CHAPTER 6

HIGHWAY INFRASTRUCTURE

INTRODUCTION

Highway infrastructure protection historically has been the primary consideration in determining
TS&W limits as the weights and dimensions of trucks in particular determine the costs that
highway agencies must bear to construct and maintain a highway system to serve present traffic and
that anticipated in the near future. This Chapter is intended to acquaint the reader with the
technical and practical side of TS&W interaction with the infrastructure elements. Pavement
deterioration increases with axle weight, the number of axle loadings, and the spacing within axle
groups. The axle loads and spacing on trucks also affects the design and fatigue life of bridges.
Truck dimensions influence roadway design -- truck width affects lane widths, trailer or load
height affects bridge and other overhead clearances, and length affects intersection and curve
design. And conversely, truck designs are determined by existing pavement and bridge strength
and roadway geometry.

Pavement types analyzed in this Study include flexible, asphaltic concrete; and rigid,
portland cement concrete. Bridge features included in the analysis are span length and type
of member support -- simple or continuous. The list of roadway geometry features analyzed
includes interchange ramps, intersections, and mainline curves. Alternative truck configurations
analyzed, in terms of their interaction with highway infrastructure features, include single-unit
or straight trucks and single- and multitrailer truck combinations.

OVERVIEW OF INFRASTRUCTURE IMPACTS

The TS&W characteristics -- axle weights, GVW, truck length, width, and height -- affect
pavements, bridges, and roadway geometry in different ways, as shown in Table VI-1.

VI-1
 Table VI-1
Highway Infrastructure Elements Affected by TS&W Limits

Highway Infrastructure Element Axle GVW Axle Truck Truck Truck

Weight Spacing Length Width Height

Pavement Flexible E E

Rigid E e

Bridge Short-Span E E E
Features
Long-Span E e E

Clearance e E

Roadway Interchange e E e
Geometric Ramps
Features
Intersections E e

Climbing Lanes E

Horizontal e e
Curvature

Vertical Curve E
Length

Intersection E E
Clearance Time

Passing Sight e
Distance

Key: E = Significant Effect

e = Some Effect

IMPACT OF WEIGHT

There are two aspects of truck weight that are interdependent and that interact with the highway
infrastructure -- axle weight (loading) and GVW. As shown in Table VI-1, the effect of axle
weight is more significant to pavements and short-span bridges, whereas GVW is of more
significance to long-span bridges.

Generally, highway pavements are stressed by axle and axle group loads directly in contact with
the pavement rather than by GVW. The GVW, taking into account the number and types of axles
and the spacing between axles, is distributed among the axles and determines axle loads. Over
time, the accumulated strains (the pavement deformation from all the axle loads) deteriorate
pavement condition, eventually resulting in cracking of both rigid and flexible pavements and
permanent deformation or rutting in flexible pavements. If the pavement is not routinely
maintained, the axle loads, in combination with environmental effects, will accelerate the cracking
and deformation. Proper pavement design relative to loading is a significant factor in pavement
life, and varies by highway system and the number of trucks in the traffic stream.

VI-2
Axle groups, such as tandems or tridems, distribute the load along the pavement, allowing greater
weights to be carried and resulting in the same or less pavement distress than that occasioned by a
single axle at a lower weight. The spread between two consecutive axles also affects pavement
life or performance; the greater the spread, the more each axle in a group acts as a single axle. For
example, a spread of 9 to 10 feet results in no apparent interaction of 1-axle with another, and each
axle is considered a separate loading for pavement impact analysis or design purposes.
Conversely, the closer the axles in a group are, the greater the weight they may carry without
increasing pavement deterioration beyond that occasioned by a single axle, dependent on the
number of axles in the group. This benefit to pavements of adding axles to a group decreases
rapidly beyond 4-axles.

Axle loads also have a beneficial effect on short-span bridges -- that is, bridge spans that are
shorter than the truck, thereby resulting in only 1-axle group, front or rear, being on the span at any
time. While spreading the axles in an axle group is beneficial to short-span bridges, it is
detrimental to pavement. It is not GVW but the distribution of the GVW over axles that impacts
pavements.

However, GVW is a factor for the life of long-span bridges -- that is, bridge spans longer than
the wheelbase of the truck. Bridge bending stress is more sensitive to the spread of axles than to
the number of axles. The FBF takes into account both the number of axles and axle spreads in
determining allowable GVW.

In the context of roadway geometrics, increasing GVW affects a truck's ability to accelerate from a
stop, to enter a freeway, or to maintain speed on a long grade. Acceleration from a stop influences
the time required to clear an intersection. Acceleration into a freeway affects the determination of
acceleration lane length requirements. Inability to maintain speed on a long grade requires the
construction of truck climbing lanes. Some of these effects can be ameliorated by changes in truck
design, primarily to engine and drive train components. The GVW also has a second order effect
on offtracking -- that is, on how the rear axle of a trailer tracks relative to the steering axle of the
truck. Other truck characteristics affected by roadway geometrics are discussed in more detail
later in this Chapter.

IMPACT OF DIMENSIONS

The dimensions of trucks and truck combinations have various effects on the three elements of
highway infrastructure. The most significant effects relate to length, particularly when combined
with GVW. Width has a limited effect on swept path -- the combination of offtracking and vehicle
width. Swept path affects highway geometrics in terms of interchange ramp or roadway
intersection design which is based on mapping a maximum swept path that the truck encroaches on
the shoulder, over the curb, or into another lane of traffic. Height regulations are intended to
ensure that trucks will clear overhead bridges, bridge members, overhead wires, traffic signals,
and other obstructions.

In general, truck length -- or more specifically wheelbase -- has a strong effect on bridge stress for
long-span bridges. The longer the wheelbase the shorter the distance from the support member to
where the load is being applied (the moment arm) when the truck is in the middle of

VI-3
the span. The shorter the truck the greater the concentration of load at the middle of the span, and
the longer the distance (moment arm) to the support member for the bridge span member. A truck
at mid-span is the loading condition for the maximum stress in a simple supported span. This is
not the case for some continuous supported spans: when a truck is straddling the center pier of a
continuous span, increasing the truck length can increase the stress in the span at the pier.

The effect of truck wheelbase on offtracking is reduced considerably if the combination is

articulated, especially in a multitrailer combination. Low-speed offtracking affects interchange
and intersection design, and high-speed offtracking affects lane width.

BRIDGES

Bridges are critical to the safe and efficient movement of people and freight on the Nation’s
highways. This section discusses the important considerations that have influenced the decision
making and investments of Federal and State transportation officials for bridges.

BRIDGE DESIGN1

Most highway bridges in the United States were designed according to the design guidelines of the
AASHTO. These guidelines provide traffic-related loadings to be used in the development and
testing of bridge designs, as well as other detailed requirements for bridge design and
construction.

Dynamic effects (vibration resulting in bridge loads that vary above and below that load
resulting trucks operating at higher speeds. In bridge design, design loadings (in the static
condition) are adjusted upward to account for dynamic effects. To minimize the dynamic effects of
extra-heavy nondivisible loads on some bridges, permits often require the truck to cross at a very
slow speed, depending on its GVW.

A key task in bridge design is to select bridge members that are sufficiently sized to support
the various loading combinations the structure may carry during its service life. These include
dead load (the weight of the bridge itself); live load (the weights of vehicles using the bridge); and
wind, seismic, and thermal forces. The relative importance of these loads is directly related to the
type of materials used in construction, anticipated traffic, climate, and environmental conditions.
For a short-span bridge (for example, span length of 40 feet), about 70 percent of the load-bearing
capacity of the main structural members may be required to support the traffic-related live load,
with the remaining 30 percent of capacity supporting the weight of the bridge itself. For a long
bridge (for example, span length of 1,000 feet), as much as 75 percent

1
A substantial amount of the background material is drawn from the TRB Special Report 225, Truck Weight
Limits: Issues and Options, 1990 and from the 1981 U.S. DOT Report to Congress under Section 161, An
Investigation of Truck Size and Weight Limits.

VI-4
of the load-bearing capacity of the main structural members may be required to support the weight
of the bridge.

In most instances, the loading event that governs bridge capacity is a design vehicle placed
at the critical location on the bridge. In certain cases, a lane loading simulating the presence of
multiple trucks on a bridge is the governing factor. Bridges are also affected by the dynamic
impact and lateral distribution of weight of trucks; dynamic impact is determined by speed and
roadway roughness, and the lateral distribution of loads varies with the position of the truck(s)
on the bridge.

The methods used to calculate stresses in bridges caused by a given loading are necessarily
conservative; therefore, the actual measured stresses are generally much less than calculated
stresses. Providing for a margin of safety is necessary to bridge design because:

C The materials used in construction are not always completely consistent in size, shape, and
quality;

C The effects of weather and the environment are not always predictable;

C Highway users on occasion violate vehicle weight laws;

C Legally allowed loads may increase during the design life of a structure; and

C Overweight loading is occasionally allowed by permit.

The adjustment of the nominal legal loading is reflected in the safety factors, which are selected so
that there is only a very small probability that a loading condition that exceeds load capacity will
be reached within the bridge’s design life.

The margins of safety used by bridge designers in the past have been reduced in recent bridge
design procedures. Use of new design procedures and computer-aided engineering and design has
enabled more precise analysis of load effects and the selection of smaller bridge members. Also,
the competition between the steel and concrete industries has led each group to foster lower costs
for their own material. For example, many designs now proposed for steel bridges reduce the
safety factor by reducing the number of girders, which increases their spacing.

Design and construction of highway bridges in the United States has been governed by the
AASHTO's Standard Specifications for Highway Bridges since 1931, with subsequent revisions.
In the early 1990s AASHTO decided to develop an entirely new bridge code to incorporate
state-of-the-art bridge engineering that is based on the load and resistance factor design (LRFD)
approach. 2 In 1993, AASHTO adopted LRFD bridge design specifications on a trial basis, as an

2
FHWA http://www.ota.fhwa.dot.gov/tech/struct/dp99lr.html, February 19, 1998.

VI-5
alternative to standard bridge design specifications. In 1996, interim LRFD specifications were
made available by AASHTO and conversion to this method was encouraged wherever practical.3

The LRFD method applies statistically determined factors to bridge design parameters, using a
series of load and resistance factors to account for variabilities in loads and material resistance.
The specifications use statistical methods and probability theory to define the variations in loading
and material properties and the likelihood that various load combinations will occur
simultaneously. 4

BRIDGE IMPACT

Past studies of the impact of truck weight limit changes on bridges were based on various
percentages of the yield stress for steel girder bridges, such as 55 percent or 75 percent. The yield
stress, a property of the particular type of steel, is the stress at the upper limit of the elastic range
for bridge strain. The elastic range of a structural member is the set of stresses over which the
deformation -- the strain of the member -- is not permanent. In the elastic range, the member
returns to its former size and shape when the stress is removed. There is no permanent set in the
structural member. For this discussion, strain is the elongation of a steel girder when (1) a portion
of the strain becomes permanent at a stress level above the yield stress; and (2) the girder
continues to elongate, or stretch, under increasing load until it ruptures or fails. Beyond the elastic
range, there is permanent elongation of the bridge girder, that is, for those stresses that are greater
than the yield stress. However, in structural steel there is considerable strain before failure
occurs.

BRIDGE INVENTORY AND OPERATING RATINGS

States rate bridges, at their discretion, at either an inventory rating (55 percent of the yield stress)
or operating rating (75 percent of the yield stress).5 Bridges are never intentionally loaded to
yield stress in order to provide an adequate margin of safety. The design stress level for bridges
is the same as the inventory rating, 55 percent of the yield stress. These two ratings are also used
for posting bridges; either may be used under AASHTO guidelines, at the option of the State. A
sign specifying weight limits is posted on bridges when it is determined that a vehicle above the
specified weight would overstress the bridge. This weight could be that which stresses the bridge
at either the 55 percent or 75 percent level of the yield stress.

3
AASHTO http://www2.epix.net/~lrfd/develop.html, February 19, 1998.
4
Ibid.
5
According to the AASHTO Manual for Maintenance Inspection of Highway Bridges (1983) an operating
rating is defined as RF = 0.75-D/L(1+I) where RF= rating factor arrived at with the equation 0.55R= D + L (1 +
I) where R= the limiting stress (often the stress at which steel will undergo permanent deformation, or “yield”),
D= stress due to dead load (the effect of gravity on bridge components), L= stress due to live load (vehicles on
the bridge), I= an adjustment to the static effect of live loads to account for dynamic effects. An inventory
bridge rating is arrived at by selecting the most highly stressed bridge component and inserting the rating factor
(RF) into the Equation, RF= 0.55R- D/L(1 + I), as a multiplier on the live load of the rating truck.

VI-6
As States have the option to use either level for posting purposes, both ratings have been used in
past studies to assess the bridge impacts for evaluating TS&W policy scenarios. Significant cost
differences result from choice of rating. Use of the lower stress level (inventory rating) results in
more bridges being identified as needing to be upgraded to accommodate increased weights or
decreased lengths.6

Following the reviews of the TRB Special Reports 225 and 227 the FHWA determined that
the stress level most representative of all State bridge posting practices was the inventory rating
(55 percent of the yield stress) plus 25 percent, which gives a level of 68.8 percent of yield stress.
The FHWA used this 68.8 percent of yield to estimate the bridge cost impacts of LCVs. The
resulting cost estimate reported by the FHWA in May 1991 was much closer to that based on the
75 percent rating, the TRB findings.

BRIDGE STRESS

Bridge stresses caused by vehicles depend on both GVW and the distances between the axles
that act as point loads. Trucks having equal weight but different wheelbases produce different
bridge stresses. The shorter the wheelbase, the greater the stress. On a simple-span bridge, the
length of a truck relative to the length of bridge span is also important. For relatively short spans
(20 feet to 40 feet), all axles of a truck combination will not be on the bridge at the same time. The
maximum bending moments determine stresses in the main load-carrying members of simple span
bridges.

Figure VI-1 shows the maximum bending moments, by span lengths between 40 and 160 feet, for
two trucks: a 50,000-pound single unit truck with a wheelbase of 19 feet, and an 80,000-pound
combination with a wheelbase of 54 feet. For shorter bridges, the 50,000-pound single unit truck
produces slightly higher stresses than the 80,000-pound combination; however, for longer bridges,
the combination produces higher stresses.

TS&W REGULATION RELATED TO BRIDGE PROTECTION

The TS&W regulation to protect bridges generally takes the form of a bridge formula or table.
Federal bridge protection regulation, which became effective in 1975, uses a formula. Some
States still use bridge tables, which were grandfathered by the 1975 Federal law. Other States use
bridge tables for issuing overweight permits. The FBF is based on overstress criteria, the amount
of bridge stress above the design stress to be allowed.

6
The TRB Special Reports 225, Truck Weight Limits: Issues and Options and 227, New Trucks for Greater
Productivity and Less Road Wear: an Evaluation of the Turner Proposal estimated the bridge costs of the
TS&W changes under study based on the operating rating of 75 percent of yield stress, whereas reviewers of
those reports found much higher bridge costs resulting from the use of the inventory rating of 55 percent of
yield stress.

VI-7
 Figure VI-1
Maximum Bending Moments on a Simple Span Bridge:
50,000-pound Single Unit Truck vs. 80,000-pound Truck Combination

3000
Single Unit
2500 Combination

2000

1500

1000

500

0
40' 60' 80' 100' 120' 140' 160'
Span Length (ft.)

OVERSTRESS CRITERIA AND LEVEL OF RISK

The level of risk to accept in determining acceptable loadings for a given bridge, or acceptable
bridge design requirements for given loadings, is an element of TS&W regulation. A less
conservative bridge formula, one that did not preserve the underlying FBF criteria, would
reduce the margin of safety, thereby increasing somewhat the likelihood of bridge damage
due to overstress. An overstress sufficient to damage a bridge would necessitate bridge repair
and/or replacement sooner than anticipated.

BRIDGE FATIGUE

Another factor to be considered is fatigue life, which is related to repetitive loadings. Each truck
crossing produces one or more stress cycles in bridge components, which use up a portion of the
components' fatigue lives. The magnitude of stress depends on vehicle weight and the size of the
bridge component. The occurrence of a fatigue failure is signaled by cracks developing at points
of high stress concentration.

Generally, only steel bridges are susceptible to fatigue, although some studies suggest that
commonly used prestressed concrete spans, if overloaded, are similarly susceptible. The
governing damage law for steel components has a third-power relationship between stress and
damage, so that a doubling of stress causes an eight-fold increase in damage.7

7
Fisher, 1977.

VI-8
Bridge details that are particularly susceptible to fatigue include weld connections in tension
zones, pin and hanger assemblies, and cover plates on the bottom flanges of steel beams.8 Many
fatigue failures result from stresses induced indirectly by the distortion of the structure due to poor
design details or unforeseen restraints. Most steel cracks reported to date probably fall into the
category of distortion induced. Some of the worst detailing can be corrected by repair and retrofit.

FEDERAL BRIDGE FORMULA

In 1975 along with axle and maximum GVW limits for Interstate highways, Federal law adopted a
bridge formula that restricts the maximum weight allowed on any group of consecutive axles based
on the number of axles in the group and the distance from the first to the last axle. The AASHO
proposed the formula concept in the 1940s. It was further developed and presented in a 1964
Report to Congress from the Secretary of Commerce.9 That Study recommended a table of
maximum weights for axle groups to protect bridges (see Appendix A). The values in the table are
derived from the following formula, that is, FBF:

W = 500 [ L N / ( N - 1 ) + 12 N + 36 ]

where:

W = maximum weight in pounds carried on any group of two or more consecutive

axles

L = distance in feet between the extremes of the axle group

N = number of axles in the axle group

Current Federal law specifies exceptions to the results given by the above formula: 68,000 pounds
may be carried on two sets of tandem axles spaced at least 36 feet apart, and a single set of tandem
axles spread no more than 8 feet is limited to 34,000 pounds.

The FBF is based on assumptions about the amount by which the design loading can be
exceeded for different bridge designs. Specifically, this formula was designed to avoid
overstressing HS-20 bridges by more than 5 percent and H-15 bridges by more than 30 percent.
The FHWA established a bridge stress level of not more than 5 percent over the design stress
for HS-20 bridges to preserve the significantly large investment in these bridges by Federal, State,
and local governments, and because these bridges carry high volumes of truck traffic.

8
AASHTO specifications give different allowable fatigue stresses for different categories of detail. These
fatigue rules were initiated in the mid-1970s, therefore many older bridges were never checked during their
original design for fatigue life. Further, the AASHTO fatigue rules apply to welded and bolted details with
stresses induced directly by load passages (Moses, 1989).
9
Maximum Desirable Dimensions and Weights of Vehicles Operated on the Federal-Aid System, 1964 Study
Report to Congress, U.S. Department of Commerce.

VI-9
Although a level of up to 30 percent is considered a safe level for overstressing an H-15 bridge
in good condition, the fatigue lives of these structures may be shortened by repeated loadings at
this level.

The FBF reflects the fact that increasing the spacing between axles generally results in less
concentrated loadings and lower stresses in bridge members. For example, the bridge formula
would allow a 3-axle single-unit truck with a wheelbase of 20 feet to operate at 51,000 pounds. If
the wheelbase of this truck is increased to 24 feet, the maximum weight allowed under FBF would
increase to 54,000 pounds as shown in Table VI-2.

Table VI-2
FBF 3-axle, 4-axle, And 5-axle Single-unit Truck Limit

Distance10 GVW (Pounds)

(Feet)
3-Axles 4-Axles 5-Axles

20 51,000 55,500 60,500

24 54,000 58,000 63,000
28 57,000 60,500 65,500
32 60,000 63,500 68,000
36 66,000 70,500
40 68,500 73,000

As noted, there is a greater gain in allowable load by adding an axle than by increasing the
distance between axles. For instance, at 30 feet a 3-axle vehicle is allowed a maximum GVW
of 58,500 pounds and by adding 2 feet can gain only 1,500 pounds. If the same 3-axle vehicle at 30
feet adds an axle there is a gain of 3,500 pounds -- or 2,000 pounds more than by increasing
distance by 2 feet. Increasing the number of axles in an axle group without increasing the overall
length of the group has very little effect in reducing bridge stress. However, more axles do
provide substantial benefits to pavements.

POTENTIAL ALTERNATIVES TO FBF

Actually, the FBF is not just one formula but a series of formulas with the appropriate one chosen
by a parameter, N, the number of axles in the group in question. However, bridge stress is affected
more by the total amount of load than by the number of axles. Thus the FBF is not effective in
modeling the actual physical phenomenon, and it results in loads, especially for long combinations,
that overstress bridges more than intended. More importantly, it encourages the addition of axles
to obtain more payload even though one or both bridge stress criteria are exceeded. At other
times, the equation restricts allowable loads for some short trucks below that

10
Between the outside axles of any group of 2 or more axles.

VI-10
allowed by the stress criteria themselves. In summary, the FBF actually results in overstressing
some of the bridges it is intended to protect.

Since 1975, there have been a number of proposals to revise the FBF and reduce its shortcomings.
However, significant areas of concern have been identified with respect to the alternatives as
well. Three alternative formulas proposed in recent years are discussed here: a TRB (a
combination of the Texas Transportation Institute (TTI) and FBFs) alternative, an AASHTO
alternative, and a Goshen alternative.

TRB ALTERNATIVE

In 1990, the TRB recommended adoption of the formula developed by the TTI which would allow
a 5 percent overstress for HS-20 bridges, in conjunction with existing Federal axle limits for
vehicles with GVWs of 80,000 pounds or less.11 The TRB Report further recommended the FBF
continue to be applied to vehicles weighing more than 80,000 pounds. The effect of this proposal
would be an increase in maximum weights allowed for shorter vehicles, while the maximum
weight limits for the longer wheelbase trucks would remain unchanged. It was asserted that the
TTI formula was overly conservative at heavier weights.

The TTI formula is in the form of two equations for straight lines that meet at a wheelbase length of
56 feet. For wheelbases less than 56 feet, it is:

W = 1,000 ( L + 34 )

For wheelbases equal to or greater than 56 feet, it is:

W = 1,000 ( L/2 + 62 )

where:

W = allowable weight

L = wheelbase for truck configuration

11
TRB Special Report 225.

VI-11
AASHTO ALTERNATIVE

In 1993, AASHTO issued a report which recommended that its member committees (1) evaluate
nationwide adoption of the TTI bridge formula as a replacement for FBF; (2) consider a limit
on maximum extreme axle spacing of 73 feet in the short term; (3) retain existing single- and
tandem-axle limits; (4) control tridem-axle weights -- and the special permitting of vehicles with
GVWs more than 80,000 pounds -- using the original TTI bridge formula which protects both
H-15 and HS-20 bridges, as opposed to the TTI formula mentioned above, which protects only
HS-20 bridges. The recommendation was reviewed by the AASHTO Highway Subcommittees on
Bridges and Structures and Highway Transport, accepted in resolution form, and approved by the
Standing Committee on Highways. The AASHTO Board of Directors considered the
recommendations at its 1996 Fall Meeting. The board expressed concern that the impact on
pavements was not adequately addressed and remanded it for further consideration to the
Subcommittees on Design and on Bridges and Structures.

GHOSN ALTERNATIVE

In 1995 a research study by Ghosn and others for FHWA, proposed a new formula based on
structural reliability theory as a replacement for the FBF.12 Structural reliability theory more
explicitly accounts for the uncertainties associated with bridge design and load evaluation. The
proposed formula, however, is considerably more permissive than the FBF when applied to long
vehicles. It results in bridge stresses well above the criteria selected for this Study. Therefore, it
was not considered.

ALLOWABLE WEIGHTS BASED ON FBF STRESS CRITERIA

Original research conducted for this Study suggests that a series of look-up tables may be
developed based on the underlying the FBF stress criteria -- that is, a maximum overstress of
5 percent for HS-20 bridges, and 30 percent for H-15 bridges. These stresses were computed
for both simple and continuous spans for the most critical span lengths for truck configurations.
The following discussion illustrates how this approach might be applied to three vehicles: (1) a
tractor-semitrailer combination vehicle with a 3-axle tractor and 2-axle semitrailer, (2) a tractor-
semitrailer combination vehicle with a 3-axle tractor and a semitrailer with a tridem-axle group,
and (3) a RMD. The GVWs for each configuration with varying semitrailer lengths were
calculated based on axle spacing.

Table VI-3 presents the weight values for the first vehicle combination under the FBF, TTI, and
FBF stress criteria; and Figure VI-2 graphically displays maximum GVW from the Table, for
semitrailers of varying lengths.

12
Bridge Overstress Criteria, Michael Ghosn, Charles G. Schilling, Fred Moses, and Gary Runco, Report by the
City College of the City University of New York for the FHWA (Washington, D.C., FHWA, 1995).

VI-12
 Table VI-3
Maximum GVW For 5-axle Semitrailer Combination Applying
Federal And TTI Bridge Formulas And FBF Stress Criteria

Maximum GVW Maximum GVW

(1,000 Pounds) (1,000 Pounds)
Semitrailer Semitrailer
Length FBF Length FBF
(Feet) FBF TTI Stress (Feet) FBF TTI Stress
Criteria Criteria

28.0 70.0 70.1 78.4 45.0 80.0 80.0 80.0

35.0 74.5 77.1 80.0 48.0 80.0 80.0 80.0

40.0 78.0 80.0 80.0 53.0 80.0 80.0 80.0

NOTE: GVWs specific to 22.5-foot tractor wheelbase, 52-inch tractor tandem spread, and trailer 48-inch tandem
spread. The distance from the first drive axle (on the tractor to the last trailer axle is the trailer length minus
6 feet.

Figure VI-2
Maximum GVW For 5-axle Semitrailer Combination
Applying Federal and TTI Bridge Formulas And FBF Stress Criteria

100

95

90

85

80

75

70

65

60
23 25 27 29 31 33 35 37 39 41 43 45 50 60
Semitrailer Length (feet)

FBF TTI
FBF Criteria

VI-13
Table VI-4 presents weight values and maximum GVWs for the 6-axle semitrailer combination
with the semitrailer supported at the rear by a tridem-axle group. In this case, both the tractor
wheelbase and semitrailer length are varied (common descriptive dimensions). The allowable
GVW for varying semitrailer lengths is shown in Figure VI-3.

Table VI-4
Maximum GVW For 6-axle Semitrailer Combination Applying
Federal And TTI Bridge Formulas And FBF Stress Criteria
Tractor Wheelbase = 22.5 Feet

Semitrailer Maximum GVW Semitrailer Maximum GVW

Length (1,000 Pounds) Length (1,000 Pounds)
(Feet) (Feet)
FBF TTI FBF FBF TTI FBF
Stress Stress
Criteria Criteria

28.0 75.0 70.1 73.4 45.0 85.5 87.1 88.6

35.0 79.5 77.1 84.5 48.0 87.5 90.1 90.0

40.0 82.5 82.1 88.7 53.0 90.5 92.0 94.2

Figure VI-3
Maximum GVW For 6-axle Semitrailer Combination
Applying Federal And TTI Bridge Formulas And FBF Stress Criteria
Tractor Wheelbase = 22.5 Feet
100

95

90

85

80

75
70

65

60
23 25 27 29 31 33 35 40 45 50 55 58
Semitrailer Length (feet)

Legend
FBF TTI FBF Criteria

VI-14
Table VI-5 presents the values and maximum GVWs for the RMD combination, a tractor-
semitrailer combination with a 3-axle tractor pulling a 2-axle semitrailer and a 2-axle full trailer.
The tractor and semitrailer length of this double are varied, with the trailer remaining constant at
28 feet. The limiting axle loads and maximum GVW for the entire vehicle are easily read from
a table. This approach negates the need to compute the many axle group combinations inherent in
the use of the existing and proposed formulas (which can amount to as many as 36 different
combinations in the case of a 9-axle vehicle). The GVW for varying semitrailer lengths is shown
in Table VI-5.

Table VI-5
Maximum GVW for RMD with Semitrailer of Variable Length
And 28' Trailer Applying Federal and TTI Bridge Formulas
And FBF Stress Criteria
Tractor a = 18.2 Feet, Tractor B = 22.5 Feet

Semitrailer GVW
Length (1,000 Pounds)
(Feet)
FBF TTI FBF Stress Criteria

Tractor A Tractor B Tractor A Tractor B Tractor A Tractor B

45 109.5 109.5 105.16 107.3 111.4 112

48 111 111 106.6 108.8 112.8 113.4

53 111 111 109.1 111.3 115.2 116

In summary, there is significant variation in the results derived from the three formulaic
approaches by vehicle configuration. In general, the TTI formula is better matched than the
FBF for bridges, and there is a significant amount of load capacity available before limits are
exceeded for the 5- and 6-axle semitrailer and 7-axle RMD configurations. This is not the case,
however, for larger vehicles such as the 9-axle turnpike doubles -- FBF allows too much weight
for these in terms of the stress criteria. The TTI curve for that vehicle is on the low side of the
FBF stress criteria curve. Also, FBF is conservative for multiaxle short straight trucks.

There are benefits to adhering to the criteria on which the FBF is based and incorporating the
consideration of continuous beams into the control. Tools such as user-friendly computer software
programs can be designed to assess allowable loading configurations for any vehicle, and standard
(bridge formula) tables for the more common vehicles can be generated. The use of the FBF stress
criteria described in this section addresses the documented drawbacks of FBF and provides a
basis for truck weight control that conforms to the criteria upon which both FBF and TTI are based
-- but to which they do not always adhere.

VI-15
It should be noted that the FBF, by design, incorporates a degree of control for pavement damage
by explicitly including the number of axles in the formula. The TTI formula and FBF stress
criteria indirectly control for pavement damage by adhering to axle weight limits -- the higher
GVW limits, such as for LCVs, require more axles to avoid exceeding axle limits.

PAVEMENTS

The condition and performance of highway pavements depend on many factors, including
the thickness of the various pavement layers, quality of construction materials and practices,
maintenance, properties of the roadbed soil, environmental conditions (most importantly rainfall
and temperature), and the number and weights of axle loads to which the pavements are
subjected.13

WEIGHT

While pavement engineers traditionally have used ESAL factors estimated from the AASHO Road
Test (started in 1956 and completed in 1962) as the basis for designing pavements, there
is increasing recognition that better relationships between axle load and pavement deterioration
are needed. Pavement distress models used in both the 1982 and 1997 Federal HCA Studies
(HCAS) abandoned the use of ESALs to relate axle loading to pavement deterioration, and
AASHTO will be replacing its ESAL-based pavement design formula with one that more directly
relates axle loads to factors that determine pavement life. While ESALs were not used as the
basis for estimating pavement costs for this Study, they are widely understood by highway
administrators, pavement engineers, and others concerned with the pavement impacts of TS&W
scenarios. Therefore, they are used here as a benchmark for comparing relative pavement impacts
of various truck configurations with different numbers and types of axles.

Pavement deterioration increases sharply with increases in axle load. On both flexible and
rigid pavements, the load equivalence factor for a 20,000-pound single axle is about 1.5. Thus,
100 passes across a pavement by a 20,000-pound axle would have the same effect on pavement
life as 150 passes by an 18,000-pound axle.

The number of axles is also important in estimating pavement impact, other things being equal, as a
vehicle with more axles has less effect on pavements. For example, a 9-axle combination vehicle
carrying 80,000 pounds has less effect on pavements than a 5-axle combination vehicle carrying
80,000 pounds. A significant amount of additional weight can be carried by the 9-axle vehicle
without causing greater pavement consumption relative to the 5-axle vehicle. Comparing vehicles
in terms of ESALs provides information on load-related pavement impact, but it does not include
an offsetting benefit gained by a reduction in the number of trips required to transport

13
TRB Special Report 225, Truck Weight Limits: Issues and Options, 1990.

VI-16
the same amount of freight. Vehicles are often compared in terms of ESALs per unit of freight
carried as a means of including the reduction in pavement deterioration from fewer trips.

The increase in pavement costs per added ESAL mile can vary by several orders of magnitude
depending upon pavement thickness, quality of construction, and season of the year. Thinner
pavements are much more vulnerable to traffic loadings than thicker pavements.14 Additionally,
pavements are much more vulnerable to traffic loadings during spring thaw in areas subject to
freeze-thaw cycles.

AXLE SPACING

The primary load effect of axle spacing on flexible pavement performance is fatigue.
Axle spacing is a major concern for fatigue. When widely separated loads are brought closer
together, the stresses they impart to the pavement structure begin to overlap, and they cease to
act as separate entities. While the maximum deflection of the pavement surface continues to
increase as axle spacing is reduced, maximum tensile stress at the underside of the surface layer
(considered to be a primary cause of fatigue cracking) can actually decrease as axle spacing is
reduced. However, effects of the overlapping stress contours also include increasing the duration
of the loading period. Thus, the beneficial effects of stress reduction are offset to an unknown
degree by an increase in the time or duration of loading. The net effect of changes in axle spacing
on pavement deterioration is complex and highly dependent on the nature of the pavement
structure.15

TIRE CHARACTERISTICS

In recent years, several studies on the impact of tire characteristics on pavement have raised
concern over the possibility of accelerated pavement deterioration, particularly rutting, caused
by increasing tire pressures. The tires of the AASHO Road Test trucks of the 1950s were bias-ply
construction with inflation pressures between 75 pounds and 80 pounds per square inch (psi). The
replacement of bias-ply tires with radial tires and higher inflation pressures, averaging 100 psi,
result in a smaller size tire “footprint” on the pavement and, consequently, a concentration of
weight over a smaller area.16 These changes hasten the wear of flexible pavements, increasing
both the rate of rutting and the rate of cracking.

14
Results of a study by Hutchinson and Haas compare the average and marginal costs per ESAL on highways with
500,000 ESALs per year and 2 million ESALs per year. The cost per ESAL for highways with 500,00 ESALs
is almost four times as great as the cost per ESAL on highways designed for 2 million ESALs. One important
implication of this finding is that a policy that encourages heavy trucks to shift from highways with thicker
pavements, such as the Interstate or NHS, to highways with thinner pavement can have a significant impact on
pavement costs.
15
TRB Special Report 225.
16
A study by Bartholomew (1989) summarized surveys of tire pressure conducted in seven States between 1984
and 1986 and found that 70 to 80 percent of the truck tires used were radials and that average tire pressures
were about 100 psi.

VI-17
The AASHTO load equivalency factors apply only to axles supported at each end by dual tires.
Recent increases in steering axle loadings and more extensive use of single tires on load-bearing
axles have precipitated efforts to examine the effect on pavement deterioration of substituting
single for dual tires. Both standard and wide-based tires have been considered. Past
investigations of the pavement deterioration effects of single versus dual tires have found that
single tires induce more pavement deterioration than dual, but that the differential wear effect
diminishes with increases in pavement stiffness, in the width of the single tire, and in tire load.17

A general finding from the studies is that wide-base single tires appear to cause about 1.5 times
more rutting than dual tires on flexible pavements (the most common type of pavement) as they do
not have good rut resistance. Another finding is that one of the wheels in a dual tire assembly is
frequently overloaded due to variability in the roadway cross-section and that the average
overload causes an increase in rutting similar to that caused by wide-based single and dual tire
assemblies.

Based upon past studies, single tires have more adverse effects on pavements than dual tires,18
it appears likely, however, that past investigations have overstated the adverse effects of single
tires by neglecting two potentially important effects: (1) unbalanced loads between the two tires of
a dual set, and (2) the effect of randomness in the lateral placement of the truck on the highway.
Unbalanced loads between the tires of a dual set can occur as a result of unequal tire pressures,
uneven tire wear, and pavement crown. As with unequal loads on axles within a multiaxle group,
pavement deterioration increases as the loads on the two dual tires become more unbalanced.

The second neglected factor, sometimes termed “wander,” is the effect of randomness in the lateral
placement of trucks within and sometimes beyond lane boundaries. Less than perfect tracking is
beneficial to pavement deterioration, as the fatiguing effect is diminished because the repetitive
traffic loads are distributed over wider areas of the pavement surface. The greater overall width
of dual tires naturally subjects a greater width of pavement to destructive stresses, therefore,
wander is expected to have a smaller beneficial effect for dual than for single tires. Once rutting

17
Gillespie (1993) found that a steering axle carrying 12,000 pounds with conventional single tires is more
damaging to flexible pavements than a 20,000-pound axle with conventional dual tires. Gillespie proposed that
road damage from an 80,000-pound vehicle combination would be decreased by approximately 10 percent if a
mandated load distribution of 10,000 pounds on the steering axle and 35,000 pounds on tandems. Since the
operating weight distribution of a 5-axle tractor-semitrailer at 80,000 pounds GVW generally has less than
11,000 pounds on the steering axle, the practical effect of the proposal would be to increase tandem axle
weights without a compensating decrease in steering axle weights.
18
Bauer (1994) summarized several recent studies on the effects of single versus dual tires: “Smith (1989), in a
synthesis of several studies . . . evaluated at 1.5 on average the relationship of the damage caused by wide base
single assemblies and that caused by traditional dual tire assemblies with identical loading at the axle. Sebaaly
and Tabataee (1992) found rutting damage ratios between wide base and dual tire assemblies varying between
1.4 and 1.6 . . . Bonaquist (1992), reporting on results obtained from a study . . . on two types of roadway, using
a dual tire assembly with 11 R 22.5 and a wide base with 425/65 R 22.5, indicates rutting damage ratios varying
from 1.1 to 1.5, depending on the layers of the roadway.”

VI-18
begins, however, tires -- especially radial tires -- tend to remain in the rut, thereby greatly
reducing the beneficial effects of wander for both single and dual tires.19

Another consideration in evaluating wide-base single versus dual tires is dynamic loadings that
arise from the vertical movement of the truck caused by surface roughness. Thus, peak loads
are applied to the pavement that are greater than the average static load.20 Signs of pavement
damage from dynamic loadings are typically localized, at least initially. Because of the localized
nature of the dynamic loading, its severity is much greater than previously thought. 21 A further note
on wide-base single tires is that those having only two sidewalls are much more flexible than a
pair of dual tires with four sidewalls. This means the tire absorbs more of the dynamic bouncing
of the truck, and less of the dynamic load is transmitted to the pavement.

SUSPENSION SYSTEMS

The subject of road-friendly suspensions -- within the context of the broader subject of vehicle-
pavement interaction -- was researched as an Organization for Economic Cooperation and
Development (OECD) Project -- the Dynamic Interaction between Vehicles and Infrastructure
Experiment (DIVINE) Project -- involving the United States and 16 other countries.22 The work
focused on (1) how well different suspension systems distribute load among axles in a group (the
more evenly, the better); (2) how well different suspension systems dampen vertical dynamic

19
The TRB Special Report 225 examined the importance of loading imbalance and wander. The TRB Study
examined two types of pavement deterioration: surface cracking due to fatigue and permanent deformation or
rutting in the wheel tracks. Fatigue was found to be more sensitive to the differences between single and dual
tires than rutting. Both balanced and unbalanced dual-tire loads were considered in analyzing the affect on
wander. The analysis indicated that the adverse effects of single tires on pavement deterioration were reduced
when wander was taken into account, although the effects were still significant.
20
From research summarized by the Midwest Research Institute (MRI) that suggests dynamic loadings are a
consideration in assessing the relative merits of wide base single versus dual tires. Gyenes and Mitchell report
that the magnitude of the added dynamic components was earlier thought to increase road damage over that of
the static loading alone between 13 and 38 percent, according to research reported by Eisenmann. The MRI
research noted that many recent studies have pointed out the fallacy in the earlier work, which assumed that the
dynamic component of loading was distributed uniformly over the pavement in the direction of travel. The
research found, however that the dynamic component is very localized, arising out of pavement surface
irregularities and therefore is spatially correlated with these irregularities.
21
Gillespie, et. al. estimate that damage due to the combination of static and dynamic loading can be two to four
times that due to static loading locally. Von Becker estimates the combined loading produces a “shock factor”
between 1.3 and 1.55, depending upon suspension characteristics. Applying the fourth power law would
translate these figures into relative damage estimates ranging from 2.8 to 4.8 times the static loading damage.
Gyenes and Mitchell suggest impact factors in the range of 1.3 to 1.5 for relative damage estimates of 2.8 to
5.1.
22
TRB Special Report 225 noted that a heavy truck travels along the highway, axle loads applied to the pavement
surface fluctuate above and below their average values. The degree of fluctuation depends on factors such as
pavement roughness, speed, radial stiffness of the tires, mechanical properties of the suspension system, and
overall configuration of the vehicle. On the assumption that the pavement deterioration effects of dynamic
loads are similar to those of static loads and follow a fourth-power relationship, increases in the degrees of
fluctuation increase pavement deterioration.

VI-19
loads (the more, the better); and (3) spatial repeatability of dynamic loads. The research also
examines how road and bridge characteristics act to excite a truck, and in turn influence the loads
received by the road and bridge.

The findings of the DIVINE research primarily relate to the physical interaction between heavy
vehicles and the highway infrastructure -- pavements and bridges. The research breaks new
ground, providing scientific evidence of the effects of heavy vehicles. Conclusions that relate to
vehicle and pavement interaction are summarized from the final report.

Pavement wear -- the gradual loss of functional condition -- is expressed in permanent

deformations to the longitudinal profile of the pavement surface. Whereas, pavement damage
results from an accumulation of rutting and cracking distress from repeated applications of vehicle
loads. "Road research . . . has historically tended to over-emphasize pavement damage, and the
true importance and nature of pavement wear has not yet been recognized."23 The DIVINE
research focused primarily on examining pavement wear rather than damage.

Two scientific breakthroughs resulted from the DIVINE accelerated pavement tests: "the effects of
dynamic loading were measured for the first time, and a detailed statistical analysis of both the
pavement and vehicle variables was undertaken."24 Conclusions reached are:

• Changes in pavement profile under dynamically-active steel suspensions relate to: local
structural compliance (the opposite of strength), and local dynamic wheel load.

• Changes in pavement profile under dynamically-quiet air suspensions are mainly related
to the local structural compliance of the pavement.

• The relationship between tensile strain at the bottom of the pavement surfacing layer
and dynamic wheel loading appears to depend on the pavement thickness. For thick pavement,
strain is directly related to dynamic wheel loading. For thin pavement, strain directly related
to dynamic wheel loading is weaker. This difference in pavement behavior is believed to
be related to changes in tire contact conditions occurring from variances in the dynamic wheel
load.

• Air suspension would increase pavement life by 60 percent for thick pavement and 15 percent
for thin pavement (based on two types of implied assumptions: selected pavement response
parameter measured and analyzed, and the "damage law" applied).

• Spatial repeatability on a relatively smooth road would increase total wheel loading at
certain locations by approximately 10 percent, reducing pavement life at those locations
by approximately 35 percent to 50 percent.

23
OECD DIVINE Programme, Final Report "Dynamic Interaction of Heavy Vehicles with Roads and Bridges,"
May 1997, p. 145.
24
Ibid.

VI-20
The findings indicate that "pavement wear is the key concept to be used in the scientific
consideration of the effect of heavy vehicles on highway pavements."25

Additionally, recent research outside the DIVINE Program evaluated the role of suspension
damping in enhancing the road friendliness of a heavy vehicle. The findings indicated an increase
in linear suspension damping tends to reduce the dynamic load coefficient and the dynamic tire
forces -- factors related to road wear. The research concluded that linear and air spring
suspensions with light linear damping offer significant potentials to enhance the road friendliness
of the vehicle with a slight deterioration in ride quality. 26 It is worth noting that approximately 90
percent of all truck-tractors and 70 percent of all van trailers sold in the United States are
equipped with air suspensions. Additional studies on various types of axle suspension systems
include studies on: torsion suspensions, four-leaf suspensions, and walking-beam suspensions.27

The research has yet to produce any compelling argument to incorporate a suspension system
determinant into U.S. regulations, although some countries have done so. Mexico is in the final
stages of preparing regulations that will allow up to 2,200 pounds of additional weight for each
trailer axle equipped with air suspension or its equivalent. For a drive axle, Mexico may allow up
to an additional 3,300 pounds. The impacts of different suspension systems on pavement
deterioration are of secondary importance compared to the static axle load levels themselves. Use
of road-friendly suspensions is beneficial, particularly for large trucking operations with well-
controlled axle loadings.

LIFT AXLES

The widespread use of lift axles in Canada and the United States raises concern for resulting
pavement deterioration when a driver, attempting to improve fuel consumption, fails to lower
the axle when loaded. A 1988 and 1989 survey conducted in Ontario and Quebec found that
approximately 17 percent and 21 percent, respectively, of trucks on highways in those Provinces
had lift axles.28 Lift axles have been adopted in response to GVW limits governed by the number

25
Ibid, p. 147.
26
In the Rakheja and Woodroofe model suspension effects are represented using a sprung mass, an unsprung
mass, and restoring and dissipative effects due to suspension and tire. The tire is modeled assuming linear
spring rate, viscous damping, and point contact with the road.
27
Sousa, Lysmer and Monismith investigated the influence of dynamic effects on pavement life for different
types of axle suspension systems. They calculated a Reduction of Pavement Life (RPL) index of 19 percent
for torsion suspensions (an ideal suspension would have RPL of 0). Similar results were found by Peterson in
a study for RTAC: under rough roads at 50 mph, air bag suspensions exhibited dynamic loading coefficients
(DLC) of 16 percent, spring suspensions had a DLC of 24 percent, and rubber spring walking beam suspensions
had a DLC of 39 percent. Problems with walking-beam suspensions were also noted by Gillespie, et. al. who
state that on rough and moderately rough roads, walking-beam suspensions without shock absorbers are
typically 50 percent more damaging than other suspension types.
28
Billing, et. al.

VI-21
of axles (such as the FBF), and because trucks with multiple widely spaced axles have difficulty
turning on dry roads and the lift axles can be raised by the driver prior to turns.

Lift axles make compliance with and enforcement of axle weight limits difficult. Improperly
adjusted lift axles can damage pavements. The lift axle can be adjusted to any level by the driver.
If the lift axle load is too high, the lift axle is overloaded. If it is too low, other axles may be
overloaded. For example, under current Federal limits, a 4-axle single unit truck
with a wheelbase of 30 feet can carry 62,000 pounds: 20,000 pounds on the steering axle and
42,000 pounds on the rear tridem. This vehicle would produce approximately 2.1 ESALs on
flexible pavements. However, if the first axle of the tridem is a lift axle carrying little or no
weight, this vehicle would produce approximately 4.0 ESALs.

PAVEMENT COST

Unit pavement costs and pavement costs per unit of payload-mile by configuration are shown
in Tables VI-6 and VI-7. They illustrate how the addition of axles allows for increased payloads
and at the same time reduces pavement deterioration. Particularly striking, are comparisons
between the 3- and 4-axle single unit trucks, the 5- and 6-axle semitrailer combinations, and the 5-
and 8-axle doubles. As shown in Table VI-7, the 4-axle truck has costs per payload ton-mile
about 75 percent of that for the 3-axle truck even though its gross weight is 10,000 pounds more
than the 3-axle truck. The comparison of the 6-axle semitrailer with the 5-axle is very similar on
non-Interstate highways. The costs for the 8-axle double-trailer are less than half those for the
5-axle double-trailer. Triples do not compare well with doubles. Generally, truck owners would
be opposed to adding axles because this increases the tare weight of the vehicle and reduces
payload capacity.

TS&W REGULATION RELATED TO PAVEMENT PRESERVATION

TIRE REGULATIONS

Federal law and most State laws, do not address truck tire pressure. Tire pressure may have a
large effect on fatigue of flexible pavements as discussed earlier (albeit a small to moderate effect
on rigid pavements), and today's tire pressures are higher than in the 1950s -- primarily the
consequence of a change from bias to radial ply tires. Concern has been raised about accelerated
pavement rutting as a result of increased tire pressures. Recent research gives conflicting views
as to whether or not pressures should be regulated.29

Federal, and most State, laws do not discourage or prohibit the use of wide-base tires. The
consensus of United States and international research is that these tires have substantially more

29
TRB Special Report 225 (1990) suggested regulation could be warranted if the more pessimistic analyses
proved to be correct. NCHRP Study (1993) suggested limiting tire pressure to the recommended cold setting
plus 15-psi; AASHTO (1993) suggested more research is required to answer all questions regarding the
relationship of tire size, contact pressure, and contact area to pavement damage.

VI-22
adverse effects on pavements than dual tires because current designs employ smaller, overall
tire-road contact patch sizes than equivalent dual tire sizes. Future tire designs could address this
issue. Wide-base tires -- which are widely used in Europe -- are being increasingly adopted by
U.S. trucking operations. The benefits of wide-base tires are reduced energy use, emissions,
tire weights, and truck operating costs. The trade off between changes in Federal pavement costs
and operating benefits that would result from permitting or prohibiting extensive adoption of wide-
base tires in the United States has not been analyzed.

Table VI-6
Unit Pavement Cost For Various Truck Types
$/1,000 MILES

Truck Type

Single-Unit Semitrailer Double-Trailer Triple

3-Axles 4-Axles 5-Axles 6-Axles 5-Axles 7-Axles 8-Axles 7-Axles

GVW 54,000 64,000 80,000 90,000 80,000 100,000 105,000 100,000 115,000
(Pounds)

Area Functional
Type Class

Rural Interstate 0.09 0.07 0.05 0.05 0.03 0.10 0.05 0.04 0.08

Prin. Art. 0.17 0.16 0.12 0.11 0.07 0.15 0.10 0.17 0.31

Min. Art. 0.37 0.33 0.29 0.22 0.32 0.41 0.21 0.39 0.75

Maj. Col. 1.38 1.35 0.90 0.80 1.17 1.03 0.65 1.46 2.95

Min. Col. 2.27 2.08 1.49 1.24 1.92 1.69 1.07 2.42 4.87

Locals 5.90 5.63 3.87 3.23 4.99 4.40 2.79 6.27 12.60

Urban Interstate 0.06 0.04 0.04 0.04 0.03 0.04 0.02 0.03 0.05

Freeway & 0.09 0.06 0.06 0.05 0.04 0.07 0.04 0.09 0.18
Expressway

Prin. Art. 0.13 0.12 0.10 0.09 0.11 0.09 0.06 0.13 0.26

Min. Art. 0.30 0.24 0.22 0.17 0.19 0.18 0.12 0.34 0.70

Collectors 0.66 0.70 0.54 0.49 0.46 0.34 0.25 0.86 1.82

Locals 2.34 2.53 1.91 1.75 1.64 1.19 0.88 3.06 6.45

Historically, many States specified some form of tire load regulation for safety. In recent years,
additional States have adopted tire load regulations to control the damage effect of wide-base
tires. They restrict the weight that can be carried on a tire based on its width. The limits range
from 550 pounds per inch (in Alaska, Mississippi, and North Dakota) to 800 pounds per inch (in
Indiana, Massachusetts, New Jersey, New York, and Pennsylvania). Such restrictions result in
lower pavement costs; however, the size of the pavement cost savings (either in absolute terms or
in relation to the increase in goods movement costs also resulting from these restrictions) have not
been estimated.

VI-23
 Table VI-7
Unit Cost per Payload-mile for Various Truck Types
$/1,000 Ton-miles

Truck Type

Single-Unit Semitrailer Double-Trailer Triple

Weights 3-Axles 4-Axles 5-Axles 6-Axles 5-Axles 7-Axles 8-Axles 7-Axles

(Pounds)

GVW 54,000 64,000 80,000 90,000 80,000 100,000 105,000 100,000 115,000

Tare 22,600 26,400 30.490 31,530 29,320 38,600 33,470 41,700 41,700

Payload 31,400 37,600 49,510 58,470 50,680 61,400 71,530 58,300 73,300

Area Functional
Type Class

Rural Interstate 0.006 0.004 0.002 0.002 0.001 0.003 0.001 0.001 0.002

Prin. Art. 0.011 0.009 0.005 0.004 0.003 0.005 0.003 0.006 0.008

Min. Art. 0.024 0.018 0.012 0.008 0.013 0.013 0.006 0.013 0.020

Maj. Col. 0.088 0.072 0.036 0.027 0.046 0.034 0.018 0.050 0.080

Min. Col. 0.145 0.111 0.060 0.042 0.076 0.055 0.030 0.083 0.133

Locals 0.376 0.299 0.156 0.110 0.197 0.143 0.078 0.215 0.344

Urban Interstate 0.004 0.002 0.002 0.001 0.001 0.001 0.001 0.001 0.001

Freeway & 0.006 0.003 0.002 0.002 0.002 0.002 0.001 0.003 0.005
Expressway

Prin. Art. 0.008 0.006 0.004 0.003 0.004 0.003 0.002 0.004 0.007

Min. Art. 0.019 0.013 0.009 0.006 0.007 0.006 0.003 0.011 0.019

Collectors 0.042 0.037 0.022 0.017 0.018 0.011 0.007 0.030 0.050

Locals 0.149 0.136 0.077 0.060 0.065 0.039 0.024 0.105 0.176

SPLIT-TANDEM VERSUS TRIDEM-AXLE LOAD LIMITS

There is increasing use of split tandem axle groups with spreads up to 10 feet, particularly in
flatbed heavy haul operations. These axles are allowed to be loaded at single axle limits --
20,000 limits -- 20,000 pounds on each of the 2 axles -- as opposed to 34,000 pounds on a
closed tandem when they are split more than 8 feet. They offer two key benefits to 5-axle tractor-
semitrailer usage: (1) flexibility in load distribution; and (2) full achievement of the 80,000-pound
GVW cap, which is limited by the ability to distribute up to 12,000 pounds on the steering axle of
a combination. But they do so at a significant cost to pavement life.

In the United States, the allowable load on a group of three axles connected by a common
suspension system (tridem) is determined by the Federal bridge formula rather than a limit set by
law (or regulation). In Europe, Canada, Mexico, and most other jurisdictions, tridem axles are
given a specific load limit in the same way the United States specifies single and tandem axle

VI-24
limits without direct reference to a bridge formula. This is not to say that these tridem limits are
not bridge-related. For example, the tridem limits prescribed by the RTAC, which vary as a
function of spacing, are based on bridge loading limitations -- not pavement limitations.

THE GVW LIMIT

The existing legal Federal maximum GVW (cap) limit for the Interstate System is 80,000 pounds,
although some States allow truck combination weights above this cap under Federal grandfathering
provisions. Axle weight limits and the FBF are designed to protect pavements and bridges,
respectively. As such, the cap may not be providing any additional protection to pavements and
bridges. Nevertheless, it is important to consider such factors as bridge design loads and criteria,
structural evaluation procedures, the age of the existing bridges, and the extent to which increased
GVWs would affect the fatigue life of bridges in the United States.

44,000-POUND TRIDEM-AXLE WEIGHT LIMIT

Original research done for this Study on the pavement and bridge impacts of tridem axles showed
how bridge stresses decrease as the axles in the tridem group are spread apart. This allows more
weight to be carried on the tridem group as the axles are spread. The opposite is true for
pavement damage. The more the axles are spread, the greater the damage. Therefore, as the axles
are spread within the group, the allowable weight must be reduced to hold pavement damage
constant.

The tridem-axle weight limit of 44,000 pounds was determined by observing where the curve of
the increasing bridge allowable load function crosses the curve of the decreasing pavement load
equivalency function (see Figure VI-4). The two curves cross at a spread of 9 feet between the
two outer axles which gives 44,000 pounds for both functions. To stop short of 9 feet would
require a lower load limit as bridge damage would be greater than at 44,000 pounds. To go
beyond 9 feet would increase pavement damage over that at 44,000 pounds.

A 6-axle semitrailer combination is more effective in reducing pavement damage than a 5-axle
semitrailer combination with a split tandem (two trailer axles spread apart), which is allowed
under the current FBF. Table VI-8 provides the weight limits for a tridem axle between 8 and 16
feet and Figure VI-4 illustrates the impact on pavement and bridges.

Table VI-8
Tridem-axle Weight Limits

Axle Distance Between Load at LEF=1 Allowable

Spreads Adjacent Axles Bridge Load
(Feet) (Feet) (1,000 Pounds)

8 4 45 43

12 6 42 48.6

16 8 40 ------

VI-25
 Figure VI-4
Pavement and Bridge Impact of Tridem-axle

TRIDEM LOADING
48
47
46
45
44
43
42
41
40
4 4.5 5 5.5 6
Axle Spacing (feet)

Pavement Bridge

USE OF TRIDEMS

The use of tridem axles could increase truck load capacity while reducing pavement damage.30
Many heavy bulk haulers have already switched from 3-axle to 4-axle single unit trucks, and as
noted above, significant pavement cost savings may be possible. The 80,000-pound GVW limit
poses a constraint on adding axles to 5-axle combinations because the extra axle would reduce the
payload.

When viewed using the AASHTO load equivalence factors, combinations with tridem axles
generally have much lower pavement costs per ton of freight carried than conventional 5-axle
combinations. To illustrate this, as shown in Figur VI-5, a 6-axle tractor-semitrailer
loaded to 90,000 pounds with a rear tridem carrying 44,000 pounds produces 2.00 ESALs on
flexible pavements and 3.83 ESALs on rigid pavements. The corresponding ESAL values for
a conventional 5-axle tractor-semitrailer carrying 80,000 pounds are 2.37 (flexible) and 3.94
(rigid).

Assuming tare weights of 28,000 and 29,500 pounds for the 5- and 6-axle combinations,
respectively, and using the AASHTO load equivalence factors, the ESALs per million pounds
of payload for the trucks shown in Figure VI-5 are shown in Table VI-9.

30
Both the TRB Special Report 225 and the AASHTO TS&W Subcommittee suggest consideration of the TTI
bridge formula which could allow about 90,000 pounds for a 6-axle tractor-semitrailer combination.

VI-26
 Figure VI-5
ESAL Comparison of 5-axle and 6-axle Combinations on Pavement

5-Axle Tractor-Semitrailer

TOTAL
Weight (pounds) 34,000 34,000 12,000 80,000
ESALs
Flexible 1.09 1.09 0.19 2.37
Rigid 1.88 1.88 0.18 3.94

6-Axle Tractor-Semitrailer

TOTAL
Weight (pounds) 44,000 34,000 12,000 90,000
ESALs
Flexible 0.72 1.09 0.19 2.00
Rigid 1.77 1.88 0.18 3.83

Table VI-9
ESALS per Million Pounds Payload for 5- and 6-axle Combinations

Flexible Pavement Rigid Pavement

5-Axle Tractor-Semitrailer 46 76

6-Axle Tractor-Semitrailer 33 63

VI-27
 ROADWAY GEOMETRY

ELEMENTS OF ROADWAY GEOMETRY AFFECTING TRUCK OPERATIONS

INTERCHANGE RAMPS

Access and exit ramps for controlled access highways are intended to accommodate design
vehicles at certain design speeds. Otherwise, trucks heavier than the design vehicle have
an increased probability of rolling over, and trucks longer than the design vehicle will have trailer
wheels that travel off the pavement to the inside of a curve. The TS&W, configuration, and speed
influence the potential for rollover on short loop ramps. The AASHTO policy recommends
widening ramps to accommodate combination vehicles. For example, the width of a 1-lane ramp,
with no provision for passing a stalled vehicle, would be 15 feet on a tangent section.

The extreme case for design consideration occurs when traffic is congested and stop-and-go
conditions exist. The speed component to the offtracking equation is negligible and maximum
offtracking to the inside of the curve occurs. Under this condition, the turnpike doubles analyzed in
this study offtrack 20 percent more than a 5-axle 53-foot semitrailer combination and as a result,
encroach on adjacent lanes or shoulders and necessitate widening beyond AASHTO standards.

INTERSECTIONS

Most truck combinations turning at intersections encroach on either the roadway shoulder
or adjacent lanes. For example, the turning path of a truck making a right turn is generally
controlled by the curb return radius, whereas the turning path in left turns is not constrained by
roadway curbs, but may be constrained by median curbs and other traffic lanes. Combination
vehicles with long semitrailers are critical in determining needed intersection improvements to
accommodate offtracking requirements. Additionally, the increased time required for a large truck
to complete its turn requires longer traffic signals and affects pedestrian safety and intersection
efficiency. Figure VI-6 illustrates the intersection maneuver.

Proper design and operation requires that no incursion into the path of vehicles traveling in
opposing directions be allowed. A higher standard is often used in design, especially in urban
areas, where no incursion into any adjacent lane is allowed. This is particularly critical at
signalized intersections where heavy traffic is a prevailing condition. A substantial number of
intersections on the existing highway and street network cannot accommodate even a 5-axle
tractor-semitrailer combination with a 48-foot semitrailer. Even more intersections would be
inadequate to accommodate vehicles that offtrack more than the standard 48-foot semitrailer
combination.

VI-28
 Figure VI-6
Path of Tractor Semitrailer Keeping Tires Within Lanes

NOTE: Distance from kingping to rear axle is 40 feet; distance from rear axle to rear of trailer is 14.5 feet

VI-29
Currently, there are a substantial number of intersections on the highway and street network where
improvements for combinations with semitrailers over 48 feet are not feasible and where controls
on vehicles, routing, or travel times are needed. Examples of common constraints to intersection
improvements are bridges, buildings and sensitive environmental or historic plots. The use of
permits in such cases can provide a desirable level of control. Another option for States might be
the provision of staging areas where routes and intersections have prohibitive constraints off
Interstate-type highways.

CLIMBING LANES

The ability of a truck to maintain speed on a grade is described by the term “gradeability;” the
truck’s ability to start on a grade from a standstill is termed “startability.” The ability of various
trucks to start and to maintain speeds on grades is a complex subject that primarily depends on net
engine horsepower, torque, gearing, drive train efficiency, friction, GVW, and minimum allowable
speed. Gradeability and startability are discussed in Chapter 5, Safety and Traffic Operations.
The AASHTO recommends that separate climbing lanes be provided on grades that have
substantial truck traffic or that cause typical trucks to slow by more than 10 miles per hour.31

CROSS-SECTION

Cross-section refers to the shape of the surface of the roadway perpendicular to the direction of
traffic.32 Under normal operating conditions, cross-section is not a dominant factor in increased
TS&W, but under extreme icing conditions, a superelevated cross slope can be a significant
problem for vehicles with greater offtracking. The presence of cross-slope discontinuities can
also be a problem for vehicles more prone to rollover because of the dynamic forces that they tend
to introduce.

HORIZONTAL CURVATURE

The rear wheels of trucks and truck combinations traversing horizontal curves generally offtrack to
one side or the other of the paths of the wheels on the steering axle. When a truck is traveling at
higher speeds the rear wheels can follow a path outside that of the steering wheels. This effect is
relatively small and virtually never results in the need to make geometric improvements beyond
those normally made in the design process. On the other hand, when offtracking is to the inside of
the curve at lower speeds and in stop-and-go traffic, it is usually more substantial and must be
accommodated. Truck combinations with longer trailers are often prone to producing relatively
large amounts of offtracking beyond that provided for in AASHTO

31
Substantial is not defined by AASHTO. There is no universally acceptable standard and it is left to the States to
define.
32
The major determinants of the cross section are the number of lanes, the presence of curbing or shoulders, and
cross slope. Generally, a slight cross slope is designed into the cross section to assist in proper drainage of
precipitation. Often this slope breaks to a steeper slope at the shoulder line, on a divided multilane highway the
grade or elevation is generally highest at the centerline.

VI-30
standards. For roadways not constructed to AASHTO standards more improvement would be
required to accommodate longer combinations where offtracking would exceed normal lane width.

VERTICAL CURVE LENGTH

The height of the truck driver's eye is a distinct advantage of trucks over passenger vehicles for
crest vertical curves that are designed to maximize stopping sight distance. Vertical curves are
generally designed for passenger cars, as a passenger car driver's eye is lower than is a truck
driver’s. For a sag vertical curve going from a downgrade to an upgrade, headlight coverage and
passenger comfort usually control. The vehicles considered in this study have braking distances
similar to vehicles in common use at this time; therefore, no geometric adjustments would be
required.

PASSING SIGHT DISTANCES

Distances required for passing trucks can be significantly longer than for automobiles and pickups.
Longer trucks increase the distance required for a car or truck to pass and require more care in
order do so safely. Drivers of passenger cars passing trucks, and drivers of trucks who desire to
pass other vehicles, are expected to follow the rules of the road and exercise discretion, passing
only where sight distance is adequate. On multilane highways, passing is not as critical as
passing on a 2-lane highway with traffic in opposing directions. Sight distance criteria for
marking passing and no-passing zones on 2-lane highways are more appropriate for a passenger
car passing another passenger car: they do not consider trucks, even the standard truck-and 48-foot
semitrailer combination vehicle at 80,000 pounds.

The additional lengths of LCVs could require as much as 8 percent more passing sight distance for
cars passing LCVs on 2-lane roads; longer and/or heavier trucks would require incrementally
longer passing sight distances to pass cars safely on 2-lane roads.

DIMENSIONAL LIMITS IMPACTING TRUCK MANEUVERS

LENGTH LIMITS FOR SEMITRAILERS

The STAA of 1982 requires States to allow the operation of a semitrailer of at least 48 feet long
on the NN. All States now allow up to 53 feet on at least some highways. The majority of States
prohibit semitrailers longer than 53 feet, the exceptions being Alabama, Arizona, Arkansas,
Colorado, Florida, Kansas, Louisiana, New Mexico, Oklahoma, Texas, and Wyoming. 33 Most of
these States allow trailers in the 57- to 60-foot range to operate.

33
Federal Size Regulations for Commercial Motor Vehicles, U.S. DOT, Publication Number FHWA-MC-96-
03.

VI-31
LENGTH LIMITS FOR DOUBLE TRAILERS IN COMBINATION

The STAA of 1982 also established a requirement for States to allow, at a minimum, the
operation of two 28-foot trailers (twins) in combination on the Interstate and NN. About
one-fourth of the States prescribe 28 feet as a maximum; the others allow additional length up to 30
feet with 28.5 feet being the most common. Prior to passage of the ISTEA, Federal law allowed
States to permit longer trailers in combination (commonly referred to as doubles) but did not
require States to do so.

OVERALL LENGTH LIMITS

The STAA of 1982 established a prohibition against State laws specifying a maximum
length for semitrailer and STAA double combinations operating on the Interstate and NN.
Consequently, most States control total length on the NN by limiting semitrailer and trailer lengths.
About two-thirds of the States have some form of control of total combination length for non-NN
highways. While there are no proposals that the Federal law prescribe a total length limit at this
time, offtracking standards could effectively limit overall lengths for single- and double-trailer
combinations.

VEHICLE WIDTH AND HEIGHT LIMITS

Vehicle widths and heights are important from the standpoint of safety and traffic operations. The
effect on roadway geometric design relates to lane and shoulder width and vertical clearances. A
1-lane ramp with a narrow shoulder would result in a blockage if a truck were disabled. Many
older structures (overpasses) were constructed with minimal vertical clearances. The addition of
pavement overlays over the years may have further reduced these clearances. Increases in vehicle
height increases the potential for striking these overhead structures as well as vehicle rollover.

ROADWAY GEOMETRY AND TRUCK OPERATING CHARACTERISTICS

When a vehicle makes a turn, its rear wheels do not follow the same path as its front wheels.
The magnitude of this difference in path, known as “offtracking,” generally increases with the
spacing between the axles of the vehicle and decreases for larger radius turns. Offtracking
of passenger cars is minimal because of their relatively short wheel bases; however, many
trucks offtrack substantially. The magnitude of the offtracking is often measured by the differences
in the paths of the centerlines of the front and subsequent axles. The maximum extent of offtracking
for a turn of a given radius and length occurs at the rearmost axle or the center of the rearmost axle
group.

Offtracking develops gradually as a vehicle enters a turn and, if the turn is long enough, eventually
reaches what is termed as fully-developed offtracking. The offtracking does not continue to
increase beyond this point for curves that are any longer. The extent of this fully-developed
offtracking is used to determine if the nominal lane width can accommodate the offtracking or how
much the lane should be widen through the curve to accommodate the offtracking characteristics of
the trucks using the highway.

VI-32
In contrast, for a short radius 90-degree turn such as a truck would make at an intersection, the turn
is too short for fully-developed offtracking to occur. Nevertheless, the maximum extent of
offtracking may be readily calculated for designing an intersection that can accommodate the trucks
expected to make right turns at the intersection.

LOW-SPEED OFFTRACKING

When a combination vehicle makes a low-speed turn -- for example a 90-degree turn at an
intersection -- the wheels of the rearmost trailer axle follows a path several feet inside the path of
the tractor steering axle. This is called low-speed offtracking. Excessive low-speed offtracking
may make it necessary for the driver to swing wide into adjacent lanes to execute the turn (that is,
to avoid climbing inside curbs or striking curbside fixed objects or other vehicles). When
negotiating exit ramps, excessive offtracking can result in the truck tracking inboard onto the
shoulder or up over inside curbs.

This performance attribute is affected primarily by the distance from the tractor kingpin to the
center of the trailer rear axle, or the wheelbase of the semitrailer. In the case of multitrailer
combinations, the effective wheelbase(s) of all the trailers in the combination, along with the
tracking characteristics of the converter dollies, dictate this property. In general, longer
wheelbases worsen low-speed offtracking. However, other factors including the use of tandem or
tridem axles, the kingpin offset from the center of the supporting axle group, the cross slope of the
roadway, the loads of the axles, and the truck suspension have small, generally negligible, effects
on low-speed offtracking. Figure VI-7 illustrates low-speed offtracking in a 90-degree turn for a
tractor-semitrailer combination.

The standard double-trailer combination (two 28-foot trailers) and triple combinations (three
28-foot trailers) exhibit better low speed offtracking performance when compared to a standard
tractor and 53-foot semitrailer combination. This is because they have more articulation points in
the vehicle combination, and use trailers with shorter wheelbases.

HIGH-SPEED OFFTRACKING

High-speed offtracking, on the other hand, is a dynamic, speed-dependent phenomenon. It results

from the tendency of the rear of the truck to move outward due to the lateral acceleration of the
vehicle as it makes a turn at higher speeds. High-speed offtracking is actually the algebraic
combination of the low-speed offtracking toward the inside of the turn and the
outward displacement due to the lateral acceleration. As the speed of the truck increases, the total
offtracking decreases until, at some particular speed, the rear trailer axles follow exactly
the tractor steering axle. At still higher speeds, the rear trailer axles will track outside of the
tractor steering axle. The speed-dependent component of offtracking is primarily a function of the
spacing between truck axles, the speed of the truck, and the radius of the turn; it is also dependent
on the loads carried by the truck axles and the truck suspension characteristics. Figure VI-8
illustrates offtracking maneuver for a standard tractor-semitrailer.

VI-33
 Figure VI-7
Low-speed Offtracking

Source: Roaduser Research

OFFTRACKING ON MAINLINE HORIZONTAL CURVE AND INTERCHANGE RAMPS

An analysis of offtracking and swept path width for horizontal curves designed in accordance with
AASHTO's high-speed design criteria (1994) was completed for the vehicle configurations
considered in this study. Such curves are typically found on mainline roadways and higher speed
ramps. Alternative design criteria that permit higher unbalanced lateral acceleration and, thus,
tighter radii can be used under AASHTO policies for horizontal curves with design speeds of
40 mph or less, which are typically found on ramps and turning roadways at intersections.

Under AASHTO policy (1994), the minimum radius for a horizontal curve varies with the
roadway design speed and the maximum superelevation rate. For horizontal curves with a

VI-34
 Figure VI-8
High-Speed Offtracking

Source: Roaduser Research

maximum superelevation rate of 0.06 feet/foot (the maximum superelevation rate most commonly
used by State highway agencies), the minimum radii permitted by the AASHTO high-speed design
criteria vary with design speed, as shown in Table VI-10.

Table VI-10
AASHTO High-speed Design Criteria

Design Speed Minimum Radius (Feet)

(Mph)

30 273

50 849

70 2,083

VI-35
The AASHTO policy for horizontal curve design specifies pavement widening on sharp radius
horizontal curves for which truck offtracking is a concern. For the minimum-radius curves listed
above on a highway with a lane width of 12 feet on tangent sections, only the 273-foot radius curve
(for a 30-mph design speed) would require widening. The AASHTO criteria call for such a curve
to be widened from 12 to 14.5 feet.

An analysis was conducted to determine whether minimum-radius curves with the widths
described above, designed in accordance with AASHTO policies, would be capable of
accommodating each of the vehicle configurations considered in this Study. This analysis
was conducted by comparing the lane or ramp width to the swept path width of the truck making a
turn with the specified radius. Tables VI-11 and VI-12 present this comparison for selected truck
configurations.

The swept path widths in Table VI-11 are based on fully-developed offtracking determined with
the Glauz and Harwood Model for a truck traversing the curve with a travel speed equal to the
roadway design speed. None of the swept path widths shown in Table VI-11 exceed the
corresponding lane width for mainline roadways or the corresponding ramp widths, although
the turnpike double with 53-foot trailers does require nearly all of the (widened) 14.5 feet of the
30-mph AASHTO horizontal curve. Thus, there is no indication that any of the Study vehicles,
traveling at the roadway design speed, would necessarily offtrack into an adjacent lane or shoulder
of the roadway or ramps designed in accordance with AASHTO policies.

Table VI-11
Swept Path Width for Selected Trucks on Horizontal Curves
At AASHTO Design Speed Criteria

Maximum Swept Path Width (Feet) at the Design Speed

on the Sharpest Horizontal Curve Allowed by AASHTO
Design Policy

Design Speed (Mph) 30 40 60

Curve Radius (Feet) 273 509 1,348

Truck Configuration Length

(Feet)

3-Axle Single Unit Truck 39.5 8.12 8.00 8.00

5-Axle Tractor Semitrailer 64.3 10.09 8.56 8.50

5-Axle Tractor Semitrailer 76.8 11.88 9.43 8.50

6-Axle Tractor Semitrailer 76.8 11.79 9.48 8.50

7-Axle Truck-Full Trailer 61.3 8.44 8.00 8.00

7-Axle Rocky Mtn Double 99.3 11.62 9.21 8.50

8-Axle B-Train Double 84.3 10.39 8.70 8.50

9-Axle Turnpike Double 124.3 14.29 10.54 8.50

7-Axle Triple 109.0 9.69 8.50 8.50

VI-36
Table VI-12 presents comparable results when the trucks travel at very slow speeds on these same
curves, such as they may be required to do in congested traffic. The swept path widths at low
speed in Table VI-12 are generally greater than those in Table VI-11, but except for the turnpike
doubles, none of the study vehicles would encroach on adjacent lanes or shoulders. Both turnpike
doubles would encroach on adjacent lanes or shoulders on 30-mph design speed horizontal curves;
the turnpike double with 53-foot trailers would offtrack at low speeds into adjacent lanes or
shoulders on 40-mile per hour design speed horizontal curves and on 30-mile per hour design
speed ramps.

Table VI-12
Swept Path Width for Selected Trucks on Horizontal Curves
At AASHTO Design Speed Criteria

Maximum Swept Path Width (Feet) at Very Low Speed on

the Sharpest Horizontal Curve Allowed by AASHTO
Design Policy

Design Speed (Mph) 30 40 60

Curve Radius (Feet) 273 509 1,348

Truck Configuration Length

(Feet)

3-Axle Single Unit Truck 39.5 8.80 8.26 8.00

5-Axle Tractor Semitrailer 64.3 11.54 9.95 8.80

5-Axle Tractor Semitrailer 76.8 13.65 11.12 9.30

6-Axle Tractor Semitrailer 76.8 13.22 10.85 9.14

7-Axle Truck-Full Trailer 61.3 8.98 8.34 8.00

7-Axle RMD 99.3 13.65 11.15 9.35

8-Axle B-Train Double 84.3 11.92 10.16 8.89

9-Axle Turnpike Double 124.3 16.69 12.83 10.05

7-Axle Triple 109.0 12.15 10.40 9.14

The analyses assume that the turn is made at the intersection of two 2-lane or two 4-lane streets
and that the truck making the turn positions itself as far to the left as possible on the approach to the
intersection without encroaching on the opposing lanes, and completes the turn as far to the left as
possible without encroaching on the opposing lanes. In other words, the truck does encroach on
adjacent lanes for traffic moving in the same direction (on 4-lane roads), but does not encroach on
lanes used by traffic moving in the opposing direction. The maneuver specified above requires a
turning radius for the truck tractor which is 8 feet longer than the curb return radius on a 2-lane
road and 20 feet longer than the curb return radius on a 4-lane road, if all lanes are 12 feet wide.

VI-37
Table VI-13 presents estimates of encroachment on the curb return for selected trucks for right
turns at corners with curb return radii of 30, 60, and 100 feet. The data in these exhibits are based
on the maximum value of the partially developed offtracking because, in most cases, offtracking
will not develop fully as a large truck proceeds through an intersection turning maneuver.

Table VI-13
Curb Encroachment for 90-degree Right-turn Maneuvers
At Intersection of 4-lane Roads

Encroachment on Curb Return

Length 30-Foot Curb Return 60-Foot Curb 100-Foot Curb

Truck Configuration (Feet) Radius Return Radius Return Radius

3-Axle Single Unit Truck 39.5 -9.97 -12.07 -13.37

5-Axle Tractor Semitrailer 64.3 -0.09 -4.47 -7.88

5-Axle Tractor Semitrailer 76.8 6.42 1.11 -3.49

6-Axle Tractor Semitrailer 76.8 5.34 0.16 -4.25

7-Axle Truck-Full Trailer 61.3 -8.10 -10.82 -12.54

7-Axle RMD 99.3 6.73 1.23 -3.48

8-Axle B-Train Double 84.3 1.58 -3.23 -7.02

9-Axle Turnpike Double 124.3 15.38 8.83 2.69

7-Axle Triple 109.0 1.97 -2.97 -6.87

The encroachment columns in Table VI-13 indicates the amount of encroachment on the curbline by
the rear axles of the turning truck. A negative value indicates that the truck does not encroach on
the curbline. A positive value indicates that encroachment does occur, and the magnitude of the
value indicates the maximum encroachment distance. Where a positive value is shown for the
encroachment distance, that particular truck could make the turn without encroaching on the
curbline only if it encroached on an opposing lane(s) instead.

The turn from a 4-lane street to another 4-lane street was chosen as the case of interest because
none of the trucks considered -- baseline or study vehicles -- are capable of making a short-radius
turn from one 2-lane street to another without encroaching on either the curbline or an opposing
lane, unless the curb return radius is very large (say, 100 feet), and then only by short trucks.

With a 30-foot curb return radius, many of the truck configurations will encroach on the curb
return, with a few exceptions. The single unit trucks, the tractors with a 45-foot semitrailer,
the truck-full trailers, and the Western twins can successfully negotiate these turns. The
encroachment of the 5-axle semitrailer configuration with a 45-foot trailer is very marginal,
however, as is the triple with 28-foot trailers.

VI-38
By expanding the curb return radius to 60 feet, nearly all configurations examined can negotiate the
turn without encroaching on the curb return. The exceptions that cannot successfully complete the
turn are the tractors with 57.5-foot semitrailers, the longer RMD, and (especially) the turnpike
doubles. At an even larger curb return radius of 100 feet, all but the turnpike double with 53-foot
trailers can properly negotiate the turn.

TS&W REGULATION RELATED TO ROADWAY GEOMETRY

CURRENT REGULATIONS ON OFFTRACKING

Federal law does not address offtracking-related characteristics of trucks and combinations. In
particular, it specifies no requirements for kingpin setting, kingpin setback, and rear overhang. In
nearly half of the States, regulations require a kingpin setting for semitrailers over 48 feet in
length. Although there is no one uniform standard, the most common setting is 41 feet.

REGULATORY APPROACHES

Control of offtracking can be accomplished in either of two ways. The first requires specifying the
length limit(s) of the combination units within the context of overall combination length,
restrictions on the kingpin setback, wheelbase, and effective rear overhang, as in Canadian
regulations. The second approach is a performance specification requiring that a truck be
able to turn through a given angle, at a given speed, within a defined swept path as in European
regulations. Such a regulation would require matching truck equipment with trailer equipment for
operation based on knowledge of specific system characteristics, which would require extensive
documentation and signage to implement and enforce.

VI-39