Sổ tay kết cấu thép - Section 2

2.1
SECTION 2
FABRICATION AND ERECTION
Thomas Schflaly*
Director, Fabricating & Standards
American Institute of Steel Construction, Inc.,
Chicago, Illinois
Designers of steel-framed structures should be familiar not only with strength and service-
ability requirements for the structures but also with fabrication and erection methods. These
may determine whether a design is practical and cost-efficient. Furthermore, load capacity
and stability of a structure may depend on design assumptions made as to type and magnitude
of stresses and strains induced during fabrication and erection.
2.1 SHOP DETAIL DRAWINGS
Bidding a structural fabrication project demands review of project requirements and assembly
of costs. A take-off is made listing each piece of material and an estimate of the connection
material that will be attached to it. An estimate of the labor to fabricate each piece is made.
The list is sorted, evaluated, and an estimate of the material cost is calculated. The project
estimate is the sum of material, fabrication labor, drafting, inbound and outbound freight,
purchased parts, and erection.
There are many issues to consider in estimating and purchasing material. Every section
available is not produced by every mill. Individual sections can be purchased from service
centers but at a premium price. Steel producers (mills) sell sections in bundle quantities
that vary by size. A bundle may include five lighter weight W18 shapes or one heavy W14.
Material is available in cut lengths but some suppliers ship in increments of 4 to 6 in.
Frequently material is bought in stock lengths of 30 to 60 ft in 5 ft increments. Any special
requirements, such as toughness testing, add to the cost and must be shown on the order.
Advance bills of material and detail drawings are made in the drafting room. Advance
bills are made as early as possible to allow for mill lead times. Detail drawings are the means
by which the intent of the designer is conveyed to the fabricating shop. They may be prepared
by drafters (shop detailers) in the employ of the fabricator or by an independent detailing
firm contracted by the steel fabricator. Detail drawings can be generated by computer with

software developed for that purpose. Some computer software simply provides a graphic
*Revised Sect. 2, previously authored by Charles Peshek, Consulting Engineer, Naperville, Illinois, and Richard W.
Marshall, Vice President, American Steel Erectors, Inc., Allentown, Pennsylvania.
2.2
SECTION TWO
tool to the drafter, but other software calculates geometric and mechanical properties for the
connections. Work is underway to promote a standard computer interface for design and
detail information. The detailer works from the engineering and architectural drawings and
specifications to obtain member sizes, grades of steel, controlling dimensions, and all infor-
mation pertinent to the fabrication process. After the detail drawings have been completed,
they are meticulously checked by an experienced detailer, called a checker, before they are
submitted for approval to the engineer or architect. After approval, the shop drawings are
released to the shop for fabrication.
There are essentially two types of detail drawings, erection drawings and shop working
drawings. Erection drawings are used by the erector in the field. They consist of line diagrams
showing the location and orientation of each member or assembly, called shipping pieces,
which will be shipped to the construction site. Each shipping piece is identified by a piece
mark, which is painted on the member and shown in the erection drawings on the corre-
sponding member. Erection drawings should also show enough of the connection details to
guide field forces in their work.
Shop working drawings, simply called details, are prepared for every member of a steel
structure. All information necessary for fabricating the piece is shown clearly on the detail.
The size and location of all holes are shown, as well as the type, size, and length of welds.
While shop detail drawings are absolutely imperative in fabrication of structural steel,
they are used also by inspectors to ascertain that members are being made as detailed. In
addition, the details have lasting value to the owner of the structure in that he or she knows
exactly what he or she has, should any alterations or additions be required at some later
date.
To enable the detailer to do his or her job, the designer should provide the following
information:

For simple-beam connections: Reactions of beams should be shown on design drawings,
particularly when the fabricator must develop the connections. For unusual or complicated
connections, it is good practice for the designer to consult with a fabricator during the design
stages of a project to determine what information should be included in the design drawings.
For rigid beam-to-column connections: Some fabricators prefer to be furnished the mo-
ments and forces in such connections. With these data, fabricators can develop an efficient
connection best suited to their practices.
For welding: Weld sizes and types of electrode should, in general, be shown on design
drawings. Designers unfamiliar with welding can gain much by consulting with a fabricator,
preferably while the project is being designed.
If the reactions have been shown, the engineer may show only the weld configuration. If
reactions are not shown, the engineer should show the configuration, size, filler metal
strength, and length of the weld. If the engineer wishes to restrict weld sizes, joint config-
urations, or weld process variables, these should be shown on the design drawings. Unnec-
essary restrictions should be avoided. For example, full joint penetration welds may only be
required for cyclic loads or in butt splices where the full strength of the member has to be
developed. The AWS D1.1 Welding Code Structural permits differing acceptance criteria
depending on the type of load applied to a weld. The engineer may also require special
testing of some welds. Therefore to allow proper inspection, load types and special testing
requirements must be shown on design drawings.
For fasteners: The type of fastening must be shown in design drawings. When specifying
high strength bolts, designers must indicate whether the bolts are to be used in slip-critical,
fully tightened non-slip critical, or snug tight connections, or in connections designed to slip.
For tolerances: If unusual tolerances for dimensional accuracy exist, these must be clearly
shown on the design drawings. Unusual tolerances are those which are more stringent than
tolerances specified in the general specification for the type of structure under consideration.
Typical tolerances are given in AISC publications ‘‘Code of Standard Practice for Steel
Buildings and Bridges,’’ ‘‘Specification for Structural Steel Buildings, Allowable Stress De-
sign and Plastic Design,’’ and ‘‘Load and Resistance Factor Design Specification for Struc-
FABRICATION AND ERECTION

2.3
tural Steel Buildings’’; in AASHTO publications ‘‘Standard Specifications for Highway
Bridges,’’ and ‘‘LRFD Bridge Design Specifications’’; and in ASTM A6 General Require-
ments for Delivery of Rolled Steel Plates, Shapes, Sheet Piling, and Bars for Structural Use.’’
The AISC ‘‘Code of Standard Practice for Steel Building and Bridges’’ shows tolerances in
a format that can be used by the work force fabricating or erecting the structure. Different,
unusual or restrictive tolerances often demand specific procedures in the shop and field. Such
special tolerances must be clearly defined prior to fabrication in a method that considers the
processes used in fabrication and erection. This includes clearly labeling architecturally ex-
posed structural steel and providing adjustment where necessary. One of the issues often
encountered in the consideration of tolerances in buildings is the relative horizontal location
of points on different floors, and the effect this has on parts that connect to more than one
floor, such as stairs. Room must be provided around these parts to accommodate tolerances.
Large steel buildings also move significantly as construction loads and conditions change.
Ambient environmental conditions also cause deflections in large structures.
For special material requirements: Any special material requirements such as testing or
toughness must be shown. Fracture critical members and parts must be designated. The AISC
specifications require that shapes defined as ASTM A6 Group 4 and Group 5, and those
built from plates greater than 2 in thick, that will be spliced with complete joint penetration
welds subject to tension, be supplied with a minimum Charpy V-notch toughness value. The
toughness value, and the location on the cross section for specimens, is given in the speci-
fications. This requirement also applies when Group 4 and Group 5 shapes, or shapes made
from plate greater than 2 in thick, are connected with complete joint penetration welds and
tension is applied through the thickness of the material. Other requirements may apply for
seismic structures.
2.2 CUTTING, SHEARING, AND SAWING
Steel shops are commonly organized into departments such as receiving, detail material,
main material cut-and-preparation, assembly and shipping. Many shops also have paint de-
partments. Material is received on trucks or by rail, off loaded, compared to order require-
ments, and stored by project or by size and grade. Material is received from the mill or

warehouse marked with the size, specification, grade, and heat number. The specification
and grade marks are maintained on the material that is returned to stock from production.
Material handling is a major consideration in a structural shop and organized storage is a
key to reducing handling.
Flame cutting steel with an oxygen-fed torch is one of the most useful methods in steel
fabrication. The torch is used extensively to cut material to proper size, including stripping
flange plates from a wider plate, or cutting beams to required lengths. The torch is also used
to cut complex curves or forms, such as those encountered in finger-type expansion devices
for bridge decks. In addition, two torches are sometimes used simultaneously to cut a member
to size and bevel its edge in preparation for welding. Also, torches may be gang-mounted
for simultaneous multiple cutting.
Flame-cutting torches may be manually held or mechanically guided. Mechanical guides
may take the form of a track on which is mounted a small self-propelled unit that carries
the torch. This type is used principally for making long cuts, such as those for stripping
flange plates or trimming girder web plates to size. Another type of mechanically guided
torch is used for cutting intricately detailed pieces. This machine has an arm that supports
and moves the torch. The arm may be controlled by a device following the contour of a
template or may be computer-controlled.
In the flame-cutting process, the torch burns a mixture of oxygen and gas to bring the
steel at the point where the cut is to be started to preheat temperature of about 1600
Њ
F. At
this temperature, the steel has a great affinity for oxygen. The torch then releases pure oxygen
2.4
SECTION TWO
under pressure through the cutting tip. This oxygen combines immediately with the steel.
As the torch moves along the cut line, the oxidation, coupled with the erosive force of the
oxygen stream, produces a cut about
1
⁄

8
in wide. Once cutting begins, the heat of oxidation
helps to heat the material.
Structural steel of certain grades and thicknesses may require additional preheat. In those
cases, flame is played on the metal ahead of the cut.
In such operations as stripping plate-girder flange plates, it is desirable to flame-cut both
edges of the plate simultaneously. This limits distortion by imposing shrinkage stresses of
approximately equal magnitude in both edges of the plate. For this reason, plates to be
supplied by a mill for multiple cutting are ordered with sufficient width to allow a flame cut
adjacent to the mill edges. It is not uncommon to strip three flange plates at one time using
4 torches.
Plasma-arc cutting is an alternative process for steel fabrication. A tungsten electrode may
be used, but hafnium is preferred because it eliminates the need for expensive inert shielding
gases. Advantages of this method include faster cutting, easy removal of dross, and lower
operating cost. Disadvantages include higher equipment cost, limitation of thickness of cut
to 1
1
⁄
2
in, slightly beveled edges, and a wider kerf. Plasma is advantageous for stainless
steels that cannot be cut with oxyfuel torches.
Shearing is used in the fabricating shop to cut certain classes of plain material to size.
Several types of shears are available. Guillotine-type shears are used to cut plates of mod-
erate thickness. Some plate shears, called rotary-plate shears, have a rotatable cutting head
that allows cutting on a bevel. Angle shears are used to cut both legs of an angle with one
stroke. Rotary-angle shears can produce beveled cuts.
Sawing with a high-speed friction saw is often employed in the shop on light beams and
channels ordered to multiple lengths. Sawing is also used for relatively light columns, be-
cause the cut produced is suitable for bearing and sawing is faster and less expensive than
milling. Some fabricators utilize cold sawing as a means of cutting beams to nearly exact

length when accuracy is demanded by the type of end connection being used. Sawing may
be done with cold saws, band saws, or in some cases, with hack saws or friction saws. The
choice of saws depends on the section size being cut and effects the speed and accuracy of
the cut. Some saws provide a cut adequate for use in column splices. The adequacy of
sawing is dependent on the maintenance of blades and on how the saw and work piece is
set up.
2.3 PUNCHING AND DRILLING
Bolt holes in structural steel are usually produced by punching (within thickness limitations).
The American Institute of Steel Construction (AISC) limits the thickness for punching to
the nominal diameter of the bolt plus
1
⁄
8
in. In thicker material, the holes may be made by
subpunching and reaming or by drilling. Multiple punches are generally used for large groups
of holes, such as for beam splices. Drilling is more time-consuming and therefore more
costly than punching. Both drill presses and multiple-spindle drills are used, and the flanges
and webs may be drilled simultaneously.
2.4 CNC MACHINES
Computer numerically controlled (CNC) machines that offer increased productivity are
used increasingly for punching, cutting, and other operations. Their use can reduce the time
required for material handling and layout, as well as for punching, cutting, or shearing. Such
FABRICATION AND ERECTION
2.5
machines can handle plates up to 30 by 120 in by 1
1
⁄
4
in thick. CNC machines are also
available for fabricating W shapes, including punching or drilling, flame-cutting copes, weld

preparation (bevels and rat holes) for splices and moment connections, and similar items.
CNC machines have the capacity to drill holes up to 1
9
⁄
16
in in diameter in either flanges
or web. Production is of high quality and accuracy.
2.5 BOLTING
Most field connections are made by bolting, either with high-strength bolts (ASTM A325 or
A490) or with ordinary machine bolts (A307 bolts), depending on strength requirements.
Shop connections frequently are welded but may use these same types of bolts.
When high-strength bolts are used, the connections should satisfy the requirements of the
‘‘Specification for Structural Joints Using ASTM A325 or A490 Bolts,’’ approved by the
Research Council on Structural Connections (RCSC) of the Engineering Foundation. Joints
with high strength bolts are designed as bearing-type, fully-tightened, loose-to-slip or slip-
critical connections (see Art. 5.3). Bearing-type connections have a higher allowable load or
design strength. Slip-critical connections always must be fully tightened to specified mini-
mum values. Bearing-type connections may be either ‘‘snug tight’’ or fully tightened de-
pending on the type of connection and service conditions. AISC specifications for structural
steel buildings require fully tensioned high-strength bolts (or welds) for certain connections
(see Art. 6.14.2). The AASHTO specifications require slip-critical joints in bridges where
slippage would be detrimental to the serviceability of the structure, including joints subjected
to fatigue loading or significant stress reversal. In all other cases, connections may be made
with ‘‘snug tight’’ high strength bolts or A307 bolts, as may be required to develop the
necessary strength. For tightening requirements, see Art. 5.14.
2.6 WELDING
Use of welding in fabrication of structural steel for buildings and bridges is governed by
one or more of the following: American Welding Society Specifications Dl.1, ‘‘Structural
Welding Code,’’ and D1.5, ‘‘Bridge Welding Code,’’ and the AISC ‘‘Specification for Struc-
tural Steel Buildings, ’’ both ASD and LRFD. In addition to these specifications, welding

may be governed by individual project specifications or standard specifications of agencies
or groups, such as state departments of transportation.
Steels to be welded should be of a ‘‘weldable grade,’’ such as A36, A572, A588, A514,
A709, A852, A913, or A992. Such steels may be welded by any of several welding pro-
cesses: shielded metal arc, submerged arc, gas metal arc, flux-cored arc, electroslag, electro-
gas, and stud welding. Some processes, however, are preferred for certain grades and some
are excluded, as indicated in the following.
AWS ‘‘Structural Welding Code’’ and other specifications require the use of written,
qualified procedures, qualified welders, the use of certain base and filler metals, and inspec-
tion. The AWS Dl.1 code exempts from tests and qualification most of the common welded
joints used in steel structures which are considered ‘‘prequalified’’. The details of such pre-
qualified joints are shown in AWS Dl.1 and in the AISC ‘‘Steel Construction Manual—
ASD’’ and ‘‘Steel Construction Manual—LRFD.’’ It is advantageous to use these joints where
applicable to avoid costs for additional qualification tests.
Shielded metal arc welding (SMAW) produces coalescence, or fusion, by the heat of
an electric arc struck between a coated metal electrode and the material being joined, or
base metal. The electrode supplies filler metal for making the weld, gas for shielding the
2.6
SECTION TWO
molten metal, and flux for refining this metal. This process is commonly known also as
manual, hand, or stick welding. Pressure is not used on the parts to be joined.
When an arc is struck between the electrode and the base metal, the intense heat forms
a small molten pool on the surface of the base metal. The arc also decomposes the electrode
coating and melts the metal at the tip of the electrode. The electron stream carries this metal
in the form of fine globules across the gap and deposits and mixes it into the molten pool
on the surface of the base metal. (Since deposition of electrode material does not depend on
gravity, arc welding is feasible in various positions, including overhead.) The decomposed
coating of the electrode forms a gas shield around the molten metal that prevents contact
with the air and absorption of impurities. In addition, the electrode coating promotes elec-
trical conduction across the arc, helps stabilize the arc, adds flux, slag-forming materials, to

the molten pool to refine the metal, and provides materials for controlling the shape of the
weld. In some cases, the coating also adds alloying elements. As the arc moves along, the
molten metal left behind solidifies in a homogeneous deposit, or weld.
The electric power used with shielded metal arc welding may be direct or alternating
current. With direct current, either straight or reverse polarity may be used. For straight
polarity, the base metal is the positive pole and the electrode is the negative pole of the
welding arc. For reverse polarity, the base metal is the negative pole and the electrode is the
positive pole. Electrical equipment with a welding-current rating of 400 to 500 A is usually
used for structural steel fabrication. The power source may be portable, but the need for
moving it is minimized by connecting it to the electrode holder with relatively long cables.
The size of electrode (core wire diameter) depends primarily on joint detail and welding
position. Electrode sizes of
1
⁄
8
,
5
⁄
32
,
3
⁄
16
,
7
⁄
32
,
1
⁄

4
, and
5
⁄
16
in are commonly used. Small-size
electrodes are 14 in long, and the larger sizes are 18 in long. Deposition rate of the weld
metal depends primarily on welding current. Hence use of the largest electrode and welding
current consistent with good practice is advantageous.
About 57 to 68% of the gross weight of the welding electrodes results in weld metal.
The remainder is attributed to spatter, coating, and stub-end losses.
Shielded metal arc welding is widely used for manual welding of low-carbon steels, such
as A36, and HSLA steels, such as A572 and A588. Though stainless steels, high-alloy steels,
and nonferrous metals can be welded with this process, they are more readily welded with
the gas metal arc process.
Submerged-arc welding (SAW) produces coalescence by the heat of an electric arc
struck between a bare metal electrode and the base metal. The weld is shielded by flux, a
blanket of granular fusible material placed over the joint. Pressure is not used on the parts
to be joined. Filler metal is obtained either from the electrode or from a supplementary
welding rod.
The electrode is pushed through the flux to strike an arc. The heat produced by the arc
melts adjoining base metal and flux. As welding progresses, the molten flux forms a protec-
tive shield above the molten metal. On cooling, this flux solidifies under the unfused flux as
a brittle slag that can be removed easily. Unfused flux is recovered for future use. About 1.5
lb of flux is used for each pound of weld wire melted.
Submerged-arc welding requires high currents. The current for a given cross-sectional
area of electrode often is as much as 10 times as great as that used for manual welding.
Consequently, the deposition rate and welding speeds are greater than for manual welding.
Also, deep weld penetration results. Consequently, less edge preparation of the material to
be joined is required for submerged-arc welding than for manual welding. For example,

material up to
3
⁄
8
in thick can be groove-welded, without any preparation or root opening,
with two passes, one from each side of the joint. Complete fusion of the joint results.
Submerged-arc welding may be done with direct or alternating current. Conventional
welding power units are used but with larger capacity than those used for manual welding.
Equipment with current ratings up to 4000 A is used.
The process may be completely automatic or semiautomatic. In the semiautomatic pro-
cess, the arc is moved manually. One-, two-, or three-wire electrodes can be used in automatic
FABRICATION AND ERECTION
2.7
operation, two being the most common. Only one electrode is used in semiautomatic oper-
ation.
Submerged-arc welding is widely used for welding low-carbon steels and HSLA steels.
Though stainless steels, high-alloy steels, and nonferrous metals can be welded with this
process, they are generally more readily welded with the gas-shielded metal-arc process.
Gas metal arc welding (GMAW) produces coalescence by the heat of an electric arc
struck between a filler-metal electrode and base metal. Shielding is obtained from a gas or
gas mixture (which may contain an inert gas) or a mixture of a gas and flux.
This process is used with direct or alternating current. Either straight or reverse polarity
may be employed with direct current. Operation may be automatic or semiautomatic. In the
semiautomatic process, the arc is moved manually.
As in the submerged-arc process, high current densities are used, and deep weld penetra-
tion results. Electrodes range from 0.020 to
1
⁄
8
in diameter, with corresponding welding

currents of about 75 to 650 A.
Practically all metals can be welded with this process. It is superior to other presently
available processes for welding stainless steels and nonferrous metals. For these metals,
argon, helium, or a mixture of the two gases is generally used for the shielding gas. For
welding of carbon steels, the shielding gas may be argon, argon with oxygen, or carbon
dioxide. Gas flow is regulated by a flowmeter. A rate of 25 to 50 ft
3
/hr of arc time is
normally used.
Flux-cored arc welding (FCAW) is similar to the GMAW process except that a flux-
containing tubular wire is used instead of a solid wire. The process is classified into two
sub-processes self-shielded and gas-shielded. Shielding is provided by decomposition of the
flux material in the wire. In the gas-shielded process, additional shielding is provided by an
externally supplied shielding gas fed through the electrode gun. The flux performs functions
similar to the electrode coatings used for SMAW. The self-shielded process is particularly
attractive for field welding because the shielding produced by the cored wire does not blow
off in normal ambient conditions and heavy gas supply bottles do not have to be moved
around the site.
Electroslag welding (ESW) produces fusion with a molten slag that melts filler metal
and the surfaces of the base metal. The weld pool is shielded by this molten slag, which
moves along the entire cross section of the joint as welding progresses. The electrically
conductive slag is maintained in a molten condition by its resistance to an electric current
that flows between the electrode and the base metal.
The process is started much like the submerged-arc process by striking an electric arc
beneath a layer of granular flux. When a sufficiently thick layer of hot molten slag is formed,
arc action stops. The current then passes from the electrode to the base metal through the
conductive slag. At this point, the process ceases to be an arc welding process and becomes
the electroslag process. Heat generated by resistance to flow of current through the molten
slag and weld puddle is sufficient to melt the edges at the joint and the tip of the welding
electrode. The temperature of the molten metal is in the range of 3500

Њ
F. The liquid metal
coming from the filler wire and the molten base metal collect in a pool beneath the slag and
slowly solidify to form the weld. During welding, since no arc exists, no spattering or intense
arc flash occurs.
Because of the large volume of molten slag and weld metal produced in electroslag
welding, the process is generally used for welding in the vertical position. The parts to be
welded are assembled with a gap 1 to 1
1
⁄
4
in wide. Edges of the joint need only be cut
squarely, by either machine or flame.
Water-cooled copper shoes are attached on each side of the joint to retain the molten
metal and slag pool and to act as a mold to cool and shape the weld surfaces. The copper
shoes automatically slide upward on the base-metal surfaces as welding progresses.
Preheating of the base metal is usually not necessary in the ordinary sense. Since the
major portion of the heat of welding is transferred into the joint base metal, preheating is
accomplished without additional effort.
2.8
SECTION TWO
The electroslag process can be used to join plates from 1
1
⁄
4
to 18 in thick. The process
cannot be used on heat-treated steels without subsequent heat treatment. AWS and other
specifications prohibit the use of ESW for welding quenched-and-tempered steel or for weld-
ing dynamically loaded structural members subject to tensile stresses or to reversal of stress.
However, research results currently being introduced on joints with narrower gaps should

lead to acceptance in cyclically loaded structures.
Electrogas welding (EGW) is similar to electroslag welding in that both are automatic
processes suitable only for welding in the vertical position. Both utilize vertically traveling,
water-cooled shoes to contain and shape the weld surface. The electrogas process differs in
that once an arc is established between the electrode and the base metal, it is continuously
maintained. The shielding function is performed by helium, argon, carbon dioxide, or
mixtures of these gases continuously fed into the weld area. The flux core of the electrode
provides deoxidizing and slagging materials for cleansing the weld metal. The surfaces to
be joined, preheated by the shielding gas, are brought to the proper temperature for complete
fusion by contact with the molten slag. The molten slag flows toward the copper shoes and
forms a protective coating between the shoes and the faces of the weld. As weld metal is
deposited, the copper shoes, forming a weld pocket of uniform depth, are carried continu-
ously upward.
The electrogas process can be used for joining material from
1
⁄
2
to more than 2 in thick.
The process cannot be used on heat-treated material without subsequent heat treatment. AWS
and other specifications prohibit the use of EGW for welding quenched-and-tempered steel
or for welding dynamically loaded structural members subject to tensile stresses or to reversal
of stress.
Stud welding produces coalescence by the heat of an electric arc drawn between a metal
stud or similar part and another work part. When the surfaces to be joined are properly
heated, they are brought together under pressure. Partial shielding of the weld may be ob-
tained by surrounding the stud with a ceramic ferrule at the weld location.
Stud welding usually is done with a device, or gun, for establishing and controlling the
arc. The operator places the stud in the chuck of the gun with the flux end protruding. Then
the operator places the ceramic ferrule over this end of the stud. With timing and welding-
current controls set, the operator holds the gun in the welding position, with the stud pressed

firmly against the welding surface, and presses the trigger. This starts the welding cycle by
closing the welding-current contactor. A coil is activated to lift the stud enough to establish
an arc between the stud and the welding surface. The heat melts the end of the stud and the
welding surface. After the desired arc time, a control releases a spring that plunges the stud
into the molten pool.
Direct current is used for stud welding. A high current is required for a very short time.
For example, welding currents up to 2500 A are used with arc time of less than 1 sec for
studs up to 1 in diameter.
(O. W. Blodgett, Design of Welded Structures, The James F. Lincoln Arc Welding Foun-
dation, Cleveland, Ohio.) See also Arts. 5.15 to 5.23.
2.7 CAMBER
Camber is a curvature built into a member or structure so that when it is loaded, it deflects
to a desired shape. Camber, when required, might be for dead load only, dead load and
partial live load, or dead load and full live load. The decision to camber and how much to
camber is one made by the designer.
Rolled beams are generally cambered cold in a machine designed for the purpose, in a
large press, known as a bulldozer or gag press, through the use of heat, or a combination of
mechanically applied stress and heat. In a cambering machine, the beam is run through a
multiple set of hydraulically controlled rollers and the curvature is induced in a continuous
FABRICATION AND ERECTION
2.9
operation. In a gag press, the beam is inched along and given an incremental bend at many
points.
There are a variety of specific techniques used to heat-camber beams but in all of them,
the side to be shortened is heated with an oxygen-fed torch. As the part is heated, it tries to
elongate. But because it is restrained by unheated material, the heated part with reduced
yield stress is forced to upset (increase inelastically in thickness) to relieve its compressive
stress. Since the increase in thickness is inelastic, the part will not return to its original
thickness on cooling. When the part is allowed to cool, therefore, it must shorten to return
to its original volume. The heated flange therefore experiences a net shortening that produces

the camber. Heat cambering is generally slow and expensive and is typically used in sections
larger than the capacity of available equipment. Heat can also be used to straighten or
eliminate warping from parts. Some of these procedures are quite complex and intuitive,
demanding experience on the part of the operator.
Experience has shown that the residual stresses remaining in a beam after cambering are
little different from those due to differential cooling rates of the elements of the shape after
it has been produced by hot rolling. Note that allowable design stresses are based to some
extent on the fact that residual stresses virtually always exist.
Plate girders usually are cambered by cutting the web plate to the cambered shape before
the flanges are attached.
Large bridge and roof trusses are cambered by fabricating the members to lengths that
will yield the desired camber when the trusses are assembled. For example, each compression
member is fabricated to its geometric (loaded) length plus the calculated axial deformation
under load. Similarly, each tension member is fabricated to its geometric length minus the
axial deformation.
2.8 SHOP PREASSEMBLY
When the principal operations on a main member, such as punching, drilling, and cutting,
are completed, and when the detail pieces connecting to it are fabricated, all the components
are brought together to be fitted up, i.e.,temporarily assembled with fit-up bolts, clamps, or
tack welds. At this time, the member is inspected for dimensional accuracy, squareness, and,
in general, conformance with shop detail drawings. Misalignment in holes in mating parts
should be detected then and holes reamed, if necessary, for insertion of bolts. When fit-up
is completed, the member is bolted or welded with final shop connections.
The foregoing type of shop preassembly or fit-up is an ordinary shop practice, routinely
performed on virtually all work. There is another class of fit-up, however, mainly associated
with highway and railroad bridges, that may be required by project specifications. These
may specify that the holes in bolted field connections and splices be reamed while the
members are assembled in the shop. Such requirements should be reviewed carefully before
they are specified. The steps of subpunching (or subdrilling), shop assembly, and reaming
for field connections add significant costs. Modern CNC drilling equipment can provide full-

size holes located with a high degree of accuracy. AASHTO specifications, for example,
include provisions for reduced shop assembly procedures when CNC drilling operations are
used.
Where assembly and reaming are required, the following guidelines apply:
Splices in bridge girders are commonly reamed assembled. Alternatively, the abutting
ends and the splice material may be reamed to templates independently.
Ends of floorbeams and their mating holes in trusses or girders usually are reamed to
templates separately.
For reaming truss connections, three methods are in use in fabricating shops. The partic-
ular method to be used on a job is dictated by the project specifications or the designer.
2.10
SECTION TWO
Associated with the reaming methods for trusses is the method of cambering trusses.
Highway and railroad bridge trusses are cambered by increasing the geometric (loaded)
length of each compression member and decreasing the geometric length of each tension
member by the amount of axial deformation it will experience under load (see Art. 2.7).
Method 1 (RT, or Reamed-template, Method ). All members are reamed to geometric an-
gles (angles between members under load) and cambered (no-load) lengths. Each chord is
shop-assembled and reamed. Web members are reamed to metal templates. The procedure
is as follows:
With the bottom chord assembled in its loaded position (with a minimum length of three
abutting sections), the field connection holes are reamed. (Section, as used here and in
methods 2 and 3, means fabricated member. A chord section, or fabricated member, usually
is two panels long.)
With the top chord assembled in its loaded position (with a minimum length of three
abutting sections), the field connection holes are reamed.
The end posts of heavy trusses are normally assembled and the end connection holes
reamed, first for one chord and then for the other. The angles between the end post and the
chords will be the geometric angles. For light trusses, however, the end posts may be treated
as web members and reamed to metal templates.

The ends of all web members and their field holes in gusset plates are reamed separately
to metal templates. The templates are positioned on the gusset plates to geometric angles.
Also, the templates are located on the web members and gusset plates so that when the
unloaded member is connected, the length of the member will be its cambered length.
Method 2 (Gary or Chicago Method ). All members are reamed to geometric angles and
cambered lengths. Each chord is assembled and reamed. Web members are shop-assembled
and reamed to each chord separately. The procedure is as follows:
With the bottom chord assembled in its geometric (loaded) alignment (with a minimum
number of three abutting sections), the field holes are reamed.
With the top chord assembled in its geometric position (with a minimum length of three
abutting sections), the holes in the field connections are reamed.
The end posts and all web members are assembled and reamed to each chord separately.
All members, when assembled for reaming, are aligned to geometric angles.
Method 3 (Fully Assembled Method). The truss is fully assembled, then reamed. In this
method, the bottom chord is assembled and blocked into its cambered (unloaded) alignment,
and all the other members are assembled to it. The truss, when fully assembled to its cam-
bered shape, is then reamed. Thus the members are positioned to cambered angles, not
geometric angles.
When the extreme length of trusses prohibits laying out the entire truss, method 3 can
be used sectionally. For example, at least three abutting complete sections (top and bottom
chords and connecting web members) are fully assembled in their cambered position and
reamed. Then complete sections are added to and removed from the assembled sections. The
sections added are always in their cambered position. There should always be at least two
previously assembled and reamed sections in the layout. Although reaming is accomplished
sectionally, the procedure fundamentally is the same as for a full truss assembly.
In methods 1 and 2, field connections are reamed to cambered lengths and geometric
angles, whereas in method 3, field connections are reamed to cambered lengths and angles.
To illustrate the effects of these methods on an erected and loaded truss, Fig. 2.1a shows by
dotted lines the shape of a truss that has been reamed by either method 1 or 2 and then fully
connected, but without load. As the members are fitted up (pinned and bolted), the truss is

forced into its cambered position. Bending stresses are induced into the members because
their ends are fixed at their geometric (not cambered) angles. This bending is indicated by