1, 2003

EditorDuane K. Miller,

Sc.D., P.E.

The James F. Lincoln Arc Welding Foundation

Omer W. Blodgett, Sc.D., P.E.Design Consultant

Features

Departments

Visit us online at www.WeldingInnovation.com

THE JAMES F. LINCOLN ARC WELDING FOUNDATION

Dr. Donald N. Zwiep, ChairmanOrange City, Iowa

John Twyble, TrusteeMosman, NSW, Australia

Roy L. MorrowPresident

Duane K. Miller, Sc.D., P.E., Trustee, Cleveland Ohio

Carl PetersExecutive Director

7 Opportunities: Lincoln Electric Technical Programs

8 Design File: Pay Attention to Tack and Temporary Welds

13 Opportunities: Lincoln Electric Professional Programs

14 Lessons Learned in the Field: Understanding Distortion

is a Never Ending Challenge

2 Repair and Maintenance Procedures for Heavy Machinery ComponentsThe hardfacing process is a cost-effective tool that can minimize wear and increase service life of heavy machinery components.

17 Making the Essential Connections on "A Bridge to the 21st Century"Fabrication and erection of the $160 million Clinton Presidential Library represents the value of meticulous planning, coordination and precision welding.

www.weldinginnovation.com

2 Welding Innovation Vol. XX, No. 1, 2003

Repair and Maintenance Proceduresfor Heavy Machinery Components

A version of this paper was publishedat the 50th WTIA Annual Conferenceheld in Sydney, Australia, 26-30August 2002.

Introduction

Heavy machinery components aresubjected to severe destructive condi-tions of environmental wear. The hardfacing process is a cost-effectivetool that can minimize wear andincrease service life of heavy machinerycomponents.

Types of Wear

The OECD (Organisation forEconomic Cooperation andDevelopment) defines wear as: “Theprogressive loss of substance from theoperating surface of a body occurringas a result of relative motion at thesurface” [1]. Commonly recognizedwear categories and their respectiveestimated shares of heavy machinerywear [2] are shown in Figure 1.

Usually, there are several wear mech-anisms that act simultaneously onheavy machinery components. Thetwo most common types are abrasiveand metal to metal wear.

Metal to Metal Wear

Metal to metal wear occurs when twometallic surfaces slide against eachother under the pressure. True metalto metal wear is the most often foundunder nonlubricated or dry conditions.Archard’s Metal to Metal Theory hasbeen widely accepted since the rela-tionship established between the wearvolume (V), sliding distance (L), nor-mal load (N) and hardness (H) is con-sistent with experimentally observedresults:

V=(KxLxN)/H (1)K is coefficient of wear.

When shear stresses overcome thecohesive strength of the metal matrix,cracks and voids can be nucleatedand wear particles can form [4].

Abrasive Wear

Abrasive wear occurs when non-metallic materials slide or roll, underpressure, across a metallic surface.This type of wear is determined by:• The properties of the wear material,• The properties of the abrasive

material, and• The nature and severity of the

interaction between the abrasiveand wear material.

Abrasive wear can be classified as (a)gouging abrasion, (b) high stressgrinding abrasion and (c) low stressscratching abrasion or erosion. Inabrasive wear, there are two extrememechanisms of material removal, onein which plastic deformation plays adominant role, and the other in whichfracture with limited plastic deforma-tion dominates. According to the sim-plified abrasion wear theory, equation2, volume loss, Q, is proportional tothe applied load (N) and is inverselyproportional to the hardness (H) of theabraded surface [5].

Q = N/H (2)

Figure 3 illustrates the mechanism ofabrasive wear.

Figure 1. Illustration of the ratio ofdifferent wear categories in industry.

Figure 2. Alloy content as a function of wear. [3]

By Milo DumovicManager, Welding Technology CentreThe Lincoln Electric Company Australia

Impact16%

Corrosive10%

Metalto

Metal22%

Abrasive52%

Welding Innovation Vol. XX, No. 1, 2003 3

Figure 3. (a) Idealized representation of abrasive wear resulting from mechanicalapplication of force to an abrasive particle. (b) Idealized representation of abrasivewear resulting from kinetic application of force to an abrasive particle.

Table 2. Engineering methods for surface hardening of steel. [7]

Table 3. Hardnesses of the most common materials used in hardfacing [8]

Table 1. Typical properties of selectedmaterials. [6]

Impact Wear During service, heavy equipment components can also be subjected toimpact wear. Toughness can be regard-ed as the capacity of a material toabsorb energy by deforming plasticallybefore fracture. Toughness is indepen-dent of the strength and ductility of thematerial and is measured by Charpyand Izod tests. Table 1 gives the impacttoughness of selected materials.

Hardfacing

Hardfacing is a surfacing process usedto improve the wear resistance ofheavy machinery components withoutaffecting the interior of the component.Hardfacing is a process of applying, bywelding, a layer, edge or point of wear

resistant metal onto a metal compo-nent. Table 2 illustrates engineeringmethods for surface treatment of steel.

Selection of Hardfacing WiresSelection of hardfacing wires is based on:• The wear mechanism acting on the

component;• Tribological conditions: load, temper-

ature and impact;• Comparison with prior experience;• Compatibility with substrate materials;• Requirements for heat treatment

and machining after welding;• Availability of materials, equipment

and skilled personnel; and• Cost.

Table 3 gives the hardnesses of themost common materials used in hard-facing wires.

(a) (b)

LAYER ADDITIONS

HARDSURFACING Fusion hardfacing (welded overlays)Thermal spray (bonded overlay)

COATINGSElectrochemical platingChemical vapour depositionThin films (physical vapour deposition)Ion mixing

SUBSTRATE TREATMENT

DIFFUSION PROCESS CarburisingNitridingCarbonitridingNitrocarburisingBoridingTitanium-carbon diffusionToyota diffusion process

SELECTIVE HARDENING METHODS Flame hardeningInduction hardeningLaser hardeningElectron beam hardeningIon implantationSelective carburising and nitriding

Austenitic (11-13%Mn) Steel 200-250 140

Austenitic (6% Mn) Steel 200-250 30

Cast Martensitic Steel 400-600 15-25

Wrought Martensitic Steel 300-550 20-70

Cast Pearlitic Steel 250-420 5-10

Alloy White Cast Irons 600-900 2-5

ImpactToughness

(J)Material Hardness

HVMaterial Formula Hardness HV

Ferrite Alpha-Fe 70 – 200 Pearlite (nonalloyed) Alpha Fe + Fe3C 250 – 320Pearlite (alloyed) Alpha Fe + Fe3C 300 – 460Austenite Cr- alloyed Gamma- Fe 300 – 600Austenite low alloyed Gamma- Fe 250 – 350Nickel Ni 560Bainite Alpha Fe + Fe3C 250 – 450Martensite Alpha Fe + Fe3C 500 – 1010Cementite Fe3C 840 – 1100Chromium Carbide CrxCy 1330 – 1700Titanium Nitride TiN 1800Tungsten Carbide WC 1900 – 2000Vanadium Carbide VC 2300Titanium Carbide TiC 2500Boron Carbide B4C 2800


Design and Selection ofHardfacing Consumables

The design and selection of weldingconsumables for build-up and wear-resistance applications is based on thefollowing principles:• Addition of carbon;• Addition of alloys;• Providing hard particles in a soft

weld metal matrix.

Table 4 shows the influence of alloyingelements on the properties of welddeposits.

The influence of carbon and percent-age of martensite (cooling rate) on thehardness of steel weld metal deposits[7] is shown in Figure 4. The influenceof alloying elements on the microstruc-tures of weld metal deposits is given inFigure 5 [9].

Repair ProceduresPreheatingGenerally, weld metal and parent metalproperties such as chemical composi-tion, hardenability, joint geometry andrestraint determine the desired proper-ties for a repaired component. One ofthe widely adopted approaches fordetermining weldability is to review thehardenability of the material. The car-bon equivalent (CE) formula wasdeveloped to indicate how the chemicalcomposition would affect hardenability.The maximum interpass temperaturefor the repair of austenitic manganesecastings is 260°C. Table 5 gives guide-lines for preheating temperature as afunction of carbon equivalent [10]. Thecarbon equivalent formula is given inequation 3 [13].

CE=C+(Mn+Si)/6+(Cr+Mo+V)/5+(Ni+Cu)/15(3)

Table 5. Guideline preheat tempera-tures as a function of carbon equivalent(CE).

Special precautions should be takenon applications that are crack sensi-tive, such as high carbon or alloysteels, previously hardfaced parts andhighly stressed parts. The repair (hard-facing) of heavy cylinders, massiveparts and parts having complexshapes are all examples of applica-tions producing high internal stressesthat may result in delayed cracking(Figure 7). These applications mayrequire one or more of the following:• Higher preheating temperatures 150

to 260°C (Figure 6).• Higher interpass temperatures up to

480°C. In general this high interpasstemperature will not cause a drop inthe hardness of weld deposit.Establishing interpass temperaturesshould also take into considerationthe previous heat treatment historyof the component.

• Controlled, slow cooling betweenpasses.

Table 4. Influence of alloying elements on the properties of weld deposits.

Figure 5. Map of alloying elementsand properties of build-up and wear-resistant weld deposits.

Figure 6. Preheat of massive part.

Figure 4. Influence of carbon contentand % martensite (cooling rate) on thehardness of steel weld metal deposits.

CARBON• Reduces ductility

(increases brittleness)• Increases tensile strength• Increases hardness• Increases hardenability

NICKEL• Increases strength &

toughness• Prevents grain growth• Lessens distortion• Increases hardenability

MANGANESE• Increases hardness• Promotes a finer grain size• Acts as deoxidiser• Minimizes sulphur,

hot cracking

MOLYBDENUM• Increases tensile strength

and toughness• Increases resistance

to creep

CHROMIUM• 1-2% increases the hardness and

toughness without loss of ductility• 4-6% increases resistance

to tarnishing• Above 11% becomes corrosion

resistant• Promotes carbide formation

VANADIUM• Increases tensile strength• Increases resistance to fatigue• Resistant to high stresses

Carbon Equivalent Suggested preheat (°C)

Up to 0.45 Optional 0.45 to 0.6 95 to 210 Above 0.6 210 to 370


A soaking time of 1 hour per 25 mm ofcross section at the recommendedtemperature is required in order toobtain maximum benefit from preheat-ing. The maximum interpass tempera-ture for the repair of austeniticmanganese castings is limited to260°C.

Postweld Heat TreatmentThe iron based hardfacing alloys areamong the few engineering alloys thatcan be heat treated in order to varytheir mechanical properties. Heat

treatment can be applied to a steel notonly to harden it but also to improve itsstrength, toughness, ductility, decreasethe stresses caused by welding and toavoid undesirable microstructures inthe heat affected zone. The variousheat treatment processes can be clas-sified as : a) annealing; b) normalisinghardening; c) tempering; d) stressrelieving.

A summary of the influence of stressrelieving temperature on the hardnessof weld deposits resistant to metal-to-metal wear is illustrated in Figure 8[11], while a summary of the influenceof annealing on the hardness of welddeposits resistant to metal to metalwear is illustrated in Figure 9 [12].

Figures 10 and 11 [12] summarize therelationship between the percentageof carbon and alloying elements andas quenched hardness of hardfacingweld metal deposits.

WIRE CHARACTERISTICS• Typically less than 0.3%C, less than 6% alloy

(Cr, Mn, Mo, Ni);• Pearlitic/Ferritic weld deposit• Hardness up to 35 HRc;• Two distinct applications;• Provide high compressive strength to support

a harder top layer – Build-up layer;• Final surface for metal to metal wear.

Figure 7. Transverse crack of therepaired idler.

Figure 9. Influence of annealing on thehardness of iron based weld deposits[12].

Figure 11. Relationship between totalcontent of alloying elements (Cr, Mo,V and W) and as hardfacing quenchedhardness of iron based hardfacingdeposits [12].

Figure 12. (a) Idler rebuild; (b) Worn internal surface of dragline chain;(c) chain repaired.

(a) (b) (c)

WELDING PROCEDURE• Preheat 50-210°C;• Maximum interpass can run as high

as 370-430°C;• Stringers or weave are acceptable;• Unlimited number of layers;• Slow cool to avoid cracking;• Hardness will depend on the cooling rate.

Figure 8. Summary of influence ofstress relieving temperature on hard-ness of weld deposits resistant tometal-to-metal wear (see Table 3).

Figure 10. Relationship between car-bon content and as quenched hard-ness of iron based weld deposits [12].

Typical Application of a Build-up Product

Examples


SummaryAlthough wear of machinery parts rep-resents a significant economic cost tothe owners and operators of heavyequipment, the option of using hard-facing products to restore worn materi-al is a very cost effective alternative toparts replacement. In many cases, thehardfaced deposit will wear better thanthe original part. The hardfacing solu-tion is successful when the type ofwear is properly identified, and theoptimal material is selected for theapplication. Care should be taken toensure that adequate ventilationand/or local exhaust is used to controloperator exposure to welding fumesand its constituents per the materialsafety data sheet for the consumablesbeing used. Finally, regardless of thehardfacing material selected, thematerial must be properly deposited toensure that it performs as intended.

REFERENCES[1] Yumaguchi Y. TRIBOLOGY OF PLASTIC

MATERIALS, Elsiver 1990, pp 92-102[2] WTIA TECHNICAL NOTE 4. The industry

Guide to Hardsurfacing for the Control ofWear

[3] D.J. Kotecki and J.S. Ogborn, AbrasionResistance of Iron Based HardsurfacingAlloys, Welding Journal, August 1995, pp269-278

[4] Suh N.P., THE DELAMINATION THEORY OFWEAR, Wear, 25 1973, pp 111-124

[5] ASM Metals Handbook, Vol. 10, 8th Edition,1975, pp 134-153

[6] Mutton P.J. ABRASION RESISTANT MATE-RIALS, AMIRA, 1988, pp 15-44

[7] ASM Metals Handbook, Vol. 4, Heat Treating,1991

[8] Askeland D.R., THE SCIENCE AND ENGI-NEERING OF MATERIALS, PWSEngineering, Boston, 1985, pp 510-550

[9] Kotecki D. “Hardfacing Benefits Maintenanceand Repair Welding”, AWS Welding Journal,November 1992, Volume 71/Number 11

[10] W. F. Newell, Jr, “Understanding and UsingCarbon Equivalent Formulas”, AWS WeldingJournal Sept. 1995.

[11] M. Dumovic “The effect of stress relieving onthe hardness of iron based weld depositsresistant to metal to metal wear”,Australasian Welding Journal, Volume 45,Third Quarter, 2000

[12] M. Dumovic “Effect of annealing and quench-ing on the hardness of iron based welddeposits resistant to metal to metal wear”,Australasian Welding Journal, Volume 46,First Quarter, 2001

[13] AWS D 1.1/D1.1M:2002 “Structural WeldingCode - Steel”, Annex XI

Figure 13. Shaft repaired using spread arc technique.

Figure 14. Repaired austenitic manganese steel casting; no preheating applied; maximuminterpass temperature was kept below 260°C by immersing component in water bath.

Typical Metal to Metal Wear Application

Typical Manganese Repair

Typical Abrasion and Impact Application

WELDING PROCEDURE• Preheat 150-315°C is recommended;• Max. interpass can go as high as 370-430°C;• Stringers or weaves are acceptable;• Usually limited to 3-4 layers maximum;• Slow cool to prevent cracking;• Post weld heat treatment required to toughen

and soften weld/component after welding.

WIRE CHARACTERISTICS• Typically less than 0.4%C and 6% total alloy;• Hardness typically 35-45 HRc;• Low alloy martensitic weld deposit;• Austenite transforms to martensite below 371°C;• Hardness doesn’t depend upon cooling rate

unless extremely slow;• Main application is metal to metal wear, especially

sliding; also abrasion from softer materials (dirt,limestone).

WELDING PROCEDURE• No preheat required on austenitic base metal;• Preheat 148-204°C on carbon and low alloy

to steel to prevent pullout;• Limited heat build up to 260°C maximum to avoid

embitterment due to Mn-carbide precipitation;• Unlimited layers;• No post weld heat treatment required.

WIRE CHARACTERISTICS• Suitable for severe impact applications;• Typically 0.4 to 0.6 %C, 13 to 20% alloy, mainly

manganese;• Typically 20 to 25 Rockwell C as-welded, work

hardens rapidly to 45 to 55 HRC;• High dilution on mild steel will be martensitic.• Non-magnetic alloys.

Figure 15. Bucket sides protected with hardfacing.

WELDING PROCEDURE• No preheat on austenitic substrate;• Preheat at 204°C on carbon steel, low alloy steel,

or cast iron;• First run several beads fast enough to establish

tight check crack spacing (6.0 to 19.0 mm) mayrequire ≥1000 mm/min travel speed;

• For a single layer, use heavy overlap (about 70%)to get primary carbides – dilution can lead toprimary austenitic or near eutectic structure whichhas inferior abrasion resistance.

WIRE CHARACTERISTICS• 2 to 6%C, 14 to 35% total alloy content,

mainly chromium;• Typically 58 to 63 HRC;• Used primarily to resist abrasion, abrasion &

impact – shovel and bucket lips, conveyor screws,blast furnace bells, coal crushers, asphalt mixers etc.


Lincoln Electric Technical Programs

Opportunities

Space is limited, so register early to avoid disappointment. For full details, see

www.lincolnelectric.com/knowledge/training/seminars/

Or call 216/383-2240, or write to Registrar, Professional Programs, The Lincoln Electric Company, 22801 Saint Clair Avenue, Cleveland, OH 44117-1199.

Welding of Aluminum Alloys,Theory and PracticeOctober 21-24, 2003Designed for engineers, technologists, techni-cians and welders who are already familiar withbasic welding processes, this technical trainingprogram provides equal amounts of classroomtime and hands-on welding.Fee: $495.

High Productivity Welding

Submerged Arc WeldingNovember 12-13, 2003The program will provide the basic theoretical concepts that support the process, advanced top-ics, and new developments. It will also emphasizewelding process optimization and welding costreduction opportunities. This seminar is designedto benefit welding engineers, technicians, super-visors, instructors, quality assurance personnel,and manufacturing engineers.Fee: $395.



Pay Attention to Tack and Temporary WeldsPractical Ideas for the Design Professional by Duane K. Miller, Sc.D., P.E.

Design File

Introduction

Ever noticed how sometimes the smallest details cancause the biggest problems? This situation can be com-pounded when a variable or factor is considered insignifi-cant and accordingly ignored. Such can be the case whentack welds and temporary welds are improperly made.

AWS A3.0 Standard Terms and Definitions defines a tackweld as: “A weld made to hold the parts of a weldment inproper alignment until the final welds are made.”

The term “temporary weld” is defined as: “A weld made toattach a piece or pieces to a weldment for temporary usein handling, shipping, or working on the weldment.”

The clear difference is that the tack weld joins “the parts ofa weldment,” whereas the temporary weld joins “a piece orpieces to a weldment.” Thus, a temporary weld will alwaysjoin to the weldment something foreign to the weldmentproper.

The term implies, but the A3.0 definition does not actuallymandate, that temporary welds have a limited life.Specifically, the definition does not require that the weld beremoved after its function has been performed. However,the implication is that after the weld and the associatedattachment have performed their function during “handling,shipping, or working on the weldment,” the attachment andthe associated weld will be removed. Thus, the weld thatjoins a lifting lug onto a weldment could be either a perma-nent weld (if the lug was to remain in place for future han-dling of the weldment) or a temporary weld (if the lug wasto be removed after handling the weldment). In these twosituations, the welds may be otherwise identical, but theyare called by different names.

Semantics and definitions aside, the important issue is this:both tack welds and temporary welds must be properlymade. Since there are no secondary members in weldedconstruction, improperly made tack or temporary welds

may create problems that result in the propagation ofcracks into main members. Further, temporary welds mayprovide the metallurgical path for cracks (if present) inattachments to propagate through the weld, into the mainmember. Accordingly, these seemingly unimportant weldsmay be critical, especially in weldments subject to cyclicloading. Let’s look into the issues that should be consid-ered when tack and temporary welds are designed andfabricated.

Same Quality Required

In general, the same quality requirements that would applyto final welds should apply to both tack and temporarywelds. The AWS D.1.1: 2002 Structural Welding Code—Steel requires this in 5.18.1, which states “Temporarywelds shall be subject to the same WPS requirements asthe final welds.” For tack welds, 5.18.2 reads “Tack weldsshall be subject to the same quality requirements as thefinal welds…” The provision goes on to list some excep-tions, to be discussed below. But the basic starting point fortack and temporary welds is that they are to be of thesame quality as the final welds.

More on Tack Welds

The A3.0 definition does not define the length or size of atack weld, but rather addresses the purpose of the weld.This definition does not, nor should it, preclude the use ofa continuous weld in the root of a joint. It does not, andagain should not, mandate a certain maximum size for thetack weld. Colloquial usage would suggest, however, that atack weld must be small, and intermittent. But as we willsee, small intermittent welds may be undesirable in somecircumstances.

Tack welds may be placed within the weld joint, and thensubsequently welded over with the final weld. Alternately,tack welds may be made outside the weld joint. For weldsmade within the weld joint, the tack weld may be complete-


Figure 1. Tack welds in joints.

ly remelted and become part of the final weld. Alternatively,part or most of the tack weld may remain within the joint,and become part of the final weld. Tack welds made out-side the joint may remain in place, and become part of thepermanent weldment, or they may be removed after thejoint has been partially or completely welded. The place-ment of the tack weld, its relationship to the fill passes inthe weld, and the final disposition of the tack—all will affecthow the tack weld is to be treated.

A tack weld must be sufficiently strong to resist the loadsthat will be transmitted through it. Some weldments have

individual components that are massive, and the weight ofsuch parts may be transferred through tack welds while theweldment is handled during fabrication. Careful sizing oftack welds that are used for this purpose is essential. Tackwelds are often required to hold parts in alignment whileassemblies are being preheated for final welding. Thermalexpansion, the corresponding strains, and resultant stress-es may necessitate tack welds of significant strength. Thestrength of tack welds, like other welds, is proportional tothe throat size, and the length. Thus, a stronger tack weldmay be made by making it with a larger throat, or longerlength, or both. In most cases, tack welds are intermittent,and the strength across the joint can be made greater byincreasing the number of intermittent tack welds, even tothe point of a continuous tack weld. Other factors, dis-cussed below, should be considered when determiningwhether the tack weld should be made longer, or larger,when additional strength is required.

Tack Welds Within the Joint

Examples of tack welds within a joint, shown in Figure 1,would include:• A tack weld in a T-joint that will receive a final fillet weld• A tack weld in the root of a CJP groove weld preparation

that attaches the steel backing• A tack weld in the root of a PJP groove weld preparation

In each case, the final weld is placed over the tack weld.The subsequent weld passes may totally remelt the tackweld, and significantly reheat the surrounding heat affectedzone (HAZ). Or, the following passes may partially melt thetack weld, and reheat the surrounding HAZ, but the reheat-ing may be to a level such that the previous HAZ propertiesare not much changed. The first condition will be referred toas “remelted tack welds” and the second as “incorporatedtack welds.” Fundamentally different approaches should betaken to remelted versus incorporated tack welds.

(a)

(b)

(c)


Remelted Tack WeldsThe basic concept behind remelted tack welds is that thesubsequent weld passes will effectively eliminate all evi-dence that the tack weld ever existed (see Figure 2).Accordingly, it is reasonable that quality criteria associatedwith tack welds that will be remelted would be more relaxedthan for the situation where the tack welds become part ofthe completed weld. This is reflected in D1.1 as follows:

5.18.2 General Requirement for Tack Welds. Tack weldsshall be subject to the same quality requirementsas the final welds, with the following exceptions:

(1) Preheat is not required for single-pass tack welds which are remelted and incorporated into continuous SAW welds.

(2) Removal of discontinuities, such as undercut, unfilled craters, and porosity before the final SAW is not required.

The two exceptions apply only for the conditions of (a)remelting and incorporation, and (b) subsequent weldingby SAW (submerged arc welding).

It should not be assumed that remelting will automaticallyoccur when SAW is used. With the high amperage levelstypically associated with larger electrode diameters, remelt-ing may routinely occur. However, SAW may be performedwith smaller diameter electrodes, composite (cored) versussolid electrodes, DC- polarity, long contact tip to work dis-tances, lower current levels, and other conditions that mayinhibit the ability of the SAW process to remelt tack welds.

Nor is it appropriate to assume that other welding processescannot remelt tack welds. Electroslag (ESW) and electro-gas (EGW) are obvious examples of deep penetratingprocesses where tack welds are expected to be remelted.While D1.1 does not permit the consideration of otherwelding processes for remelting tack welds, for work notgoverned by codes with such a restriction, the capability of other processes could be evaluated. The AWS/AASHTOBridge Welding Code D1.5: 2002, for example, recognizesthe remelting capability of ESW and EGW and extends the exception to these processes as well (D1.5: 2002,3.3.7.1[1]).

Heavy sections of steel, and higher strength steels withtheir corresponding higher carbon and/or alloy levels, typi-cally require preheat. Maintence of WPS (welding proce-dure specification) preheat levels is required for tack welds,unless the exception conditions are met. Even though asmall tack weld on non-preheated thicker sections mayresult in a hard, crack sensitive heat affected zone aroundthe tack weld, the high heat input levels of SAW passesthat remelt the tack welds will also reheat the HAZ. If theHAZ created by the tack weld is heated above the transfor-mation temperature, and permitted to slowly cool, the hard,crack sensitive HAZ will be softened. Discontinuities in thetack weld that will be remelted are not a concern, as theremelting process eliminates the discontinuities as well.

If the intent is to remelt the tack weld, then the tack weldshould be made with a geometry that is conducive toremelting. Remelting is facilitated when the tack weld is rel-atively small. Large tack welds are more difficult to remelt,and excessively large tack welds will not be remelted evenwith high energy SAW procedures. Thus, to gain therequired joint strength with tack welds that will be remelted,emphasis should be placed on making small welds that arelonger in length. Not only will this encourage remelting ofthe tack weld, it also minimizes the tendency to disrupt thesurface appearance of the final weld.

Incorporated Tack Welds

When tack welds are placed within the joint and not remelt-ed, they are automatically incorporated into the subsequentfinal weld (see Figure 3). When this is the case, the tackweld should be treated in a manner much like the root passof a final weld: everything associated with an incorporatedtack weld should be the same as would apply to the weldroot pass. The preheat, filler metal selection, WPS parame-ters, weld size, heat input, and the quality of deposit shouldmeet the same standards as would apply to the root pass.In terms of quality, this would include undercut levels,porosity limits, bead shape criteria, and the absence ofcracks. Remember: these tack welds will be incorporated,and therefore, will be part of the final weld.

TACK

HEATAFFECTEDZONE

FINALWELD

HEATAFFECTEDZONE

Figure 2. Remelted tack welds.

(a)

(b)


A major shift in thinking is required when tack welds are tobe incorporated, as compared to the remelted alternative.For example, incorporated tack welds should be made of asize, and with a heat input level, that will ensure goodfusion. These welds should meet the minimum size require-ments that would be imposed on any final weld. This willnaturally result in larger sized tack welds than was encour-aged for remelted tack welds. Thus, for a required jointstrength, incorporated tack welds will be larger in size, butperhaps shorter in length, as compared to the remeltedoption.

Large, intermittent tack welds may require that the gapsbetween the tack welds be completely welded before thesubsequent layers are made. Welding over large tack weldsmay disrupt the arc, or may affect the appearance of thesubsequent final weld. The ends of the tack weld may bepoints where fusion into the weld root is difficult to achieve.Thus, the acceptable geometry of the tack weld is depen-dent on the ability of the final weld procedure to properlyincorporate the tack weld into the final weld. This is thereason, for example, that 5.18.2.1 of D1.1 requires thatmultipass tack welds have cascaded ends.

Considerations for Both Types of Tack Welds

Irrespective of whether the tack weld will be remelted orincorporated, the interaction of the tack weld and the finalweld must be considered. Tack welds are often made with

a different welding process, or even when the sameprocess is used, with a different filler metal than will beused for the final weld. Chemical interactions between thetwo types of materials should be considered.

Some self shielded flux cored arc welding (FCAW-S) fillermetals will have slag removal problems when welding overtack welds made with certain shielded metal arc welding(SMAW) electrodes. The electrode manufacturer can becontacted for a list of compatible electrodes.

A second type of interaction that must be considered is thepotential effect of intermixed weld metals on mechanicalproperties. The Charpy V-notch toughness of subsequentfinal passes of normally tough welds may be reduced dueto negative interaction with tack welds made with weldingprocesses using a different shielding system. The typicalcombination that should be investigated is when FCAW-Sis used to tack weld under non-FCAW-S deposits, such asSAW, or gas shielded FCAW. This does not mean that allcombinations are unacceptable. However, these should beinvestigated on a case-by-case basis. The James F. LincolnFoundation publication The Fabricator’s and Erector’sGuide to Welded Construction, available as a free PDFdownload from www.jflf.org, addresses a variety of combinations.

Tack Welds Outside the Weld Joint and Temporary Welds

When tack welds are placed outside the weld joint, and forall temporary welds, other factors must be considered.Simply put, these welds too should be treated as any finalweld. They should be made with materials, procedures,techniques and quality levels that would be acceptable forfinal welds. Tack welds outside the weld joint fit into twocategories: permanent, and removed. Temporary welds, bydefinition, will be removed.

Permanent Tack Welds Outside the Weld Joint

Tack welds outside the weld joint must be evaluated todetermine if they can remain in place without causing unin-tended consequences. Of necessity, D1.1 places thisresponsibility on the Engineer as follows:

5.18.2.3 Nonincorporated Tack Welds. Tack welds notincorporated into final welds shall be removed, exceptthat, for statically loaded structures, they need not beremoved unless required by the Engineer.

Thus, for statically loaded structures, the normal practicewill be to leave such tack welds in place, unless otherwiseindicated by the Engineer. For dynamically loaded struc-tures, such nonincorporated tack welds would be removed.

TACKWELD

FINALWELD

TACKWELD

HEATAFFECTEDZONE

Figure 3. Incorporated tack welds.

(a)

(b)


Consider the potential tack welds that could be used toattach longitudinal backing under a CJP groove weld. Oneoption would be to tack weld the backing in the root of thejoint as shown in Figure 1b. This may pose some practicalproblems, prompting the need to tack weld the backing out-side the joint. Intermittent tack welds would be sufficient tohold the material in place, as shown in Figure 4a. Thiswould be an acceptable option under D1.1 for staticallyloaded members. However, if the same member was putinto a cyclic loading situation, the intermittent tack weldswould behave as Category E fatigue details, greatly limitingthe allowable stress range. An acceptable alternative(although not one specifically given in 5.18.2.3 of D1.1) isto make a continuous tack weld, which would behave as aCategory B detail, much like the longitudinal CJP grooveweld (Category B) shown in Figure 4b.

It is good to remember the adage “There are no secondarymembers in welded design” when evaluating the suitabilityof leaving in place tack welds that are made outside theweld joint.

Removing Tack and Temporary WeldsWhen tack welds are required to be removed, and whentemporary welds are removed, it is important that that theweld be fully removed without damaging the base metal. Atypical approach is to thermally cut the weld or attachmentoff (using air arc gouging, oxy fuel cutting, or plasma cutting),and follow up with grinding. When cutting is performed tooclose to the final surface, one may inadvertently gouge thebase metal.

The procedure described above assumes, however, thatthe tack weld, or the temporary weld, was properly made inthe first place. Consider the improper procedure wherein atack weld is placed outside the joint, but the weld is madewithout preheat, or the needed minimum heat input, or withan improper electrode. Such a procedure could result in anunderbead crack, an excessively hard HAZ, or other welddefects. Simple removal of weld metal from the surface ofthe steel will not automatically remove the defect that mayreside in the base metal. This can result in performanceproblems for the weldment, particularly when subject tocyclic loading.

In the case of Fracture Critical Members (FCMs), D1.5requires that, when weld removal is required, the weld plus1/8 in. [3mm] of adjacent metal be removed. The surfacesare faired in at a slope not steeper than 1 in 10 on the sur-face (see D1.5: 2002, 12.13.3). This conservative provisionensures that the whole weld, plus any affected base metal,is completely removed, along with any unacceptably hardor cracked material.

Summary

Neither tack welds nor temporary welds should be viewedas inconsequential, secondary welds, particularly whenapplied to cyclically loaded weldments. Whether the tackweld is to be made in the joint or not will affect the overallapproach to the weld. If it is made in the joint, whether thetack weld is to be remelted or incorporated will determinethe ideal configuration for that tack weld. For tack weldsand temporary welds that will be removed, care must betaken to protect the base metal. There are plenty of oppor-tunities to make tack and temporary welds improperly.Fortunately, it is not difficult to make them correctly.

Figure 4. Non-incorporated, intermittent (a) and continuous (b) tack welds.

(a)

(b)


Lincoln Electric Professional Programs

Opportunities

Blodgett’s Design of WeldmentsOctober 7-9, 2003Blodgett’s Design of Weldments is an intensive 3-day program for those concerned with manufactur-ing machine tools, construction, transportation, mate-rial handling, and agricultural equipment, as well asmanufactured metal products of all types. Seminarleaders: Omer W. Blodgett and Duane K. Miller. 2.0CEUs. Fee: $595.

Fracture & Fatigue Control in Structures:Applications of Fracture MechanicsOctober 28-30, 2003Fracture mechanics has become the primary approach to analyzing and controlling brittle fractures and fatiguefailures in structures. This course willfocus on engineering applications usingactual case studies. Guest seminarleaders: Dr. John Barsom and Dr. StanRolfe. 2.0 CEUs. Fee: $595.

Blodgett’s Design of WeldedStructuresNovember 11-13, 2003Blodgett’s Design of Welded Structures is an intensive 3-day program whichaddresses methods of reducing costs,improving appearance and function, andconserving material through the efficientuse of welded steel in a broad range of structural applications. Seminar leaders: Omer W. Blodgett and Duane K. Miller. 2.0 CEUs. Fee: $595.

Space is limited, so register early to avoid disappointment. For full details, see


Or call 216/383-2240, or write to Registrar, Professional Programs, The Lincoln Electric Company, 22801 Saint Clair Avenue, Cleveland, OH 44117-1199.



The ProjectMy company was asked to fabricatesixteen catcher beams to be installedon the Commodore Barry Bridge out-side Philadelphia. In the event that themain pins that carry the deck of thebridge should ever break, thesebeams are designed to “catch” thedeck and keep the bridge from plung-ing into the water. The creation andinstallation of the catcher beams waspart of a retrofit project on the bridge,which was built in 1974 and is ownedby the Delaware Port Authority. Theproject was completed in November,2002.

Background

The steel selected for the catcherbeam project, a 2 in. [50 mm] thickA710 Grade A Class 3 material, waschosen for its high Charpy value, 60 ft-lbs [80 J] at minus 80°F [minus 62°C].When I tried to develop a history ofthe A710 steel, information was verylimited. I could not find that it had everbeen used on a bridge anywhere inthe United States. The U.S. Navy hadused A710 in the past, but had a his-tory of it cracking at thicknessesgreater than 1 in. [25 mm].

The structure of each catcher beamconsisted of a box with two flanges 2-

1/2 in. thick by 17 in. [65 mm by 430mm] wide and two webs that were 2in. thick by 18 in. high [50 mm by 460mm]. The outside corner welds were1-3/8 in. [35 mm] partial joint penetra-tion (PJP) groove welds, with an insidecorner weld of 3/8 in. [9.5 mm]. This ofcourse was only on the bottom flange,which could be welded inside. Oncethe cap was put on, welding wasrestricted to the outside, dictating PJPgroove welds only.

The Initial Challenge

With this very limited information, I puttogether a game plan according towhich we would preheat and rotatethis assembly after it was tacked up.We built a preheating station in whichthe part could be heated and rotated.It became known as the “rotisserie.”We brought the beam up to 400°F

[200°C], let it soak for one hour at thattemperature, then let the weldmentcool down to 350˚F [175˚C], beforewelding the first root pass on the twobottom outside partial joint penetrationgroove welds. After that, the spreaderbeam was taken out of the rotisserie,

and we ran the two inside 3/8 in. [9.5mm] fillet welds. The temperature wasstill maintained at 350°F until the com-pletion of the 3/8 in. fillet. Then theassembly was allowed to cool down toroom temperature.

Now the spreader beam was fittedwith six internal bearing stiffeners.Both ends of the stiffeners had to havea mill-to-bear fit to each flange, and noweld was initially specified. The mill-to-bear tolerance was zero clearanceacross the full width of the stiffener.Anticipating that this fit requirementwould be difficult to achieve, we fabri-cated a yoke-type fixture to tightlyclamp the stiffener to the flange. Thefixture ensured that this fit was main-tained before welding began.

Despite the fixturing, the zero clear-ance tolerance proved to be a realchallenge. We would start to weld, andthe stiffener would start rising out ofthe box, creating a 0.015 to 0.020 in.[0.381 to 0.508 mm] gap. We changedthe direction of welding, and tried dri-ving the weld in from the outside cor-ner. Ultimately, we ended up running ashort 6 in. [150mm] weld, skipping a 6in. space and completing the joint byusing a back-step welding techniqueto complete the fillet welded joint.

Understanding Distortion is a Never Ending Challenge

Lessons Learned in the Field

The zero clearance toleranceproved to be a real challenge

By Byron HornWelding SpecialistMichelman-Cancelliere Iron Works, Inc.Bath, Pennsylvania


We also used differential preheat tocontrol this dimension. We only pre-heated the web, with no direct heatapplied to the stiffener. The stiffenerreceived its 250°F [120°C] preheat viaheat conduction through the web. The100°F [40°C] differential allowed theweb volume to expand to a greaterdimension than the stiffener. Then,after welding and overall cooling, theweb volume contracted more than thestiffener volume, which drove the stiff-ener into contact with the bottomflange.

The combination of the weldingsequence and the differential preheat

were sufficient to get the stiffener intocontinuous contact with the flange. Atthis point, one consultant we spokewith about the project remarked, “Youmight as well be making a Swisswatch out of a steel beam!” It certainlyfelt that way at times. But our problemsstill were not over.

Another Problem

The surface of the flange plate wasn’tflat enough for the tolerance. We dis-covered that the way these platescame from the steel mill, there was avery slight waviness and grooves onthe surface of the plate, and that

accounted for the 0.015 in. [0.381 mm].Since the plate surface rolling toler-ances exceed 0.015 in., no amount ofpreheat temperature differentialbetween the web and stiffener couldcorrect the out of tolerance gaps. Wewent back to the Engineer, who raisedthe tolerance to 0.010 in. [0.54 mm].Still, we had rejections. Then theEngineer raised the tolerance to 0.015in. [0.381 mm]. More rejections. So weended up receiving permission fromthe Engineer to weld the bottom of thestiffeners to the flange.


Now, The Top FlangeTo control the gap between the stiffen-er and the top flange, both compo-nents were machined in a big planerto a flat and level condition.

On the first spreader beam that waswelded, the top flange distorted in aconvex direction approximately 0.030in. [0.76 mm]. To reduce this distortionwe used a post heat of 350°F [175°C]for a minimum of 8 hours. The postheat relaxed the welding residualstresses to a point where the flangeflattened down to under 0.015 in.[0.381mm].

The other thing I did to control distor-tion was to build in a 0.010 in. [0.254mm] gap between the flange and theweb, fixing this dimension by usingsmall tack welds. The tacks created agap which gave room for the shrink-age, to bring the top plate down beforeit started to distort. Creating this smallgap did about as much for us to con-trol distortion as anything. The heatdefinitely helped flatten the flange outover time, but the small gap gave itsome place to go.

Hydrogen ControlThe second challenge was to preventdelayed weld cracking by controllinghydrogen in the weld. This was donevia two methods:

First, we used a controlled hydrogenprocess: SAW (Lincoln Mil800-H fluxand LS3 wire combination) with dif-fusible levels between 1.5 to 2 ml/100grams of weld deposit.

Second, we filled each partial jointpenetration groove weld to only half ofits depth before rotating to the next

joint. Hydrogen takes time to diffuse,and the greater the distance of materi-al through which the hydrogen musttravel, the more time will be required.By welding only half the groove depth,the distance for hydrogen to diffusewas reduced. Also, since the wholeassembly was maintained at the pre-heat/ interpass temperature, the rateof hydrogen diffusion was greater.While hydrogen was diffusing from thepartically welded joint, another jointwas welded.This allowed a minimum

of 12 hours for the diffusible hydrogenlevel to drop even lower. This proce-dure also balanced some residualstresses, controlling distortion (sweepand camber) of the beam assemblies.

Personnel with whom I had spoken attwo Navy shipyards described delayedcracking in welds over 1 in. [25 mm]. Ibelieve that if the shipyards had fol-lowed the above with proper preheatpractices and the use of low to medi-um restrained joints, weld crackingwould have been reduced to anacceptable level.

ConclusionDespite the demanding conditions, theproject was completed successfully. Byusing principles of distortion control,the displacements were minimized,although the final solution required theapplication of a fillet weld to overcomeall the challenges. A cooperativeEngineer and careful planning over-came the dimensional control prob-lems. Selection of the propermaterials, control of procedures, andcareful preheat and interpass temper-ature controls overcame any tenden-cies toward hydrogen cracking. I guesswe did make a Swiss watch out of asteel beam.

Making a Swiss watch out of a steel beam


BackgroundThe building permit cost $185,000.The field erection consumed 1.1 tons[0.99 tonnes] of welding electrode. Inanother year, the structure will holdmore than 80 million pages of docu-ments, 40 million e-mails, 2 millionphotographs and almost 80,000 arti-facts. When it opens in November2004, the $160 million WilliamJefferson Clinton Presidential Centeris expected to put Little Rock,Arkansas on the map as a tourist des-tination. For the employees of AFCOSteel of Little Rock, Arkansas, thecompany that fabricated the structuralsteel for the building, the project holdsspecial meaning. According to BobBendigo, V.P. of Operations at AFCO,

“The people who build our projects inthe plant are seldom able to see themgo up. But in this case, we all drive byit every day, coming to and from work.

And that caused a heightened interestin the project, and a pride that wentalong with that.”

The long, slender building elevatedabove a park was designed by thePolshek Partnership of New York toexpress former President Clinton’s

favorite theme for his administration,that it was “a bridge to the 21st centu-ry.” The oblong design also echoes thesix bridges that span the ArkansasRiver in Little Rock. In addition,Architect James Polshek incorporatedan existing (now unused) railroadbridge into the site design; it will berenovated for pedestrian use. The circa1899 Choctaw railroad station depotbuilding adjacent to the presidentiallibrary is being renovated to house theClinton School for Public Policy, whichwill grant master’s degrees in publicservices starting in 2004. A 30 acre[120,000 m2] park featuring anamphitheater, a children’s playground,and trails for walking and bicycling, isthe site for the entire complex.

Making the Essential Connections on “A Bridge to the 21st Century”

By Carla RautenbergWelding Innovation Contributing WriterThe James F. Lincoln Arc Welding FoundationCleveland, Ohio

The plan required the steelpieces to fit together perfectly

the first time

Figure 1. Truss elevation.

18 Welding Innovation Vol. XIX, No. 2 2002

Developing the Game PlanThe library’s design called for twolarge parallel trusses, 420 ft. [128 m]long, and 36 ft. [11m] tall. With sup-ports at only three locations, the twoends of the building were designed tocantilever out 90 ft. [27 m]. On thenorth end of the building, a massivesteel pier supports the structure. Thisdesign not only created the architect’svision of a bridge, but also resulted inbridge-like members for the steel fabri-cator to build.

When AFCO Steel won the contract tofabricate the structural steel for theClinton library, the company knew thatall of its planning would flow from themethodology selected to erect thestructure. Proposals were submittedby four erectors, all of whom hadcomparable qualifications. But accord-ing to Gary Johnson, V.P. of Contractsfor AFCO, “This job was so unique, noone could say, ‘Yeah, I’ve built someof those before.’ So we had as manydifferent schemes to erect this build-ing as we had bids. They were eachfirmly committed to their own way oferecting the project.” In the end,AFCO chose the strategy proposed

by Derr Steel Erection Company ofEuless, Texas, which called for thetruss to be fabricated in individualpieces, then, using falsework, to putthe pieces in their proper position andelevation and to assemble the truss inthe air, in the vertical position. Thisscheme modified the normal practiceof performing more welding in theshop, versus in the field. And the planrequired that the steel pieces wouldhave to fit together perfectly, the firsttime, if the job schedule was to bemaintained.

While the truss design and its assem-bly on site echoes elements typical ofbridge construction (see front andback cover photos), the building ofcourse did not pose the dynamic load-ing challenges of a bridge. And whilebridges are usually field-bolted, thisstructure would be joined together withfield welded connections. Unlike mostbuildings made principally of wideflange shapes, the library made exten-sive use of built up, four-plate box sec-tions, fabricated in the shop andrequiring very high standards ofdimensional control.

Fabricating the MembersThe planning for the project had to bemeticulous. Every piece of structuralsteel was custom-fabricated by anAFCO shop crew that numbered from30 to 40 welders. AFCO’s Johnsonsaid the greatest challenge overallwas “To maintain the integrity of eachpiece that we were building, as wellas to make the connection prepara-tions that would fit together in the fieldto complete the truss configuration.Because of their size and weight, wecould not lay these trusses down inour shop.” To supply the Derr SteelErection Company with a steadystream of fabricated components inthe field, AFCO allowed a 10-weeklead time for fabrication of the boxsections that would make up the trusschords and braces, and 4-6 weeks forthe wide-flange members that wereused for floor beams and other mem-bers. Johnson noted that at the out-set, the planned lead time seemedmore than ample, but “in the end, itwas just about right. We wouldn’t havewanted to do it any faster.”

The first truss drawings were issued tothe shop on September 3, 2002. Trussfabrication began seven days later andwas completed on January 16, 2003.Elements of the complex fabricationtask included:

Truss MembersThe truss chords consisted of four-plate built-up steel box sections madeof A572 Grade 50 and A588 plate.A572 was used for members withthicknesses ranging from 5/8 in. [15 mm]to 4 in. [100 mm], and A588 for thick-nesses greater than 4 in., and up to 6in. [150 mm] (since the A572 Grade 50specification covers material up to andincluding 4 in. thick, whereas the A588specification governs plate up to 8 in.[200 mm] thick). The truss has 30 ver-tical members, 32 diagonal members,36 chord segments, and 36 nodes(see Figure 1). The verticals and diag-onals were shipped as individual com-ponents to be erected at the site. The

Figure 2. Typical truss member section.


36 nodes and 36 chord sections wereshop assembled into 22 shippingpieces ranging in length from 28 to125 ft. [8.5 to 35 m] long, and weigh-ing up to 40 tons [36 tonnes].

Box WeldingEach four-plate box section requiredfour full-length fillet welds, to fabricatethe member (see Figure 2). Filletwelds ranged from 5/16 to 1/2 in. [7 to13 mm], depending upon the thicknessof the plates being joined. The 4,178linear ft. [1273 m] of box membersrequired 16,712 linear ft. [5,094 m] of

shop fillet welding—or a total of 3.2miles [5.1 km]. Shop welding was per-formed using submerged arc weldingand shielded metal arc welding.

Chord / Vertical / Diagonal to Node WeldingThe steel trusses have all welded con-nections, with the verticals and diago-nals connected to the chords at nodes(see Figure 3) using complete jointpenetration groove welds in plate vary-ing from 1 to 4 in. [25 to 100 mm] thick.The 36 chords and 36 nodes wereshop assembled into shipping pieces

using 1,456 in. [37 m] of CJP groovewelds. The balance of the welding ofthe chords/verticals/diagonals to thenodes was completed on site using6,032 in. [153.2 m] of CJP groovewelds field welded in the horizontaland vertical positions. Flux cored arcwelding was used in the field.

Support Columns and PiersThe above-grade structure is support-ed at the center and the south end byfour columns that are four-plate builtup box sections similar to the trussmembers. They were fabricated of

Figure 3. Typical node configuration.


plate in thicknesses of 1-1/2 to 4 in.[38 to 100 mm], are 41 ft. [12.5 m]long and have an average weight of12.5 tons [11.34 tonnes]. The northsupport is a steel pier composed oftwo columns and two cross membersthat tie the columns together into asingle assembly 36 ft. [11 m] tall by 20ft. [6m] wide. The 72-ton [65.3 tonne]pier was fabricated outdoors due to itssize and weight, then shipped to thesite as a single unit. The massive steelpier cap shown in Figure 4, fabricatedby Capital Steel in Oklahoma City andtrucked to the job site, is 15 ft. [4.6 m]deep by 50 ft. [15.2 m] wide andweighs 95.5 tons [86.6 tonnes].

The Field Erection Process

In the field, the greatest challengeswere: supporting the structure withfalsework towers and bracing, andminimizing distortion to maintain thestraightness of the structure.

According to Jeremy Beadles, projectengineer with Derr Steel Erection, the700 ton [635 tonne] weight of the trusswas daunting: “With a truss that heavy,it’s really hard to design erection aidsthat can carry the load.”

Carl Williams, senior engineer withDerr, described efforts to maintain thestraightness of the structure: “Weworked out a sequence of welding bothsides of each joint at the same time tominimize the welding draw. Then wewent through the structure and cameup with an overall welding sequence inorder to minimize the possibility of it alldrawing the structure out of alignment.Our sequences worked well, and wewere able to keep the structure in awhole lot tighter alignment than eventhe Engineer of Record thought we

could.” Preheats from 150-350°F [65 to175°C] were used, with some weldersarriving at the site at 5:30 a.m. to startthe preheating process, ensuring therequired temperature would beachieved when the rest of the weldersarrived at 7:00 a.m.

The project was managed by CDIContractors, LLC of Little Rock,Arkansas, and their project manager,Rob Hawkins.

A “Hold Your Breath” Moment

The four-story, 420 ft. [128 m] longbuilding is supported in just three loca-tions, with the trusses on the north andsouth both cantilevered out 90 ft. [27 m].AFCO’s Gary Johnson pointed out, “Inputting the truss together, we wereinstructed by the design engineer tocant those end sections upwards 2 in.[50 mm] above true horizontal, from thelast support. There was temporary

shoring under the truss members toachieve that upward cant. After thetruss was fully erected and all the fieldwelds were made, the shoring had tobe removed, and the weight of thestructure had to be carried by the truss-es. There was a predicted drop for theweight of the steel, and a predicteddrop for the added weight of the con-crete, and a predicted final drop whenthe wall cladding was added. So wewere quite apprehensive to see if we

would maintain the proper deflection.It was supposed to deflect about 2 in.[20 mm], and that’s what it did. At thatpoint, we knew we had the job in hand.”

The Topping-Out Ceremony

On May 23, 2003, with over 3,000people cheering him on, formerPresident Clinton added his signatureto a beam already adorned with thenames of 5,000 donors to the WilliamJ. Clinton Presidential Foundation. Itwas the final piece of steel to be hoist-ed into place signifying the completionof the structural phase of the library’sconstruction. The Arkansas DemocratGazette reported that Mr. Clinton toldthe crowd, “I’ve lived a highly improba-ble life. I hope this library and museumwill capture a little of that, but in a larg-er sense.” Upon completion of thebuilding, he plans to spend at leastone week a month in Little Rock, livingin a private apartment on the top floor,and participating in educational pro-grams at the Clinton School of PublicService next door, which will be affiliat-ed with the University of Arkansas.

The Clinton Presidential Center com-plex (Figure 5) is seen as an economiccatalyst for the city of Little Rock,whose mayor, Jim Dailey, told theGazette “New businesses are cominghere, and the city is attracting a greatdeal of attention because of thelibrary.” For the engineers, foremen,ironworkers, and fabricators who builtthe steel structure, it represents notonly hometown pride, but a triumph ofmeticulous planning and precisionwelding. As Warren Lenon, qualityassurance manager for AFCO, said,“The most satisfying thing was justseeing how it welded together. It reallywent together well.”

For those who built the structure, it represents a triumph of meticulous planning and precision


Figure 4.Cap girdererectionon pier.

Figure 5.The WilliamJ. ClintonPresidentialCenter andPark.

▲▲

NON-PROFIT ORG.U.S. POSTAGE

PAIDJAMES F. LINCOLN

ARC WELDING FND.

Truss progress on the Clinton Presidential Library in Little Rock, Arkansas. See story on page 17.

P.O. Box 17188Cleveland, OH 44117-1199

TheJames F. LincolnArc WeldingFoundation

Pho

togr

aph

cour

tesy

of

Ark

ansa

s A

eria

l Pho

togr

aphy

Documents

A publication of the James F. Lincoln Arc Welding · PDF file · 2011-02-17Since 1936, the James F. Lincoln Arc Welding Foundation ... Sales and Marketing, The Lincoln Electric Company