07 ballot items 41-60

7/27/2019 07 Ballot Items 41-60

1/107

2007 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 60 (REV 1)

SUBJECT: LRFD Bridge Design Specifications: Section 11, Table 11.5.6-1

TECHNICAL COMMITTEE: T-15 Foundations

REVISION ADDITION NEW DOCUMENT

DESIGN SPEC CONSTRUCTION SPEC MOVABLE SPEC

LRFR MANUAL OTHER

DATE PREPARED: 2/5/07

DATE REVISED: 7/10/07

AGENDA ITEM:

In Table 11.5.6-1, 11th row, 3rd column, regarding the resistance factor for bearing:

Delete Article 10.5 applies and using both the 2nd and 3 rd columns in the table in the 11th row, replace it with the

following:

Gravity and semi-gravity wallswith stiff footings 0.55Walls with flexible footings (e.g., MSE walls) 0.65

In Table 11.5.6-1, 10th row, revise the table subheading as follows:

Mechanically Stabilized Earth Walls, Gravity Walls and Semi-Gravity Walls

In Table 11.5.6-1, first row below the subheading Prefabricated Modular Walls, 3rd column, regarding the

resistance factor for bearing, delete Article 10.5 applies and replace it with 0.55.

OTHER AFFECTED ARTICLES:

None

BACKGROUND:

The resistance factor for bearing capacity for walls is referenced to the resistance factor provided in Section 10,

which specifies a resistance factor of 0.45 to 0.50. The load group used to calibrate that resistance factor in Section10 does not directly apply to walls, as for walls, the dominant load source is from earth pressure, not structure dead

and live load. Furthermore, for walls that are very flexible, such as reinforced soil walls, traditionally, an overall

safety factor for bearing resistance of 2.0 has been used. The bearing resistance factors in Chapter 10 were targetedto a safety factor of at least 2.5 to 3.0. For example, for bearing for walls with a flexible footing such as MSEwalls, the resistance factor determined through calibration by fitting, for FS = 2.0, is calculated as:

67.0

0.2123.0

3.123.05.1

1

FSV

EH

EVEH

EVEH

,rounded to 0.65.

7/27/2019 07 Ballot Items 41-60

2/107

For gravity walls with concrete footings (i.e., not flexible), a FS = 2.5 is typically used for bearing resistance. In

this case, becomes:

55.0

5.2123.0

3.123.05.1

1

FSEV

EH

EVEH

EVEH

.

The proposed changes to this resistance factor will make the design of walls consistent with past practice as

specified in the AASHTO Standard Specifications for Highway Bridges, 2002, and FHWA design manuals.

ANTICIPATED EFFECT ON BRIDGES:

Footing size for walls should be reduced relative to what would be required in the current LRFD specifications due

to the higher resistance factors, but will be consistent with what was obtained in previous allowable stress designpractice (i.e., the AASHTO Standard Specifications for Highway Bridges, 2002).

REFERENCES:

None

OTHER:

None

7/27/2019 07 Ballot Items 41-60

3/107

2007 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 41 (REVISION 1)

SUBJECT: Standard Specifications for Structural Supports for Highway Signs, Luminaires andTraffic Signals: Revisions to Section 11

TECHNICAL COMMITTEE: T-12 - Structural Supports for Highway Signs, Luminaires and

Traffic Signals



LRFR MANUAL OTHER AASHTO Standard Specifications for Structural Supports forHighway Signs, Luminaires and Traffic Signals



AGENDA ITEM:

Replace Section 11 in its entirety. See Attachment C for proposed Section 11.

The majority of changes made in this section are editorial in nature such as corrections to grammar, defining terms

to be consistent with the LRFD Specifications, the rounding of coefficients to two decimal places, and minorclarifications to various articles.

The following technical changes are made.

Section 11.4 The provisions to fatigue are expanded to include non-cantilevered structures. NCHRP 494 is the

basis for the fatigue design provisions for non-cantilevered support structures.

Table 11-1 Fatigue importance factors are included for non-cantilevered structures based on the work of NCHRP494.

Section 11.6 Additional guidance is provided as commentary for the selection of the fatigue category.

Section 11.7.2 Tapered poles are required to be investigated for vortex shedding. Vortex shedding has been

observed in tapered lighting poles, and studies have shown that tapered poles can experience vortex shedding insecond or third mode vibrations. Those vibrations can lead to fatigue problems.

Section 11.7.2 The drag coefficient to be used in the calculation of the equivalent static pressure for vortex

shedding is clarified to be based on the critical wind velocity.

Section 11.7.3 The drag coefficient to be used in the calculation of the equivalent static pressure for natural wind

gust is clarified to be based on the yearly mean wind velocity of 5 m/s or (11.2 mph).

Section 11.7.4 The drag coefficient to be used in the calculation of the equivalent static pressure for truck induced

gust is clarified to be based on a truck speed of 30 m/s or (65 mph).

Section 11.8 Guidance on the allowable deflection for non-cantilevered structures is provided in the commentarybased on the work of NCHRP 494.

7/27/2019 07 Ballot Items 41-60

4/107

OTHER AFFECTED ARTICLES:None

BACKGROUND:

The revised section is the result of work completed under NCHRP 20-07 Task 209.

ANTICIPATED EFFECT ON BRIDGES:None

REFERENCES:

None

OTHER:

None

7/27/2019 07 Ballot Items 41-60

5/107

1

ATTACHMENT C 2007 AGENDA ITEM 41 T-12 (REVISION 1)

Section 11:

Fatigue Design

SPECIFICATIONS COMMENTARY

11.1 SCOPE

This section contains provisions for the fa-tigue design of cantilevered and non-cantileveredsteel and aluminum structural supports for high-

way signs, luminaires, and traffic signals.

This section focuses on fatigue, which is de-fined herein as the damage that may result in frac-ture after a sufficient number of stress fluctuations.It is based on NCHRP Report 412, Fatigue Resis-

tant Design of Cantilevered Signal, Sign and LightSupports (Kaczinski et al. 1998), NCHRP Report

469,Fatigue-Resistant Design of Cantilever Signal,Sign, and Light Supports (Dexter and Ricker 2002)and NCHRP Report 494, Structural Supports for

Highway Signs, Luminaries, and Traffic Signals(Fouad et al 2003). The study focused on criticalsupport structures that show susceptibly to fatigue

failures. A continuation of the project is underwayto further refine the proposed design criteria.

11.2 DEFINITIONS

C o n s t a n t- a m p l i t u d e f a t i g u e l i m i t ( CA F L ) a stress range below which a fatigue life appears to be infi-nite, also known as an endurance limit nominal stress range below which a particular fatigue detail can

withstand an infinite number of repetitions without fatigue failure.

Fatiguedamage resulting in fracture caused by stress fluctuations.

In-plane bendingbending in-plane for the main member (column). At the connection of an arm or armsbuilt-up box to a vertical column, the in-plane bending stress range in the column is a result of galloping or

truck-induced gust loads on the arm and/or arms attachments.

Limit state wind load effect a specifically defined load criteria.

Load bearing attachmentattachment to main member where there is a transverse load range in the at-tachment itself in addition to any primary stress range in the main member.

Non-load bearing attachmentattachment to main member where the only significant stress range is the

primary stress in the main member.

Out-of-plane bendingbending out-of-plane for the main member (column). At the connection of an arm

or arms built-up box to a vertical column, the out-of-plane bending stress range in the column is a result ofnatural wind gust loads on the arm and the arms attachments.

7/27/2019 07 Ballot Items 41-60

6/107

Standard Specifications for Structural Supports for H ighway Signs, Lumin aires and Traffi c Signals


2

Pressure rangemagnitude of force, in terms of pressure, due to of a limit state wind load effect that pro-duces a stress range.

Stress range magnitude of stress fluctuations. the algebraic difference between extreme stresses used

in fatigue design.

Yearly mean wind velocity long-term average of the wind speed for a given area.

11.3 NOTATIONS

b = flat-to-flat width of a multisided section (m, ft)

Cd = appropriate drag coefficient from Section 3, Loads, for given attachment or memberd = diameter of a circular section (m, ft)

D = inside diameter of exposed end of female section for slip-joint splice (mm, in)E = modulus of elasticity (MPa, ksi)fn = first natural frequency of the structure (cps)fn1 = first modal frequency (cps)

(F )n = fatigue strength (CAFL) (MPa, ksi)g = acceleration of gravity (9810 mm/s

2, 386 in/s

2)

H = effective weld throat (mm, in)I = moment of inertia (mm

4, in

4)

Iav g = average moment of inertia for a tapered pole (mm4, in

4)

Ito p = moment of inertia at top of tapered pole (mm4, in

4)

Ib o t t o m = moment of inertia at bottom of tapered pole (mm4

, in4

)IF = importance factors applied to limit state wind load effects to adjust for the desired level of

structural reliabilityL = length of the pole (Article 11.7.2) (mm, in)L = slip-splice overlap length (example 1 of Figure 111) (mm, in)

L = length of reinforcement at handhole (example 13 of Figure 111) (mm, in)L = length of longitudinal attachment (examples 12, 14 and 15 of Figure 111) (mm, in)

PG = galloping-induced vertical shear pressure range (Pa, psf)PNW = natural wind gust pressure range (Pa, psf)PTG = truck-induced gust pressure range (Pa, psf)PVS = vortex shedding-induced pressure range (Pa, psf)

r = radius of chord or column (mm, in)R = transition radius of longitudinal attachment (mm, in)

Sn = Strouhal numberSR = nominal stress range of the main member or branching member (MPa, ksi)t = thickness (mm, in)

tb = wall thickness of branching member (mm, in)tc = wall thickness of main member (column) (mm, in)tp = plate thickness of attachment (mm, in)

Vc = critical wind velocity for vortex shedding (m/s, ft/smph)Vm e a n = yearly mean wind velocity for a given area (m/s, mph)VT = truck speed for truck induced wind gusts (m/s, mph)

W = weight of the luminaire (N, k)w = weight of the pole per unit length (N/mm, k/in) = damping ratio = angle of transition taper of longitudinal attachment (example 14 of Figure 111) (deg) = ovalizing parameter for bending in the main member (note b of Table 112)F = constant amplitude fatigue limit stress range (MPa, ksi) = indication of stress range in member resulting from applicable axial loadings or moments

7/27/2019 07 Ballot Items 41-60

7/107

Section 11: Fatigue Design


3

11.4 APPLICABLE STRUCTURE TYPES

Design for fatigue shall be required for the

following type structures:

a) overhead cantilevered sign structures,

b) overhead cantilevered traffic signalstructures,

c) high-level, high-mast lighting structures,

d) overhead non-cantilevered sign structures,and

e) overhead non-cantilevered traffic signal

structures.

NCHRP Report 412 is the basis for the fa-

tigue design provisions for cantilevered structures.NCHRPReport 494 is the basis for the fatigue de-sign provisions for non-cantilevered support struc-

tures. The fatigue design procedures outlined inthis section may be applicable to steel and alumi-num structures in general. However, only specific

types of structures are identified for fatigue designin this article. Common lighting poles and roadsidesigns are not included because since they are

smaller structures and normally have not exhibitedfatigue problems. An exception would be square

lighting poles, as they have exhibited poor fatigueperformance. Square cross-sections have beenmuch more prone to fatigue problems than roundcross-sections. Caution should be exercised re-

garding the use of square lighting poles even whena fatigue design is performed. The provisions of

this section are not applicable for the design ofspan-wire (strain) poles.

In general, overhead cantilevered sign andtraffic signal structures should be designed for fa-tigue due to individual loadings from galloping,

natural wind gusts, and truck-induced wind gusts.High-level lighting structures should be designedfor fatigue for loadings from natural wind gusts.

Vortex shedding should be considered for single-member cantilevered members that have tapers

less than 0.0117 m/m (0.14 in/ft), such as lightingstructures or mast arms without attachments.

NCHRP Report 412, Fatigue Resistant Design

of Cantilevered Signal, Sign and Light Supports(Kaczinski et al. 1998) is the basis for the fatigue

design provisions for cantilevered structures. Otherstructures, including overhead bridge support

structures for signs and signals, are also suscepti-ble to fatigue damage. Some of the design prov i-sions of this section can also be applicable to non-cantilevered structures. A research project is cur-

rently underway to develop complete fatigue de-sign provisions for noncantilevered support struc-tures.

7/27/2019 07 Ballot Items 41-60

8/107



4

11.5 DESIGN CRITERIA

Cantilevered and non-cantilevered supportstructures shall be designed for fatigue to resist

each of the applicable equivalent static wind loadeffects specified in Article 11.7, and modified bythe appropriate importance factors given in Article

11.6. Stresses due to these loads on all compo-

nents, mechanical fasteners, and weld detailsshall be limited designed to satisfy the require-

ments of their respective detail categories withinthe constant-amplitude fatigue limits (CAFL) pro-vided in Article 11.9. Table 113. A summary of

typical fatigue-sensitive cantilevered supportstructure connection details is presented in Table112 and illustrated in Figure 111.

Accurate load spectra and life prediction tech-niques for defining fatigue loadings are generally

not available. The assessment of stress fluctua-tions and the corresponding number of cycles forall wind-induced events (lifetime loading histogram)

is practically impossible. With this uncertainty, the

design of cantilevered sign, luminaire, and trafficsignal supports for a finite fatigue life becomes im-

practical. Therefore, an infinite life fatigue designapproach is recommended and is consideredsound practice. Fatigue stress limits are It is gen-

erally based on the constant-amplitude fatigue limit(CAFL). The CAFL values provided in Table 113are approximately the same as those given in Ta-

ble 10.3.1.A of the Standard Specifications forHighway Bridges(AASHTO 1996).

An infinite-life fatigue approach was devel-oped in an experimental study that considered

several critical welded details (Fisher et al. 1993).Theinfinite -life fatigue approach can be used whenthe number of wind load cycles expected during

the lifetime of the structures is greater than thenumber of cycles at the CAFL. This is particularlythe case for structural supports where the wind

load cycles in 25 years or greater lifetimes are ex-pected to exceed 100 million cycles, whereas typi-cal weld details reach the CAFL at 10 to 20 million

cycles.

Fatigue critical details should be are designed

with nominal stress ranges that are below the ap-propriate CAFL. To assist designers, a categoriza-

tion of typical cantilevered support structure detailsto based on the existing AASHTO and AmericanWelding Society (AWS) fatigue design categories

is are provided in Table 112 and Figure 111.The above referenced details were producedbBased on a review of state departments of trans-

portation standard drawings and manufacturersliterature, the above referenced list of typical canti-levered support structure connection details was

produced. This list should not be considered as acomplete set of all possible connection details, butrather it is intended to remove the uncertainty as-

sociated with applying the provisions of the Stan-dard Specifications for Highway Bridges to the fa-tigue design of cantilevered support structures.

Choice of details improves the fatigue resistance ofthese structures, and it can eliminate or reduceincreases in member size required for less fatigue-

resistant details.

7/27/2019 07 Ballot Items 41-60

9/107



5

This detailed categorization of fatigue-sensitive connection details can be used by de-signers and fabricators to produce more fatigue-

resistant cantilevered support structures. Properdetailing will improve the fatigue resistance of

these structures, and it can eliminate or reduceincreases in member size required for less fatigue-resistant details.

The notes for Table 112 specify the use ofStress Category K2. This stress category corre-

sponds to the category for cyclic punching shearstress in tubular members specified by the AWSStructural Welding Code D1.1Steel. Fatigue de-

sign for the columns wall under this condition mayrequire sizes of the built-up box connection or col-umn wall thicknesses that are excessive for practi-

cal use. For this occurrence, an adequate fatigue-resistant connection other than the built-up boxshown in Figure 111 should be considered.

Fatigue testing has shown the advantage of

ring-stiffeners that completely encircle a pole rela-tive to a built-up box connection. For built-up boxconnections, it is recommended that the width of

the box be the same as the diameter of the column(i.e., the sides of the box are tangent to the sidesof the column).

Regarding full-penetration groove-weldedtube-to-transverse plate connections, NCHRP Re-

port 412 did not fully investigate the effects fromthe possible use of additional reinforcing filletwelds. Additional research and testing of these

types of detail configurations are needed to sup-port future updates of this section.

Stress categories in Table 11-2 for weld ter-minations at the end of longitudinal stiffeners were

based, in part, on assigned categories for attach-ments in the AASHTO Bridge Specifications. Re-

cent Ffatigue testing of many fillet-welded tube-to-longitudinal stiffener connections indicate that theangle of intersection (A), the transitional radius tothe pole wall (R), the length of the stiffener (L), and

the ratio of the stiffener thickness to pole wallthickness, for example, all have effects on the fa-tigue life of the detail. Some tube-to-stiffener con-

nections have a potential to develop very highstress concentrations in the tube wall in the vicinityof the weld termination at the end of longitudinal

stiffeners. Testing on poles having wall thicknessless than 6 mm () indicates that longitudinal stif-feners yielded little or no improvement of the fa-

tigue performance of the connection (Koenigs et al.2003). Until further research can give reliable es-

7/27/2019 07 Ballot Items 41-60

10/107



6

timates of the effects of stiffeners, all welds termi-nating at the end of longitudinal stiffeners shall beclassified as stress category E.

Equal leg welds in socket connections have

been shown by fatigue testing to have a fatiguestrength less than stress category E. The fatiguestrength of a socket-welded connection can be im-

proved by using an unequal leg fillet weld.

11.6 FATIGUE IMPORTANCE FACTORS

7/27/2019 07 Ballot Items 41-60

11/107



7

An importance factor, IF, that accounts forthe risk degree of hazard to traffic and damage toproperty shall be applied to the limit state wind

load effects specified in Article 11.7. Importancefactors for cantilevered traffic signal, sign, and

luminaire support structures exposed to the fourwind load effects are presented in Table 111.

Importance factors are introduced into theSpecifications to adjust the level of structural reli-ability of cantilevered and non-cantilevered support

structures. Importance factors should be deter-mined by the owner. For combined structures,

such aswhere traffic signals and luminaires arecombined joined structures, the use of the moreconservative importance factor is recommended.

Three categories of cantilevered supportstructures are presented in Table 111. Structures

classified as category I present a high hazard inthe event of failure and should be designed to re-sist rarely occurring wind loading and vibration

phenomena. It is intended that only the most crit i-cal cantilevered support structures be classified ascategory I. Some examples of structures that

should be considered for category I classificationinclude the following: large sign structures (includ-ing variable message signs [VMS]), traffic signal

structures with long mast arms, and high-levellighting poles in excess of 30 m (98 ft) that are in-

stalled on highways where the vehicle speed issuch that the consequences of excessive deflec-tion or a collision with a fallen structure is intoler-

able. Category II and III structures are not lesslikely to experience the full limit state wind loadsassociated with category I. If category II or III canti-

levered support structures experience the limitstate loads over a period of time, they would beexpected to experience fatigue damage. Sound

engineering judgment shall be used in the classifi-cation process.

The importance categories and importance

factors (rounded to the nearest 0.05) are resultsfrom NCHRP Reports 469 and 494. Three catego-ries of support structures are presented in Table

111. Structures classified as Category I present ahigh hazard in the event of failure and should be

designed to resist rarely occurring wind loadingand vibration phenomenon. It is recommendedthat all structures without effective mitigation de-vices on roadways with a speed limit in excess of

60 km/hr (35 mph) and average daily traffic (ADT)exceeding 10,000 or average daily truck traffic(ADTT) exceeding 1000 should be classified as

Category I structures. ADT and ADTT are for onedirection regardless of the number of lanes.

Structures without mitigation devicesshouldmaybe classified as Category I if any of the follow-

ing apply:

1. Cantileveredsignstructures with a span in ex-

7/27/2019 07 Ballot Items 41-60

12/107



8

cess of 16 m (50 ft) or high-mast towers inexcess of 30 m (100 ft).

2. Large sign structures, both cantilevered and

non-cantilevered, including va riable messagesigns.

23. Structures lLocated in an area that is known tohave wind conditions that are conducive tovibration.

Structures should be classified as Category IIIif they are located on roads with speed limits of 60

km/hr (35 mph) or less. Structures that are locatedsuch that a failure will not affect traffic may beclassified as Category III.

All structures not explicitly meeting the Cat e-gory I or Category III criteria should be classified

as Category II.

Maintenance and inspection programs should be

considered integral to the selection of the fatigueimportance category.

There are many factors that affect the selection ofthe fatigue category and engineering judgment is

required.

7/27/2019 07 Ballot Items 41-60

13/107



9

Table 111. Fatigue Importance Factors, IF

Fatigue Category Importance Factor, IFGalloping Vortex Shedding Natural Wind Gusts Truck-Induced Gusts

I SignTraffic Signal

Lighting

1.01.0

x

x*x*

1.0

1.01.01.0

1.01.0

x

II Sign

Traffic SignalLighting

0.70 .65

0.65x

x*

x*0.65

0.85 .75

0.800.75 .72

0.90 .89

0.85 .84x

Cantilevered

III SignTraffic Signal

Lighting

0.40 .310.30

x

x*x*

0.30

0.70 .490.55 .59

0.50 .44

0.80 .770.70 .68

x

I Sign

Traffic Signal

x

x

x*

x*

1.0

1.0

1.0

1.0

II SignTraffic Signal

xx

x*x*

0.850.80

0.900.85N

on-

Cantilevered

III Sign

Traffic Signal

x

x

x*

x*

0.70

0.55

0.80

0.70

Note:x - Structure is not susceptible to this type of loading.* - Overhead cantilevered and non -cantilevered sign and traffic signal components are susceptible to vortex shed-

ding prior to placement of the signs and traffic signal heads, i.e., during construction.

11.7 FATIGUE DESIGN LOADS

To avoid large-amplitude vibrations and topreclude the development of fatigue cracks invarious connection details and at other critical

locations, cantilevered and non-cantilevered sup-port structures shall be designed to resist each of

the following applicable limit state equivalent stat-ic wind loads acting separately. These loads shallbe used to calculate nominal stress ranges at

near fatigue-sensitive connection details, as de-scribed in Article 11.5 and deflections for servicelimits described in Article 11.8. The calculated

nominal stress range shall not exceed the CAFL

values given in Table 113 for a particular con-nection detail.

In lieu of using the equivalent static pres-sures provided in this specification, a dynamic

analysis of the structure may be performed usingappropriate dynamic load functions derived fromreliable data.

Cantilevered and non-cantilevered supportstructures are exposed to several wind phenomenathat can produce cyclic loads. Vibrations associ-

ated with these cyclic forces can become signifi-cant. NCHRP Report 412 has identified galloping,

vortex shedding, natural wind gusts, and truck-induced gusts as wind-loading mechanisms thatcan induce large amplitude vibrations and/or fa-

tigue damage in cantilevered traffic signal, sign,and light support structures. NCHRP Report 494identified natural wind gusts and truck-induced

gusts as wind-loading mechanisms that can induce

large amplitude vibrations and/or fatigue damagein non-cantilevered traffic signal and sign support

structures. The amplitude of vibration and resultingstress ranges are increased by the low levels ofstiffness and damping possessed by many of these

structures. In some cases, the vibration is only aserviceability problem because motorists cannotclearly see the mast arm attachments or are con-

cerned about passing under the structures. In othercases, where deflections may or may not be con-

sidered excessive, the magnitudes of stressranges induced in these structures have resulted inthe development of fatigue cracks at various con-

nection details including the anchor bolts.

7/27/2019 07 Ballot Items 41-60

14/107



10

The wind-loading phenomena specified in thissection possess the greatest potential for creatinglarge amplitude vibrations in cantilevered support

structures. In particular, galloping and vortex shed-ding are aeroelastic instabilities that will typically

induce vibrations at the natural frequency of thestructure (i.e., resonance). These conditions canlead to fatigue failures in a relatively short period of

time.

Design pressures for each of the four possible

fatigue wind-loading mechanisms are presented asan equivalent static wind pressure range, or ashear stress range in the case of galloping. These

pressure (or shear stress) ranges should be ap-plied to the structure as prescribed by this sectionin a simple static analysis to determine stress

ranges at near fatigue-sensitive details. In lieu ofdesigning for galloping or vortex-shedding limitstate fatigue wind load effects, mitigation devices

may be used as approved by the owner. Mitigationdevices are discussed in NCHRP Reports 412 and

469.

11.7.1 Galloping

Overhead cantilevered sign and traffic signalsupport structures shall be designed for gallop-ing-induced cyclic loads by applying an equiva-

lent static shear pressure vertically to the surfacearea, as viewed in normal elevation of all sign

panels and/or traffic signal heads and backplatesrigidly mounted to the cantilevered horizontalsupport. The magnitude of this vertical shear

pressure range shall be equal to the following:

FG IP 1000 (Pa) Eq. 111

P IG F21 (psf)In lieu of designing to resist periodic gallop-

ing forces, cantilevered sign and traffic signal

structures may be erected with approved effectivevibration mitigation devices. Vibration mitigationdevices should be approved by the owner, and

they should be based on historical or researchverification of its vibration damping characteris-tics.

Alternatively, for traffic signal structures,the owner may choose to install approved vibra-

tion mitigation devices if structures exhibit displaya galloping problem. The mitigation devicesshould must be installed as quickly as possible

after the galloping problem appears.

Galloping, or Den Hartog instability, results inlarge amplitude, resonant oscillations in a planenormal to the direction of wind flow. It is usually

limited to structures with nonsymmetrical cross-sections, such as sign and traffic signal structures

with attachments to the horizontal cantileveredarm. Structures without attachments to the cantile-vered horizontal support are not susceptible to gal-

loping-induced wind load effects.

The results of wind tunnel (Kaczinski et al.

1998) and water tank (McDonald et al. 1995) test-ing, as well as the oscillations observed on cantile-vered support structures in the field, are consistent

with the characteristics of the galloping phenom-ena. These characteristics include the sudden on-set of large-amplitude, across-wind vibrations that

increase with increases in wind velocity. It is impor-tant to note, however, that gGalloping is typicallynot caused by wind applied to the support structure

members, but rather applied by to the attachmentsto the horizontal cantilevered arm, such as signs

and traffic signals.

The geometry and orientation of these at-

tachments, as well as the wind direction, directlyinfluence the susceptibility of cantilevered supportstructures to galloping. Traffic signals are more

susceptible to galloping when configured with a

7/27/2019 07 Ballot Items 41-60

15/107



11

The owner may choose to exclude gal-loping loads for the fatigue design of overheadcantilevered sign support structures with quadri-

chord (i.e., four-chord) horizontal trusses.

backplate. In particular, traffic signal attachmentsconfigured with or without a backplate are moresusceptible to galloping when subject to flow from

the rear. Galloping of sign attachments is inde-pendent of aspect ratio and is more prevalent with

wind flows from the front of the structure.

By conducting wind tunnel tests and analytical

calibrations to field data and wind tunnel test re-sults, an equivalent static vertical shear of 1000 Pa(21 psf) was determined for the galloping phe-

nomenona. This vertical shear range should beapplied to the entire frontal area of each of the signand traffic signal attachments in a static analysis to

determine stress ranges at critical connection de-tails. For example, if a 2.5 m by 3.0 m (8 ft by 10 ft)sign panel is mounted to a horizontal mast arm, a

static force of 7500 x IF, N (1680 x IF, lb) should beapplied vertically to the structure at the area cen-troid center of gravity of the sign panel.

A pole with multiple horizontal cantilevered

arms may be designed for galloping loads appliedseparately to each individual arm, and need notconsider galloping simultaneously occurring on

multiple arms.

Overhead cantilevered sign support struc-

tures with quadri-chord horizontal trusses do notappear to be susceptible to galloping because oftheir inherent high degree of three-dimensional

stiffness.

Two possible means exist to mitigate gallop-

ing-induced oscillations in cantilevered supportstructures. The dynamic properties of the structureor the aerodynamic properties of the attachments

can be adequately altered to mitigate galloping.The installation of a device providing positive aero-

dynamic damping can be used to alter the struc-tures response from the aerodynamic effects on

the attachments.

A method of providing positive aerodynamicdamping to a traffic signal structure involves install-

ing a sign blank mounted horizontally and directlyabove the traffic signal attachment closest to thetip of the mast arm. This method has been shown

to be effective in mitigating galloping-induced vi-brations on traffic signal support structures with

horizontally-mounted traffic signal attachments(McDonald et al. 1995). For vertically-mounted traf-fic signal attachments, a sign blank mounted hori-

zontally near the tip of the mast arm has mitigatedlarge amplitude galloping vibrations occurring intraffic signal support structures. This sign blank is

7/27/2019 07 Ballot Items 41-60

16/107



12

placed adjacent to a traffic signal attachment, anda separation exists between the sign blank and thetop of the mast arm. In both cases, the sign blanks

are required to provide a sufficient surface area formitigation to occur. However, the installation of

sign blanks may influence the design of structuresfor truck-induced wind gusts by increasing the pro-jected area on a horizontal plane. NCHRP Reports

412 and 469 provides additional discussion on thispossible mitigation device, and on galloping sus-ceptibility and mitigation.

11.7.2 Vortex Shedding

Nontapered High-level, high-mast lighting

structures shall be designed to resist vortexshedding-induced loads for critical wind velocitiesless than approximately 20 m/s (65 fps; 45 mph).High-level, high-mast lighting structures that have

tapers less than 0.0117 m/m (0.14 in/ft) shall berequired to resist vortex shedding-induced loads.

The critical wind velocity, Vc (m/s, mph), atwhich vortex shedding lock-in can occur may be

calculated as follows:

for circular sections

n

n

cS

dfV (m/s) Eq. 112

n

nc

S

df.V 680 (mph)

for multisided sections

Vf b

Sc

n

n

(m/s) Eq. 113

n

n

cS

bf.V 680 (mph)

where fn is the first a natural frequency of thestructure (cps); dand bare the diameter and flat-

to-flat width of the horizontal mast arm or poleshaft for circular and multi-sided sections (m, ft),

respectively; and Snis the Strouhal number. TheStrouhal number shall be taken as 0.18for circu-lar sections, 0.15 for multisided sections, and0.11for square or rectangular sections. For a ta-

pered pole, d and b are the average diameterand width.

The equivalent static pressure range to beused for the design of vortex shedding-induced

The shedding of vortices on alternate sides of

a member may result in resonant oscillations in aplane normal to the direction of wind flow. Typicalnatural frequencies and member dimensions pre-clude the possibility of most cantilevered sign and

traffic signal support structures from being suscep-tible to vortex shedding-induced vibrations.

Cantilevered mast arms and lighting struc-tures that have tapers less than 0.0117 m/m (0.14

in/ft) may be required by the owner or designer toresist vortex shedding-induced loads.

NCHRP Report 469 shows that poles withtapers exceeding 0.0117 m/m (0.14 in/ft) can alsoexperience vortex shedding in lighting structures.

Observations and studies indicate that taperedpoles can experience vortex shedding in second or

third mode vibrations and that those vibrations canlead to fatigue problems. Procedures to considerhigher mode vortex shedding on tapered poles are

demonstrated in NCHRPReport 469.

Structural elements exposed to steady, uni-

form wind flows will shed vortices in the wake be-hind the element in a pattern commonly referred to

as a von Karmen vortex street. When the fre-quency of vortex shedding approaches one of thenatural frequencies of the structure, usually the firstmode (or higher modes as demonstrated in

NCHRP Report 469), significant amplitudes of vi-bration can be caused by a condition termed lock-in. The critical velocity at which lock-in will occurs

is defined by the Strouhal relationship:

n

n

cS

dfV Eq. C 111

For the first mode of vibration, a lower-boundwind speed can be established for traffic signal and

7/27/2019 07 Ballot Items 41-60

17/107



13

loads shall be:

PV C I

vsc d F 0 613

2

2.

(Pa) Eq. 114

2002560 2

Fdcvs

ICV.P (psf)where Vcis expressed in m/s (ft/smph); Cd is thedrag coefficient as specified in Section 3, Loads,

which is based on the critical wind velocity Vc;

and is the damping ratio, which is may beconservatively estimated as 0.005.

The equivalent static pressure range PVSshall be applied transversely to poles (i.e., hori-

zontal direction) and horizontal mast arms (i.e.,vertical direction).

In lieu of designing to resist periodic vortexshedding forces, effective approved vibration mi-

tigation devices may be used.

sign structures. Although vortices are shed at lowwind velocities for wind speeds less than 5 m/s (16fps, 11 mph), the vortices do not impart sufficient

energy to excite most structures. Typical naturalfrequencies and member diameters for sign and

traffic signal support structures result in criticalwind velocities well below the 5 m/s (16 fps, 11 mph)threshold for the occurrence of vortex shedding.

Because of extremely low levels of damping inher-ent in many nontapered support structures, vortexshedding may significantly excite resonant vibra-

tion. At wind speeds greater than about 20 m/s (65fps, 45 mph) enough natural turbulence is gener-ated to disturb the formation of vortices. Because Vc is

relatively low, the largest values ofCd forthe supportcomponents maybe conservatively used.

Horizontal arms may be susceptible to vortexshedding before sign and signal heads are at-

tached, i.e., during construction. Although possi-ble, recent tests (Kaczinski et al. 1998, McDonaldet al. 1995) have indicated that the occurrence of

vortex shedding from attachments to cantileveredsign and traffic signal support structures is not criti-cal. In fact,tThese attachments are more suscepti-

ble to galloping-induced vibrations. Finally, supportstructures composed of tapered members do notappear susceptible to vortex-induced vibrations

when tapered at least 0.0117 m/m (0.14 in/ft). Thedimensions of most tapered members result in crit-ical wind velocities below the threshold velocity;

and, furthermore, any vortices that may form arecorrelated over a short length of the member, and

they consequently generate insignificant vortex-shedding forces.

Calculation of the first modal frequency forsimple pole structures (i.e., without mast arms) canbe computed accomplished using the following

equations:

41

75.1

wLEI gf n Eq. C 112

(without luminaire mass)

n1 3 4

1 .7 3 2 E Ig f

2 W L 0 .2 3 6 w LEq. C 113

(with luminaire mass)

where W is the weight of the luminaire (N, k), w isthe weight of the pole per unit length (N/mm, k/in),g is the acceleration of gravity (9810 mm/s

2, 386

in/s2), L is the length of the pole (mm, in), and I is

the moment of inertia of the pole (mm4, in

4). For

tapered poles, Iavg is substituted forI, where:

7/27/2019 07 Ballot Items 41-60

18/107



14

II I

a v g

t o p b o t to m 2

Eq. C 114

Ito p is the moment of inertia at the tip of the pole

andIb o t t o m is the moment of inertia at the bottom ofthe pole.

Determining tThe first modal frequency forpoles with mast arms, however, is best accom-

plished by a finite element based modal analysis.The mass of the luminaire/mast arm attachmentsshall be included in the analysis to determine the

first mode of vibration transverse to the wind direc-tion. Poles that may not have the attachments in-stalled immediately shall be designed for this

worst-case condition. Because the natural fre-quency of a structure without an attached mass istypically higher than those with an attachment, the

resulting critical wind speed and vortex sheddingpressure range are will also be higher for this situa-tion.

11.7.3 Natural Wind Gust

Cantilevered and non-cantilevered overhead

sign, overhead traffic signal, and high-level light-ing supports shall be designed to resist an equiv-alent static natural wind gust pressure range of:

W d FP 250C I (Pa) Eq. 115

W d FP 5.2C I (psf)

where Cd is the appropriate drag coefficientbased on the yearly mean wind velocity of 5 m/s(11.2 mph) specified in Section 3, Loads, for the

considered element to which the pressure rangeis to be applied. If Eq. C11-5 is used in place ofEq. 11-5, C

d

may be based on the location spe-

cific yearly mean wind velocity Vme a n . The naturalwind gust pressure range shall be applied in the

horizontal direction to the exposed area of allsupport structure members, signs, traffic signals,and/or miscellaneous attachments. Designs for

natural wind gusts shall consider the applicationof wind gusts for any direction of wind.

The design natural wind gust pressure rangeis based on a yearly mean wind speed of 5 m/s

(11.2 mph). For locations with more detailed windrecords, particularly sites with higher windspeeds, the natural wind gust pressure may be

modified at the discretion of the owner.

Because of the inherent variability in the ve-

locity and direction of air flow, natural wind gustsare the most basic wind phenomena that may in-duce vibrations in wind-loaded structures. The

equivalent static natural wind gust pressure rangespecified for design was developed with data ob-tained from an analytical study of the response of

cantilevered support structures subject to randomgust loads (Kaczinski et al. 1998).

BecauseVcVme a n is relatively low, the largest val-ues ofCd forthe support componentsmay be conser-

vatively used.

This parametric study was based on the 0.01percent exceedance for a yearly mean wind veloc-ity of 5 m/s (11.2 mph), which is a reasonable up-

per-bound of yearly mean wind velocities for mostlocations in the country. There are locations, how-ever, where the yearly mean wind velocity is larger

than 5 m/s (11.2 mph). For installation sites withmore detailed information regarding yearly meanwind speeds (particularly sites with higher wind

speeds), the following equivalent static naturalwind gust pressure range shall be used for design:

F

m e a n

dNW I

s/m

VCP

2

5250

(Pa) Eq. C 115

7/27/2019 07 Ballot Items 41-60

19/107



15

F

mean

dNW I

m p h.

VC.P

2

21125

(psf)

The largest natural wind gust loading for an

arm or pole with a single arm is from a wind gustdirection perpendicular to the arm. For a pole withmultiple arms, such as two perpendicular arms, the

critical direction for the natural wind gust is willusually not be normal to either arm. The designnatural wind gust pressure range shall be applied

to the exposed surface areas seen in an elevationview orientated perpendicular to the assumed wind

gust direction.

11.7.4 Truck-Induced Gust

Cantilevered and non-cantilevered Oover-head sign and traffic signal support structuresshall be designed to resist an equivalent static

truck gust pressure range of:

TG d F P = 9 00C I (Pa) Eq. 116

TG d F P = 1 8. 8C I (psf)

where Cd is the appropriate drag coefficient

based on the truck speed of 30 m/s (65 mph)from Section 3, Loads, for the considered ele-

ment to which the pressure range is to be ap-plied. If Eq. C11-6 is used in place of Eq. 11-6,Cdshould be based on the considered truck speed

VT. The pressure range shall be applied in thevertical direction to the cantilevered horizontalsupport as well as the area of all signs, attach-

ments, walkways, and/or lighting fixtures pro-jected on a horizontal plane. T his pressure range

shall be applied along any 3.7 m (12 ft) length tocreate the maximum stress range, excluding anyportion of the structure not located directly above

a traffic lane.The equivalent static truck pressurerange may be reduced for locations where vehiclespeeds are less than 30 m/s (65 mph).

The magnitude of applied pressure rangemay be varied depending on the height of thehorizontal support and the attachments above the

traffic lane. Full pressure shall be applied forheights up to and including 6 m (20 19.7 ft), andthen the pressure may be linearly reduced for

heights above 6 m (20 19.7 ft) to a value of zeroat 10 m (33 32.8 ft).

The truck-induced gust loading may be ex-cluded for the fatigue design of overhead cantile-

vered traffic signal support structures, as allowed

The passage of trucks beneath cantileveredsupport structures may induce gust loads on theattachments mounted to the horizontal support of

these structures. Although loads are applied inboth the horizontal and vertical direction, horizontal

support vibrations caused by forces in the verticaldirection are most critical. Therefore, truck gust

pressures are applied only to the exposed horizon-tal surface of the attachment and horizontal sup-port.

A pole with multiple horizontal cantilever armsmay be designed for truck gust loads applied sepa-rately to each individual arm and need not consider

truck gust loads applied simultaneously to multiplearms.

Recent vibration problems on sign structureswith large projected areas in the horizontal plane,such as variable message signs (VMS) enclosures,

have focused attention on vertical gust pressurescreated by the passage of trucks beneath the sign.

The design pressure calculated from Eq. 11-6is based on a truck speed of 30 m/s (65 mph). For

structures installed at locations where the postedspeed limit is much less than 30 m/s (65 mph), thedesign pressure may be recalculated based on this

lower truck speed. The following equation may beused:

FT

dTG I

sm

VCP

2

/30900

(Pa) Eq. C 116

F

T

dTG I

mph

VCP

2

658.18

(psf)

7/27/2019 07 Ballot Items 41-60

20/107



16

by the owner.whereVTis the truck speed in m/s (mph).t and to any attachments located within this length.

A drag coefficient value of 1.20 was usedby DeSantis and Haig (1996) to determine an

equivalent static truck gust pressure range onVMS.

The given truck-induced gust loading may beexcluded for the fatigue design of overhead canti-levered traffic signal structures, as allowed by the

owner. Many traffic signal structures are installedon roadways with negligible truck traffic. In addi-tion, the typical response of cantilevered traffic sig-

nal structures from truck-induced gusts can be sig-nificantly overestimated by the design pressures

prescribed in this article (NCHRP Report 469).However, some cantilever traffic signal structureshave experienced large-amplitude vibrations from

truck-induced gusts applied under the right specificconditions.

7/27/2019 07 Ballot Items 41-60

21/107



17

11.8 DEFLECTION

Galloping and truckgust induced verticaldeflections of cantilevered single-arm sign sup-

ports and traffic signal arms and non-cantileveredsupports should not be excessive so as to resultin a serviceability problem, because motorists

cannot clearly see the arms attachments or are

concerned about passing under the structures.

11.9 FATIGUE RESISTANCE

The allowable constant amplitude fatigue lim-

its (CAFL), are provided in Table 11-3. A sum-mary of the typical fatigue sensitive connectiondetails are presented in Table 11-2 and illustrated

in Figure 11-1. Wind loads of Article 11.7 shall beconsidered in computing the fatigue stress range.

Unless noted in Table 11-2, the membercross section adjacent to the weld toe shall beused to compute the nominal stress range.

Because of the low levels of stiffness anddamping inherent in cantilevered single mast arm

sign and traffic signal support structures, evenstructures that are adequately designed to resistfatigue damage may experience excessive verti-

cal deflections at the free end of the horizontal

mast arm. The primary objective of this provisionis to minimize the number of motorist complaints.

NCHRP Report 412 recommended that thetotal deflection at the free end of single-arm sign

supports and all traffic signal arms be limited to200 mm (8 in) vertically, when the equivalent stat-ic design wind effect from galloping and truck-

induced gusts are applied to the structure.NCHRP Report 494 recommends applying the

200 mm (8 in) vertical limit to non-cantileveredsupport structures. Double-member or truss-typecantilevered horizontal sign supports were not

required to have vertical deflections checked be-cause of their inherent stiffness. There are wereno provisions for a displacement limitation in the

horizontal direction.

The CAFL were established based on fatigue

testing and the resistances were computed basedon elastic section analysis, i.e., nominal values inthe cross section. Therefore, it is assumed that

these resistances include effects of residualstresses due to fabrication, out-of-plane distor-

tions, etc. At this time, only stress range due towind is used; therefore, dead load effects may beneglected.

Residual stresses and anchor bolt pretension aregenerally not considered in the computations.

7/27/2019 07 Ballot Items 41-60

22/107


18

Table 112. Fatigue Details of Cantilevered and Noncantilevered Support Structures

Construction Detail Stress

Category

Application Example

Plain Members 1. With rolled or cleaned surfaces. Flame-cut

edges with ANSI/AASHTO/AWS D5.1 (Ar-ticle 3.2.2) smoothness of 1,000 micro-in.or less.

A - -

2. Slip-joint splice where Lis greater than or

equal to 1.5 diameters.

B High-level lighting

poles.

1

3. Net section of fully-tightened, high-strength

(ASTM A325, A490) bolted connections.

B Bolted joints. 2Mechanically

FastenedConnections 4. Net section of other mechanically fastened

connections:Steel:Aluminum:

D

E

- 3

5. Anchor bolts or other fasteners in tension;

stress range based on the tensile stressarea. Misalignments of less than 1:40 with

firm contact existing between anchor boltnuts, washers, and base plate.

D Anchor bolts.

Bolted mast-arm-to-column connections.

8,16

6. Connection of members or attachment of

miscellaneous signs, traffic signals, etc.with clamps or U-bolts.

D - -

Holes and Cutouts 7. Net section of holes and cutouts. D Wire outlet holes.

Drainage holes.Unreinforced han d-

holes.

5

8. Tubes with continuous full- or partial-

penetration groove welds parallel to the di-rection of the applied stress.

B ' Longitudinal seam

welds.

6

9. Full-penetration groove-welded splices withwelds ground to provide a smooth transi-tion between members (with or without

backing ring removed).

D Column or mast armbutt-splices.

4

10. Full-penetration groove-welded splices withweld reinforcement not removed (with or

without backing ring removed).

E Column or mast armbutt-splices.

4

11. Full-penetration groove-welded tube-to-transverse plate connections with the back-ing ring attached to the plate with a full-

penetration weld, or with a continuous filletweld around interior face of backing ring.

The thickness of the backing ring shall notexceed 10 mm (0.375 in) when a with filletweld attachment to plate is used. Full pe-

netration groove-welded tube-to-transverse

plate connections welded from both sideswith backgouging (without backing ring).

E Column-to-base-plateconnections.

Mast-arm-to-flange-

plate connections.

5

Groove Welded

Connections

12. Full-penetration groove-welded tube-to-transverse plate connections with the back-

ing ring not attached to the plate with acontinuous full-penetration weld, or with acontinuous interior fillet weld.

E' Column-to-base-plateconnections.

Mast-arm-to-flange-plate connections.

5

7/27/2019 07 Ballot Items 41-60

23/107


19

Table 112. Fatigue Details of Cantileveredand Noncantilevered Support Structures (continued)

Construction Detail StressCategory

Ap plicat ion Exa mple

13. Fillet-welded lap splices. E Column or mast arm lap

splices.

3

14. Members with axial and bending loadswith fillet-welded end connections with-

out notches perpendicular to the ap-plied stress. Welds distributed around

the axis of the member so as to bal-ance welds stresses.

E Angle- to-gusset connec-tions with welds termi-

nated short of plateedge.

Slotted tube-to-gussetconnections with copedholes (see note e).

2, 6

Fillet-Welded

Connections

15. Members with axial and bending loads

with fillet-welded end connections withnotches perpendicular to the applied

stress. Welds distributed around theaxis of the member so as to balanceweld stresses.

E' Angle- to-gusset connec-

tions.Slotted tube-to-gusset

connections withoutcoped holes.

2, 6

16. Fillet-welded tube-to-transverse plate

connections (see note j).

E' Column-to-base -plate or

mast-arm-to-flange-plate socket connec-

tions.

7, 8,16

17. Fillet-welded connections with one-sided welds normal to the direction ofthe applied stress.

E' Built-up box mast-arm-to-column connections.

8,16

18. Fillet-welded mast-a rm-to-column pass -

through connections.

E'

(See note f)

Mast-arm-to-column pass-

through connections.

9

19. Fillet-welded T-, Y-, and K-tube-to-tube,

angle-to-tube, or plate-to-tube connec-tions.

(See

notesa and b)

Chord-to-vertical or

chord-to-diagonaltruss connections(see note a).

Mast -arm directly weldedto column (see note b).

Built-up box connection(see note b).

8, 10, 11

25. Fillet-welded ring-stiffened box-to-tube connection.

See note g) Ring-stiffened built-upbox connections.

16

20. Longitudinal attachments with partial-or full-penetration groove welds, or fil-let welds, in which the main member

is subjected to longitudinal loading:

):

) ):

):

L 5 1 m m ( 2 i n

51 m m (2 i n L 12t 102 m m (4 in

L 12t o r 102 m m (4 i n) t 25 m m (1 i n

and

when

C

D

E

Reinforcement at han d-holes.

13Attachments

21. Longitudinal attachments with partial-or full-penetration groove welds, or fil-

let welds in which the main member issubjected to longitudinal loading.

E' Weld terminations atends of longitudinal

stiffeners (see notes hand i).

12, 14

22. Detail 22 has been intentionally re-

moved.

Formatted:Bullets and Num






7/27/2019 07 Ballot Items 41-60

24/107


20


Construction Detail StressCategory

Ap pli cation E xample

Attachments

(continued)

23. Transverse load-bearing fillet-welded

attachments where t m m13 (0.5in )and the main member is subjected

to minimal axial and/or flexural loads.[When t m m13 (0.5 in), see note

d].

C Longitudinal stiffeners

welded to base plates.

12, 14

24. Transverse load-bearing longitudinalattachments with partial- or full-

penetration groove welds or filletwelds, in which the nontubular mainmember is subjected to longitudinal

loading and the weld termination e m-bodies a transition radius that is

ground smooth:

:

:

R 5 1 m m ( 2 i n )

R 5 1 m m ( 2 i n )

D

E

(See note c)

Gusset-plate-to-chordattachments.

15

Notes:

a) Stress category ET with respect to stress in branching member provided that r

24

t

for the chord member. When

r4

t

> , then the fatigue strength equals:

0.7

ET

n n

24F F

r

t

x , where ET

F is the CAFL for category ET.

Stress category E with respect to stress in chord.

b) Stress category ET with respect to stress in branching member.

Stress category K2 with respect to stress in main member (column) provided that r

4tc

for the main member.

Whenc

r24

t> , then the fatigue strength equals:

2

.7

K

n n

c

24F F

r

t

x , where 2F is the CAFL for category

K2.

The nominal stress range in the main member equals: S m ain m em ber = R

S branching m em ber

b

c

t

t

Where tb is the wall thickness of the branching member,tcis the wall thickness of the main member (column), and is the ovalizing parameter for the main member equal to 0.67 for in-plane bending, and equal to 1.5 for out-of-plane bending in the main member. RS b r a n c h i n g m e m ber is the calculated nominal stress range in the branchingmember induced by fatigue design loads. (See commentary of Article 11.5.)

The main member shall also be designed for stress category E using the elastic section modulusof the main mem-

ber and moment just below the connection of the branching member.

c) First check with respect to the longitudinal stress range in the main member per the requirements for longitudinalattachments. The attachment must then be separately checked with respect to the transverse stress range in the

attachment per the requirements for transverse load-bearing longitudinal attachments.

Formatted:Bullets and Numbering

Formatted:Bullets and Numbering

7/27/2019 07 Ballot Items 41-60

25/107


21


Notes (continued):

d) When t > 13 mm ( 0.5 i n), the fatigue strength shall be the lesser of category C or the following:

x

16

c p

n

p

H0 .0 9 4 + 1 .2 3

tF F

t

MPa x

cp

1n

6p

H0 . 0 5 5 + 0 . 7 2

tF F

t

ksi

where c

F is the CAFL for category C, H is the effective weld throat (mm, in), andtpis the attachment plate thick-

ness (mm, in).

e) The diameter of coped holes shall be the greater of 25 mm (1 in), twice the gusset plate thickness, or twice thetube thickness.

f) In addition to checking the branching member (mast arm), the main member (column) shall be designed for stress

category E using the elastic section modulusof the main member and moment just below the connection of thebranching member (mast arm).

g) Stress category E with respect to stress in branching member (ring-stiffened built-up box connection). The main

member shall be designed for stress category E using the elastic section modulus of the main member and mo-ment just below the connection of the branching member.

h) Only longitudinal stiffeners with lengths greater than 102 mm (4 in) are applicable for Detail 21. On column-to-

base-plate or mast-arm-to-flange plate socket connections having a wall thickness greater than 6 mm () which

have exhibited satisfactory field performance, the use of stiffeners having a transition radius or taper with the weldtermination ground smooth may be designed at a higher stress category with the approval of the owner. Under

this exception, the owner shall establish the stress category to which the detail shall be designed. See commen-tary for Article 11.5.

i) Nondestructive weld inspection should be used in the vicinity of the weld termination of longitudinal stiffeners.

Grinding of weld terminations to a smooth transition with the tube face is not allowed in areas with fillet welds orpartial-penetration welds connecting the stiffener to the tube. Full-penetration welds shall be used in areas where

grinding may occur. See commentary for Article 11.5.

j) Fillet welds for socket connections (Detail 16) shall be unequal leg welds, with the long leg of the fillet weld along

the column or mast arm. The termination of the longer weld leg should contact the shafts surface at approxi-mately a 30-degree angle.

7/27/2019 07 Ballot Items 41-60

26/107


22

This page intentionally left blank.

7/27/2019 07 Ballot Items 41-60

27/107


23

Table 113. Constant-Amplitude Fatigue Limits

DetailCategory

SteelLimit

AluminumLimit

MPa ksi MPa ksi

A 165 24 70 10.2

B 110 16 41 6.0

B' 83 12 32 4.6

C 69 10 28 4.0

D 48 7 17 2.5

E 31 4.5 13 1.9

E' 18 2.6 7 1.0

ET 8 1.2 3 0.44

K2 7 1.0 2.7 0.38

7/27/2019 07 Ballot Items 41-60

28/107


24

Figure 111 (a). Illustrative Examples

7/27/2019 07 Ballot Items 41-60

29/107


25

Figure 111 (b). Illustrative Examples

7/27/2019 07 Ballot Items 41-60

30/107


26

Figure 111 (c). Illustrative Examples

7/27/2019 07 Ballot Items 41-60

31/107


27

Figure 111 (d). Illustrative Examples

7/27/2019 07 Ballot Items 41-60

32/107


28

Figure 111 (e). Illustrative Examples

7/27/2019 07 Ballot Items 41-60

33/107


29

Figure 111 (f). Illustrative Examples

7/27/2019 07 Ballot Items 41-60

34/107


30

11.910 REFERENCES

American Association of State Highway and Transportation Officials. AASHTO Standard Specifications for

Highway Bridges. Seventeenth Edition. Washington, D.C.: AASHTO, 1996 2002.

Amir Gilani and Andrew Whittaker (2000). Fatigue-Life Evaluation of Steel Post Structures II: Experimenta-tion. Journal of Structural Engineering, American Society of Civil Engineers , Vol. 126, Issue 3 Vol.2, March 2000, New York, NY. pp. 331-340.

Cook, R. A., D. Bloomquist, A. M. Agosta, and K. F. Taylor.Wind Load Data for Variable Message Signs .

Report no. FL/DOT/RMC/07289488. City, Fla.: University of Florida, Florida Department of Trans-portation, 1996.

Creamer, B. M., K. G. Frank, and R. E. Klingner.Fatigue Loading of Cantilever Sign Structures from Truck

Wind Gusts. Report no. FHWA/TX-79/10+2091F. Austin, Texas: Center for Highway Research,Texas State Department of Highways and Public Transportation, 1979.

DeSantis, P. V. and P. Haig. "Unanticipated Loading Causes Highway Sign Failure." Proceedings ofANSYS Convention, 1996.

Dexter, R. J. and K. W. Johns. Fatigue-Related Wind Loads on Highway Support Structures. AdvancedTechnology for Large Structural Systems, report no. 98 03. Bethlehem, Pa.: Lehigh University,

1998.

Dexter, R., and Ricker, M. Fatigue-Resistant Design of Cantilever Signal, Sign, and Light Supports . Na-

tional Cooperative Highway Research Program (NCHRP) Report 469, Transportation Research Board,

Washington D.C., 2002.

Fisher, J. W., A. Nussbaumer, P. B. Keating, and B. T. Yen. NCHRP Report 354: Resistance of WeldedDetails Under Variable Amplitude Long-Life Fatigue Loading. TRB, National Research Council,Washington, D.C., 1993.

Fouad, F. et al. Structural Supports for Highway Signs, Luminaries, and Traffic Signals. National Coopera-tive Highway Research Program (NCHRP) Report 494, Transportation Research Board, Washington D.C.,

2003.

Kaczinski, M. R.; R. J. Dexter, and J. P. Van Dien. NCHRP Report 412: Fatigue Resistant Design of Canti-

levered Signal, Sign and Light Supports. TRB, National Research Council, Washington, D.C.,1998.

Koenigs, M. T., T. A. Botros, D. Freytag, K. H. Frank,Fatigue Strength of Signal Mast Arm Connections.Report No. FHWA/TX-04/4178-2. Austin, Texas: Center for Transportation Research, Texas De-

partment of Transportation, 2003.

McDonald, J. R.; et al. Wind Load Effects on Signals, Luminaires and Traffic Signal Structures. Report no.

1303-1F. Lubbock, Texas: Wind Engineering Research Center, Texas Tech University, 1995.

NCHRP Project 10-38(2). TRB, National Research Council, Washington, D.C.

7/27/2019 07 Ballot Items 41-60

35/107

2007 AASHTO BRIDGE COMMITTEE AGENDA ITEM: 42 (REVISION 1)

SUBJECT: LRFD Bridge Design Specifications: Section 4, Articles 4.6.2.6.1 and 4.6.2.6.5

TECHNICAL COMMITTEE: T-14 Steel



LRFR MANUAL OTHER



AGENDA ITEM:

Item # 1

Revise Article 4.6.2.6.1 as follows:

Unless specified otherwise in this article or in Articles 4.6.2.6.2, 4.6.2.6.3 or 4.6.2.6.5, the effective flangewidth of a concrete deck slab in composite or monolithic construction may be taken as the tributary width

perpendicular to the axis of the member for determining cross- section stiffnesses for analysis and for determiningflexural resistances. The effective flange width of orthotropic steel decks shall be as specified in Article 4.6.2.6.4.

For the calculation of live load deflections, where required, the provisions of Article 2.5.2.6.2 shall apply.Where a structurally continuous concrete barrier is present and is included in the structural analysis as

permitted in Article 4.5.1, the deck slab overhang width used for the analysis as well as for checking the compositegirder resistance may be extended by:

st

bA

w

2

(4.6.2.6.1-1)

where:

A b = cross-sectional area of the barrier (in.2)

ts = thickness of deck slab (in.)

The slab effective flange width in composite girder and/or stringer systems or in the chords of composite deck

trusses may be taken as one-half the distance to the adjacent stringer or girder on each side of the component, orone-half the distance to the adjacent stringer or girder plus the full overhang width. Otherwise, the slab effective

flange width should be determined by a refined analysis when:

The composite or monolithic member cross-section is subjected to significant combined axial force and

bending, with the exception that forces induced by restraint of thermal expansion may be determined in beam -slab systems using the slab tributary width,

The largest skew angle in the bridge system is greater than 75o , where is the angle of a bearing line

measured relative to a normal to the centerline of a longitudinal component,

The slab spans longitudinally between transverse floorbeams, or

The slab is designed for two-way action.

7/27/2019 07 Ballot Items 41-60

36/107

In the absence of a more refined analysis and/or unless otherwise specified, limits of the width of a concretedeck slab, taken as effective in composite action for determining resistance for all limit states, shall be as specified

herein. The calculation of deflections should be based on the full flange width. For the calculation of live loaddeflections, where required, the provisions of Article 2.5.2.6.2 shall apply.

The effective span length used in calculating effective flange width may be taken as the actual span for simplysupported spans and the distance between points of permanent load inflection for continuous spans, as appropriate

for either positive or negative moments.For interior beams, the effective flange width may be taken as the least of:

One-quarter of the effective span length;

12.0 times the average depth of the slab, plus the greater of web thickness or one-half the width of the topflange of the girder; or

The average spacing of adjacent beams.For exterior beams, the effective flange width may be taken as one-half the effective width of the adjacent

interior beam, plus the least of:

One-eighth of the effective span length;

6.0 times the average depth of the slab, plus the greater of one-half the web thickness or one-quarter of the

width of the top flange of the basic girder; or

The width of the overhang.

Item # 2

Revise Article C4.6.2.6.1 as follows:

Longitudinal stresses are distributed across the deck of composite and monolithic flexural members by in-planeshear stresses. Due to the corresponding shear deformations, plane sections do not remain plane and the

longitudinal stresses across the deck are not uniform. This phenomenon is referred to as shear lag. The effectiveflange width is the width of the deck over which the assumed uniformly distributed longitudinal stresses result

approximately in the same deck force and member moments calculated from elementary beam theory assumingplane sections remain plane, as are produced by the nonuniform stress distribution.

The provisions of Article 4.6.2.6.1 apply to all longitudinal flexural members composite or monolithic with adeck slab, including girders and stringers. They are based on finite element studies of various bridge types and

configurations, corroborated by experimental tests, and sensitivity analysis of various candidate regressionequations (Chen et al. 2005). Chen et al. (2005) found that bridges with largerL/S(ratio of span length to girder

spacing) consistently exhibited an effective widthbe equal to the tributary width b. Nonskewed bridges with L/S=3.1, the smallest value ofL/Sconsidered in the Chen et al. (2005) study, exhibitedbe = bin the maximum positive

bending regions and approximately be = 0.9b in the maximum negative bending regions under service limit stateconditions. However, they exhibited be = b in these regions in all cases at the strength limit state. Bridges with

large skew angles often exhibited be< b in both the maximum positive and negative moment regions, particularlyin cases with small L /S. However, when various potential provisions were assessed using the Rating Factor (RF) as

a measure of impact, the influence of using full width ( be = b) was found to be minimal. Therefore, the use of thetributary width is justified in all cases within the limits specified in this article. The Chen et al. ( 2005) study

demonstrated that there is no significant relationship between the slab effective width and the slab thickness, asimplied by previous Specifications.

These provisions are considered applicable for skew angles less than or equal to 75o,L/Sgreater than or equal

to 2.0 and overhang widths less than or equal to 0.5 Sbased on the Chen et al. (2005) study and complementary

studies by Nassif et al. (2006). In unusual cases where these limits are violated, a refined analysis should be used to

determine the slab effective width. Furthermore, these provisions are considered applicable for slab-beam bridgeswith unequal skew angles of the bearing lines, splayed girders, horizontally curved girders, cantilever spans andvarious unequal span lengths of continuous spans although these parameters have not been investigated extensively

in studies to date. These recommendations are based on the fact that the participation of the slab in these broaderparametric cases is fundamentally similar to the participation of the slab in the specific parametric cases that have

been studied.The use of one-half the distance to the adjacent stringer or girder in calculating the effective width of the main

girders in composite girder and/or stringer systems or the truss chords in composite deck trusses is a conservativeassumption for the main structural components, since typically a larger width of the slab can be expected to

7/27/2019 07 Ballot Items 41-60

37/107

participate with the main girders or truss chords. However, this tributary width assumption may lead to anunderestimation of the shear connector requirements and a lack of consideration of axial forces and bending

moments in the composite stringers or girders due to the global effects. To utilize a larger slab width for the maingirders or truss chords, a refined analysis should be considered.

The specific cases in which a refined analysis is recommended are so listed because they are significantlybeyond the conventional application of the concept of a slab effective width. These cases include tied arches where

the deck slab is designed to contribute to the resistance of the tie girders and cable stayed bridges with a compositedeck slab. Chen et al. ( 2005) provide guidance for selection of a few case study results for simplified lower-bound

slab effective widths in composite deck systems of cable stayed bridges with certain specific characteristics.Longitudinal stresses in the flanges are spread across the flange and the composite deck slab by in-plane shear

stresses. Therefore, the longitudinal stresses are not uniform. The effective flange width is a reduced width overwhich the longitudinal stresses are assumed to be uniformly distributed and yet result in the same force as the

nonuniform stress distribution would integrated over the whole width.In calculating the effective flange width for closed steel and precast concrete boxes, the distance between the

outside of webs at their tops will be used in lieu of the web thickness, and the spacing will be taken as the spacingbetween the centerlines of boxes.

For open boxes, the effective flange width over each web should be determined as though each web was anindividual supporting element.

For filled grid, partially filled grid, and for unfilled grid composite with reinforced concrete slab, the slabdepth used should be the full depth of grid and concrete slab, minus a sacrificial depth for grinding, grooving and

wear (typically 0.5 in.).Where a structurally continuous concrete barrier is present and is included in the models used for structural

analysis as permitted in Article 4.5.1, the width of overhang for the purpose of this article may be extended by:

st

bAw

2 (C4.6.2.6.1-23)

where:

A b = cross-sectional area ofthe barrier (in.2)

ts = depth of deck slab (in.)

Item #3

Add the following references to the Section 4 reference list:

Chen, S.S., A.J. Aref, I.-S. Ahn, M. Chiewanichakorn, J.A. Carpenter, A. Nottis, and I. Kalpakidis. 2005.

Effective Slab Width for Composite Steel Bridge Members. NCHRP Report 543, Transportation Research Board,Washington, D.C., 153 pp.

Nassif, H., Talat, A.-A., and El-Tawil, S. 2006. Effective Flange Width Criteria for Composite Steel Girder

Bridges, Annual Meeting CD-ROM, Transportation Research Board, Washington, D.C., 29 pp.

Item # 4

Add the following article after Article 4.6.2.6.4:

4.6.2.6.5 Transverse Floorbeams and Integral Bent Caps

For transverse floorbeams and for integral bent caps designed with a composite concrete deck slab, the

effective flange width overhanging each side of the transverse floorbeam or bent cap web shall not exceed six timesthe least slab thickness or 1/10 of the span length. For cantilevered transverse floorbeams or integral bent caps, t hespan length shall be taken as two times the length of the cantilever span.

7/27/2019 07 Ballot Items 41-60

38/107

Item # 5

Add the following article after Article C4.6.2.6.4:

C4.6.2.6.5

The provisions for the effective flange width for transverse floorbeams and integral bent caps are based on past

successful practice, specified by Article 8.10.1.4 of the 2002 AASHTO Standard Specifications.

OTHER AFFECTED ARTICLES:

None

BACKGROUND:

In composite girders, the shear lag phenomenon can potentially result in underestimation of the deflections andflexural stresses in calculations based on line-girder analysis and the elementary theory of bending, which assume

that plane cross sections remain plane. It is traditional to obtain estimates of maximum deflection or stress fromelementary theory by utilizing an effective slab width concept in which the actual width of each flange is replaced

by an appropriate effective width (Garcia and Daniels 1971, Moffatt and Dowling 1978, ASCE 1979, Ahn et al.2004 ). The slab effective width directly affects the computed moments, shears, torques, and deflections for the

composite section and also affects the proportions of the cross-section and the number of shear connectors that arerequired. The effective slab width is thought to be particularly important for serviceability checks (e.g., fatigue and

overload), which can often govern the design.

Chiewanichakorn et al. (2004, 2005) and Aref et al. (2007) explained the need for and the prescription of a new

definition for effective width that accounts for the variation of bending stresses through the deck thickness. A finiteelement modeling approach was developed, corroborated with experimental data, and applied to a suite of bridgesdesigned according to industry guidelines. Effective widths according to the new definition were extracted from

this finite element parametric study. Principal findings from the parametric study were ( Chen et al. 2005a, 2007):

(i) Full width was typically acting at cross sections where it is most needed, i.e., where moments and henceperformance ratios would be highest, and

(ii) Where the effective width was less than full width, the corresponding cross sections had considerable

excess flexural capacity.

The proposed revisions for effective width criteria are developed based on regression analyses (NCHRP 2003),accounting for different subsets of parameters varied in an extensive parametric study of bridge finite element

models. Impacts of various potential provisions were assessed, using the Rating Factor (RF) as the measure ofimpact. Based on the impact assessment, the proposed provisions utilizing the full tributary slab width arerecommended for both positive and negative moment regions, unless specified otherwise for cases significantly

outside the scope of the parametric study.

Item # 1

The Chen et al. (2005a) study has resulted in the recommendation that the full width tributary to each girder or webmay be used for the effective deck slab width in monolithic and composite bridge members for most situations of

practical interest. This recommendation was determined to be suitable for service as well as the strength limit

states, for exterior as well as interior girders, for negative as well as positive moment regions, and for skewed aswell as right alignments.

The simplicity of this recommendation results from an extensive set of analyses on a variety of bridgeconfigurations. An extensive impact analysis based on NCHRP ( 2003) principles revealed that more cumbersome

curve-fit expressions for effective width, although more accurate, were not significantly so in terms of thegoverning rating factor (RF) of the bridges investigated.

The (Chen et al. 2005a) parametric study finite element models were validated against a suite of experimental tests.

7/27/2019 07 Ballot Items 41-60

39/107

The parametric study bridges were selected by design of experiment (DOE) concepts and included:

Simple-span cases as well as three-span continuous cases with span length ratios of 1.0 and 1.5,

L/S ranging from 3.1 to 25, where S is the girder spacing and L is the span length for a simple span, or the

shortest span length for continuous spans,

Equal skew angles of the bearing lines ranging from zero to 60 degrees, and

S/tsranging up to 20, where tsis the slab thickness.

All the bridges were straight and had solid deck thicknesses that met or exceeded the minimum depth of 7.0 in.

specified in Article 9.7.1.1. Also, a number of additional bridges were analyzed to evaluate the influence of theproposed provisions. These cases included:

Two-span continuous steel I-girder bridges with two girders in the cross-section having zero skew, a smallL/S

= 3.18 and large S/tsvalues of 27.4 and 32.0, with transverse prestressing in the slab for the case with the largerS/ts,

Simple-span and two-span continuous steel tub-girder bridges withL/S= 16.7 and 12.5, zero skew and S/ts =

18,

Simple-span and two-span continuous concrete bulb-tee girder bridges withL/S= 5.6, zero skew and S/ts= 18.

Eq. 1 and the corresponding text are moved from the commentary to the specification provisions. This groups therules for determining the effective concrete width in the Specification provisions and shortens the commentarydiscussion.

Item # 2

In the first paragraph, the first sentence is revised to make it more specific. Also, the phrase a reduced width isreplaced by the width, which refers to the reduced or unreduced width. The term and member moments is

added to reflect the fact that the Chen et al. ( 2005a) study determined the flange effective width based on theconsideration of both force and moment equilibrium.

The second through the fifth paragraphs of the revised Article C4.6.2.6.1 provide guidance on the basis for andusage of the new provisions of Article 4.6.2.6.1. Paragraph 6 discusses cases that are strictly beyond the scope ofArticle 4.6.2.6.1.

The second through the fifth paragraphs of the current Article C4.6.2.6.1 are removed since they are no longer

applicable in the context of the new provisions.

A slightly modified version of the sixth paragraph of the current Article C4.6.2.6.1 is moved to the Specificationprovisions.

ANTICIPATED EFFECT ON BRIDGES:

The proposed provisions simplify the computation of the effective flange width for the concrete deck and result inmore efficient bridge designs. As demonstrated in the resea rch studies and example design calculations (Chen et

al. 2005a, Chen et al. 2005b), the proposed revisions provide a better representation of the structural performance

and, for most of the limit-state calculations, lead to improved design economy in terms of materials for bridges withwider girder spacings. Note that for such bridges, a slightly smaller shear connector spacing may be required.Also, particularly if the concrete is taken to be fully effective in negative bending in the application of Eq.

(6.10.4.2.2-4), slightly larger web thicknesses may be required in steel-girder bridges. In such cases, the largereffective slab width increases the susceptibility of the web to bend buckling in these regions at the Service II limit

state. Thus, in these instances, it is imperative that the depth of web in compression Dc be calculated consideringthe effect of the noncomposite dead load stress on the location of the neutral axis, as specified in Article D6.3.1,

when the deck is considered to be effective in tension at the service limit state.

7/27/2019 07 Ballot Items 41-60

40/107

It should be noted that a potential downside of the proposed provisions discussed in Chen et al. (2005a ) forchecking of composite sections in negative bending has been eliminated by the unified flexural resistance equations

implemented in Article 6.10 of the AASHTO 3rd

Edition LRFD Specifications. Chen et al. ( 2005a) notes that, forcomposite sections in negative flexure, the increase in the depth of the web in compression Dc due to an increased

slab effective width may change the cross-section classification from compact to noncompact. As such, Chen et al.(2005a) indicate that the cross-section resistance would be reduced from the full plastic moment Mp to a value less

tha

07 ballot items 41-60

Documents