Part Two
Classification, Reinforcement and Repair, and Posting
A highly mobile Army will make use of existing bridges. Before using a bridge for military traffic, engineers must first determine if the bridge can safely support the loadings. For this purpose, the Army uses the MLC system. Several methods exist for determining a bridge's MLC, each with different degrees of complexity and accuracy. These methods are discussed in Chapter 3.
Often, existing bridges might be in need of repair due to years of maintenance neglect or war damage. Other bridges might be in good shape physically, but their load capacities (classifications) might be too low to support the military vehicles that must use them. Chapter 4 provides guidance for the repair of existing bridges and methods to upgrade the load capacity of typical bridges.
Once a bridge has been classified using the "analytical method" discussed in Chapter 3, signs must be posted on the bridge to inform users of its classification. Standard methods for posting bridges are provided in Chapter 5.
This chapter implements STANAG 2021 (Edition 5) 
Chapter 3
Classification
A classification number is assigned to a given bridge to represent how much military vehicular traffic the bridge can carry. Width and height restrictions also affect a bridge's classification. A bridge might have a dual classification (such as for wheeled and tracked vehicles) when its capacity is greater than Class 50. The classification procedures presented in this chapter are based on guidance found in STANAG 2021. Other criteria (not covered by the STANAG) have been adopted from the AASHTO's Standard Specifications for Highway Bridges. Follow the same procedures in this chapter when analyzing foreign bridges, but be aware of and allow for any differences in material properties and design criteria.
SECTION I  BRIDGECLASSIFICATION CONSIDERATIONS 
31. The two primary methods of classifying bridges are the expedient and the analytical. The analytical method is the most accurate and desirable. Between these two extremes are many combinations, some of which are discussed in this chapter. The classification method used will depend on the situation and the available time and information. A complete analytical classification might be needed if using a bridge is very important. Also, only the results from the analytical method can be posted and reported as a permanent load classification for a bridge.
32. Bridge reconnaissance is necessary for classification. Even if bridge plans are available, conduct an onsite inspection to ensure the accuracy of the bridge data. If key bridges are behind enemy lines, perform reconnaissance by longrange patrols or aerial photography. Local civilians are a valuable source of information and should not be overlooked during onsite reconnaissance. See FM 5170 for more information on conducting a bridge reconnaissance.
33. Engineer units are responsible for gathering and maintaining bridgeclassification data for bridges in their areas. These units report this information to higher engineer echelons for consolidation and distribution. Asbuilt plans are included as part of the bridgeclassification data.
34. Local highway authorities are the most important source of bridgeclassification information. In most jurisdictions, these authorities maintain asbuilt plans and information on material properties of the bridges. Local, state, and country officials in the US and most foreign countries often impose maximum load limits or permissible stresses on their bridges. It is very important that military units consult local officials when determining the maximum MLC of a bridge intended for use during peacetime maneuvers. If they do not, the provisions of STANAG 2021 and this chapter will govern bridge classification.
SECTION II  EXPEDIENT CLASSIFICATION 
35. In some situations, an engineer's estimate based on an expedient method might be sufficient. The classifications resulting from an expedient method will be less accurate as well as conservative. The expedient method chosen will depend on the amount of bridge data and time available for conducting a classification.
HASTY METHOD
36. A hasty classification is the most expedient but inaccurate classification method. It is based on how many and what types of civilian vehicles cross a bridge regularly. This information can be obtained through reconnaissance, from local officials, or from observation of the type of route that the bridge is on (for example, rural road or secondary or major highway).
37. Once the type of civilian vehicles that use the bridge have been identified, it can be assumed that military vehicles of similar weight and axle configurations could also cross the bridge safely. In many cases, this method will severely limit the allowable military vehicles since many of them are much heavier than typical civilian vehicles. A more indepth classification method will likely reveal that the bridge has a higher MLC than the hasty method indicates. Use the hasty method only when a rapid crossing is required. Do not post the resulting classification.
TRACK COMMANDER'S BRIDGECROSSING BOOKLET
38. Graphic Training Aid (GTA) 5712 contains information for armor personnel to use in determining if a bridge is capable of carrying their vehicles. This method is very conservative; use it only during highmobility operations. Do not post this rating.
SECTION III  CORRELATIONCURVE CLASSIFICATION 
39. A direct correlation between known civilian design loads and an equivalent MLC can be made by equating the respective design criteria and the vehicle's load effects. The bridge's original design load can be determined from the following sources (listed in decreasing order of reliability):
 Civilian bridgeinspection reports.
 Original design drawings or calculations.
 Federal or local highway authority bridgedesign standards.
 Route requirements.
 Reconnaissance on known civilian vehicle usage (similar to the hasty classification).
310. Besides knowing the original civilian design load, the bridge's span length must be known. Use the design load and the span length to determine a temporary MLC using the correlation curves shown in Figures 31 through 37.
Figure 31. Correlation Curves for the United States
Figure 32. Correlation Classification for Civilian Bridges (British)
Figure 33. Correlation Classification for Civilian Bridges (Czechoslovakian)
Figure 34. Correlation Classification for Civilian Bridges (Danish)
Figure 35. Correlation Classification for Civilian Bridges (German)
Figure 36. Correlation Classification for Civilian Bridges (Russian)
Figure 37. Correlation Classification for Civilian Bridges (Norwegian)
CORRELATIONCURVE APPLICATIONS
311. Only temporary MLCs are allowed with the correlation method. However, the method is soundly based on bridgedesign theory and, when used properly, can result in a competent rating. Use these curves only on bridges that were designed using the appropriate design loadings and appropriate design criteria. Consequently, do not use the correlationcurve method on bridges in the back country or in thirdworld locations because careful design practices were probably not followed.
BRIDGE CONDITION
312. The correlationcurve method assumes that a bridge is in good condition (no significant deterioration). Because original bridge designs were probably conservative, allow for a small degree of deterioration before lowering the bridge's MLC. However, if the bridge appears significantly deteriorated, reduce the MLC accordingly. There are no firm guidelines to use when downgrading the MLC based on deterioration. However, a simple rule of thumb is as follows: if a member appears to have lost "X" percentage of its original cross section, reduce the computed MLC by the same percentage.
SPAN LENGTH
313. The correlation curves were developed for simply supported spans; however, other types of bridges can be classified by using the adjusted or equivalent span length. The adjusted or equivalent span length takes into consideration the positive moment on continuous or cantilevered spans. For truss, arch, and suspension bridges, apply the special considerations discussed below to determine the equivalent span length.
314. The correlationcurve method can be used for all bridge types. However, for truss and suspension bridges, use the procedure very carefully. If not careful and if using the wrong span length, the results could be too liberal for classification purposes. For each of these bridge types, determine the two span lengths as follows. After determining the truss span length and the panel length, use the lowest MLC rating.
 First span length. Follow the guidelines in paragraphs 315 through 317 (using the total span length) for all other types of bridges to determine the first span length. For singlespan trusses, use the actual span length. For continuousspan trusses, use the equivalent span length (paragraph 342). For cantilever truss spans, use the span length of the suspended span. For steelarch and suspension bridges, use the single longest span length.
 Second span length. Follow the guidelines in paragraphs 315 through 317 (using a span length equal to the stringers composing the floor system). These spans will be much shorter than those from the first span length. If stringer measurements are unavailable, use the following aids:
 For trusses, floor stringers are equal in length to the trusspanel lengths, which is the distance between intersections of truss diagonals. If panel lengths vary, use the longest.
 For steel arches, floor stringers are equal in length to the horizontal distance between the vertical supports running from the main arch to the bridge deck.
 For suspension bridges, floor stringers are equal in length to the horizontal distance between hanger cables, which are the vertical cables suspended between the main suspension cable and the bridge deck.
CORRELATIONCURVE USES
315. Once the span length has been determined, use correlation curves to determine the MLC. The process differs with US and foreign bridges.
UNITED STATES BRIDGES
316. Figure 31 shows the correlation curves for bridges in the US. The four standard classes of highway loading (as defined by AASHTO) used in the US are H15, H20, HS15, and HS20. The correlation curve that is used will depend on the year in which the bridge was designed and the route it is on. For newer or reconditioned bridges on major US highways, it will most likely be HS20. If unable to determine the original design load for older US bridges, the H15 curve can usually be used.
317. Locate the appropriate span length and the AASHTO loading on the graph, and read left to get the liveload moment. If the bridge normally carries twoway civilian traffic but a oneway MLC is needed (this will be higher than a twoway MLC and is used for caution crossings), multiply the liveload moment by the appropriate adjustment factor (K) from Table 31. Do not use the adjustment factor for a twoway MLC. Determine the liveload moment from Figure 31. Using this value of liveload moment along with the span length, determine the wheeled and tracked MLC from Table B2. Adjust the MLC downward to account for width and height restrictions (see paragraph 344 for more information).
FOREIGN BRIDGES
318. Figures 32 through 37 show the correlation curves for other countries. Use the particular country's chart, locate the span length, and move up to intersect the appropriate bridge category. Then read the MLC from the scale on the left. These curves give a twoway MLC. For a oneway MLC, multiply the resulting MLC by the appropriate adjustment factor from Table 31.
SECTION IV  ANALYTICAL BRIDGE CLASSIFICATION 
319. The analytical classification method is basically the reverse of the design method. In the design method, engineers establish the desired MLC and then determine the required size and quantity of bridge components to meet that MLC. In the analytical method, the bridge already exists and engineers must determine the composition, the dimensions, and the type of construction to obtain a permanent bridge classification. The analytical classification is based on classical methods of engineering analysis, but no rigid rules apply to the techniques used.
320. The difficulty of analytical bridge classification varies with the bridge type. The degree of accuracy depends on the information available, the techniques used, and the amount of detail covered. Some information gathering might be too difficult or timeconsuming to be worth the effort. The engineer must make reasonable assumptions based on the information available. The classification techniques have been simplified as much as possible, consistent with good engineering practices, to permit a reasonably accurate bridge classification. Only qualified engineers should make permanent classifications.
CLASSIFICATION ASSUMPTIONS
321. The superstructure is almost always the controlling feature in bridge classification. Because the superstructure must span large distances, its elements must be made as lightweight as possible (designs must be optimized). This can be done effectively since superstructure loadings are fairly predictable. Substructures, however, must be more conservative (less efficient) in design to account for unpredictable loadings (stream and ice flow, barge impact) and unknown soil conditions. Do not check the substructure unless it appears to be significantly deteriorated or unstable due to scour or settlement, or is improperly designed or constructed.
CONTROLLING FEATURES
322. Figure 38 shows typical bridge components. The deck structure is generally stronger than its supporting superstructure; therefore, it is not considered in most classifications. The only exception to this rule is timber decking, which may be weaker than the rest of the superstructure.
Figure 38. Typical Bridge Components
323. Generally, connections are also stronger than the superstructure beams; therefore, they do not have to be considered in most classifications. The only exception is when a connection is deteriorated due to such factors as overstress or rust. In these cases, reduce the computed MLC (based on the superstructure beams) by an appropriate amount. If possible, study and document the connection carefully. Afterwards, an engineer can make a more accurate assessment of the connection's load capacity.
324. For bendingmoment calculations on beams, assume the midspan to be the controlling location. The list below gives examples of where the midspan may not be the controlling location. Despite these possibilities, using midspan moments in all cases should give reasonable classifications.
 Steel girders with cover plates, where the worst case may be at the ends of the cover plates.
 Variabledepth plate girders, where the worst case may be at changes in the cross section.
 Reinforced concrete beams, where the worst case may be at bar cutoff points throughout the beam.
 Interior supports of continuousspan structures.
325. Bridges with longitudinal stringers may have smaller exterior stringers than interior stringers. The reason is because exterior stringers, by virtue of location, do not receive as much of the vehicular loading as interior stringers. For rating purposes, assume that the interior stringers control. This is a reasonable assumption since military convoy loadings are generally concentrated toward the center of a bridge. If this case does not occur, consider the capacity of the exterior stringers separately.
BRIDGE CONDITIONS
326. Classification procedures assume that a bridge is sound. Because original bridge designs are generally conservative, allow for a small degree of deterioration before reducing the computed classification. However, if the bridge appears significantly deteriorated, reduce the MLC accordingly. If a careful onsite inspection can be conducted, account for deterioration by reducing the crosssectional dimensions of the members (or reinforcing steel) or by reducing material strengths. If an inspection is not possible, then compute the classification based on normal conditions and make conservative assumptions about the MLC to account for the deterioration.
MEASUREMENTS
327. Measure the spans of simply supported bridges from center to center of the supports. The supports may be bearing plates or rollers. In a multispan bridge, measure the weakest span for classification purposes. If not sure which spans to measure, measure and classify all of them. Measure the spans to the nearest 1/2 foot, always rounding up. Prepare sketches showing all the bridge's dimensions and cross sections that were used to classify it. For moment calculations, measure the crosssection dimensions at the midspan and indicate the complete details of the main structural component. For shear calculations, measure the crosssection dimensions near the span supports. If using asbuilt plans and specifications to classify a bridge, survey the bridge to verify the drawings and check the existing bridge conditions. For more information on bridge reconnaissance, see FM 5170.
DEAD LOAD
328. The dead load is usually computed as a uniformly distributed load acting along the length of a member. Compute the dead load, based on member dimensions, using the typical weights shown in Table 32. Dead load consists of the weight of—
 The main structural members (stringers, girders, or trusses).
 The decking.
 All accessories and hardware (curbs, handrails, bracing, nails, and bolts).
329. The weight of a 1foot length of the bridge span is computed to determine the uniform dead load. To compute the dead load, determine the weight per foot of all members supporting the loads and add it to the weight of all the bridge components carried by the members in a 1foot length of the span. Equations 635 through 643 are simpler for determining the dead load. Next, compute the portion of the dead load that each member is carrying as follows:


LIVE LOAD
330. Vehicle loads are assumed to be the only live load acting on a bridge. The standard NATO vehicles in Appendix B are the vehicles that should be used to rate bridges. Assume the standard NATO convoy spacing of 100 feet between vehicles when rating. Because of this large spacing, usually only one vehicle will be on any single span of the bridge at a time. If significant pedestrian traffic is expected (refugees and dismounted military units), treat these as line loads of 75 pounds per foot, each over a 1foot width. Place these line loads where the lines of people might be expected.
IMPACT LOAD
331. Increase the live loads by 15 percent to account for impact. Use this factor for all bridge types and span lengths except for timberstringer or floating bridges. The impact on these bridges is zero.
LOAD DISTRIBUTION
332. Each structural component of a bridge shares, to varying degrees, in carrying the applied live loads. This loadsharing concept is accounted for in the bridge rating procedure by the number of effective stringers. The values in Table 33 are for longitudinal stringers and depend on the deck stiffness and stringer type and spacing. The specific use of these factors for different bridge types is discussed below. Distribution equations for other members (such as girders and trusses) are provided in their respective sections.
ALLOWABLE STRESSES
333. A load classification will only be as good as the definition of the material properties (for example, yield point or modulus of elasticity). Determine these properties as accurately as possible. Obtain accurate data from property testing (nondestructive testing, concrete cores, steel coupons), an original design, or asbuilt drawings/records. Once obtained, reduce the maximum material properties by specific amounts to obtain the safe usable portion of the property, referred to as the allowable stress. Allowable stresses based on deflection, bearing stress, and fatigue are not considered in this manual. The following list tells where to find allowable stresses and specific material properties for the different elements:
 Timber, paragraph 347.
 Steel, paragraph 363.
 Concrete, paragraph 3132.
 Prestressed concrete, paragraph 3164.
MOMENT CLASSIFICATION FOR SIMPLE SPANS
334. Compute the total moment capacity of the main structural component (stringer, girder, or floor beam) as follows:


DEAD LOAD AND DEADLOAD MOMENT OF A COMPONENT
335. Determine the portion of the dead load carried by the structural component (paragraph 328) and the deadload moment that applies to that component. See Appendix D for concentrated dead loads. For uniformly distributed dead loads, use the following equation:


LIVELOAD MOMENT OF A COMPONENT
336. Compute the liveload moment of a component as follows:


NUMBER OF EFFECTIVE COMPONENTS
337. Determine the number of effective components that support the live loads for one and twolane traffic. Each bridgeclassification procedure addresses the number of effective bridge components.
TOTAL LIVELOAD MOMENT PER LANE
338. Compute the total liveload moment per lane as follows: :


MOMENT CLASSIFICATION FOR VEHICLES
339. Determine the moment classification for wheeled and tracked vehicles and for one and twolane traffic. To do this, use the hypothetical vehiclemoment tables or curves in Table B2 or Figures B1 or B2.
MOMENT CLASSIFICATION FOR CONTINUOUS SPANS
340. Continuousspan bridges are often used for multispan bridge structures (Figure 39). This type of bridge results in a savings of material or in longer spans for the same amount of material when compared to a simply supported bridge. The efficiency of a continuousspan bridge lies in the principle that the total moment imposed on a bridge is distributed between positive moment at the midspan and negative moment over the continuous supports. Due to this redistribution of moment, a smaller bridge member (stringer, girder, truss) will be required to resist the moment caused by the applied loads. Figure 310 compares a simply supported and a continuousspan bridge.
Figure 39. Typical ContinuousSpan Bridge
Figure 310. Comparison Between a SimpleSpan and ContinuousSpan Bridge
EQUIVALENT SPAN LENGTH
341. Continuousspan beams are always indeterminate. Therefore, accurate bending moments within the spans can only be determined by rigorous indeterminate analysis (such as moment distribution or matrix methods). Such an analysis is not practical in most situations; therefore, an expedient, yet reliable, method is needed. Ordinary continuousspan bridges can be rated approximately, using the concept of an equivalent simple span. A simple span is often thought of as the distance between the liveload inflection points on a continuousspan bridge. Actually, the equivalent span length is the length of a simple span that would receive the same maximum liveload moment that would be produced on a continuous span by the same loading.
342. The equivalent simplespan length is 0.80 times the length of the end span or 0.70 times the length of the interior span. If a bridge has all equal span lengths, analyze the exterior span. If a bridge has various span lengths, analyze the span that results in the longest equivalent span length. The applicability of these factors decreases for bridges with spans greater than 90 feet or with large differences in the cross section or the span length. In these cases, consider using a more indepth indeterminate analysis.
FINAL CLASSIFICATION
343. After selecting the controlling span and its equivalent span length, classify the bridge the same as a simply supported span. Note that for this situation, use the equivalent span length for calculating deadload moment. Obtain the liveload moment from the tables and curves in Appendix B.
ROADWAYWIDTH CLASSIFICATION AND CLEARANCE RESTRICTIONS
344. Minimum roadwaywidth restrictions shown in Table 34 are based on NATO militaryvehicle classifications (see STANAG 2021). If a bridge with a specific classification meets these width requirements, all standard military vehicles bearing the same or lower classification may cross the bridge. If a onelane bridge meets all the requirements except the minimum width, post a width restriction without downgrading the classification. Twolane bridges must meet the requirements in Table 34. If necessary, downgrade the twolane classification for width requirements. The desirable minimum overhead clearance for bridges is 14 feet 9 inches. Post a clearancerestriction sign and a telltale for bridges not meeting overheadclearance requirements. Posting requirements are discussed in Chapter 5.
SOLIDSAWN AND GLUELAMINATED TIMBERSTRINGER BRIDGES
345. Timberstringer bridges (Figures 311 and 312) are very common because they are often more economic and expedient than steel or concrete bridges. The timber may be native or dimensioned or gluelaminated sections. The spans are usually simply supported and rarely exceed 20 feet. The decks are either plank or laminated timber. Some civilian bridges may have asphalt wearing surfaces, which will significantly affect the dead load on the bridge.
Figure 311. TimberStringer Bridge
Figure 312. TimberStringer Bridge Components
346. The kinds and qualities of timber vary greatly, depending on such factors as geographic location, age, load history, defects, and moisture content. If the species and grade of the timber are known (asbuilt or design drawings), use the allowable stresses from Table C1 or the design drawings (if provided). In most cases for military loadings, these values may be increased by a factor of 1.33 to account for lower traffic volume (shorter cumulative load duration, which is significant for timber). Do not apply this factor for nonengineered bridges. Apply other modification factors to the allowable stress to account for such variables as lumber thickness/width ratios, edgewise or flatwise use, repetitive member use, and moisture content (see Table C1 notes). In most cases, timber decking and stringers will retain moisture on their horizontal surfaces and should be considered in wetservice conditions as shown in Table C1 .
347. If the species and grade of the timber cannot be determined, use Table C2 to get allowable moments and shears or compute them using assumed values. For solidsawn timber, assume the allowable bending stress to be 1.75 ksi and the allowable horizontalshear stress to be 0.095 ksi. For gluelaminated timber, assume the allowable bending stress to be 2.66 ksi and the allowable horizontalshear stress will be 0.200 ksi. Do not apply the increase factor of 1.33 (as discussed above) to these assumed values. However, apply the adjustment factors listed from the Table C1 notes.
TIMBER DEFECTS
348. Use extreme care when classifying timberstringer bridges. Stringers and bents are subject to rot and insect attack, especially in areas where they come into contact with the ground. In tropical or wet areas, fungus or other biological growth may weaken timber stringers considerably. Adjust the crosssectional dimensions of the member to allow for this type of damage.
STRINGER MOMENTCLASSIFICATION PROCEDURE
349. Determine the moment classification of the stringers (paragraph 334) using the allowable bending stress (paragraph 346) and the value for the number of effective stringers from Table 33. For one and twoway traffic, compute the moment classification twice, using the appropriate values for each way.
STRINGER SHEARCLASSIFICATION PROCDURE
350. Timber is relatively weak in horizontal shear. Always check the shear capacity. Determine the allowable shear stress as discussed in paragraphs 346 and 347.
Shear Capacity per Stringer
351. Compute the shear capacity per stringer as follows:


Applied DeadLoad Shear per Stringer
352. Compute the deadload shear per stringer as follows:


LiveLoad Shear Capacity per Stringer
353. Compute the liveload shear capacity per stringer as follows:

Total LiveLoad Shear for One or Two Lanes
354. For wheeled and tracked vehicles on solidsawn timber bridges and for wheeled vehicles on gluelaminated bridges, use the following equation:

355. For tracked vehicles on gluelaminated stringer bridges, use the equations below:
Shear Determination
356. Using the value for the allowable vehicle shear (equations 39, 310, or 311), refer to Table B3 or Figures B3 or B4, and find the MLC that produces a shear less than or equal to this value. If considering both one and twoway traffic, compute twice, using the appropriate values for each way.
DECK CLASSIFICATION
357. Timber decks are either plank (wide dimension, laid horizontal with no interconnection between the planks) or laminated (wide dimension, laid vertical with the boards nailed or glued together). If timber decks control the load rating, check them.
Plank Decking
358. Use Figure 313 to determine the deck classification. Read the deck thickness on the vertical axis and the stringer spacing on the horizontal axis. Interpret the values between the curves, if necessary. For multilayer plank decking, subtract 2 inches from the total deck thickness. Then use Figure 313 to find the deck classification the same as for singlelayer decks.
Figure 313. TimberDeck Classification
Laminated Plank
359. Laminate the deck material to interconnect the adjacent boards in shear and allow them to share in the applied loadings more effectively. As a result, the applied wheel loads are effectively spread out to more deck boards than with a conventional plank deck. This increases the deck's rating over that of a plank deck with the same thickness.
360. For loadrating purposes, lamination indirectly has the effect of shortening the deck span between stringers by about 25 percent. Therefore, if the deck is effectively laminated (well nailed or glued over the full length of the boards), multiply the actual stringer spacing by 0.75, and use Figure 313 to find the deck classification.
WIDTH AND FINAL CLASSIFICATIONS
361. Check the width restrictions (paragraph 344 and Table 34). The lowest of the moment, shear, deck, or twolane width classifications is the bridge's final classification.
STEELSTRINGER BRIDGES
362. Steel stringers may consist of standard rolled shapes (more common in spans that are less than 100 feet) or of builtup sections from welded, bolted, or riveted steel plates (which are used for longer spans). Figure 314 shows a steelstringer bridge. Rolled and builtup stringers may have cover plates in areas of maximum positive and negative moment (Figure 315). These plates increase the beam's section modulus and thus increase the allowable load. Because of the highly varied dimensions and details associated with steelstringer bridges, accurate analytical classification of steelstringer bridges is difficult without complete design or asbuilt data.
Figure 314. SteelStringer Bridge
Figure 315. Typical Steel Stringers
YIELD AND ALLOWABLE STRESSES
363. Allowable stresses are always given as percentages of the yield strength of steel. Consider the following:
 If the bridge's design or asbuilt drawings are available, the yield strength should be listed.
 If the yield strength is unknown, use the recommended values in Table 35.
 If the steel type or the date the bridge was built is not known, use 30 ksi.
Once the yield strength is determined, determine the allowable stresses using Table 36.
STRINGER MOMENTCLASSIFICATION PROCEDURE
364. Compute the moment classification of the stringers (paragraph 334). Use the allowable bending stress (paragraph 363) and the appropriate value for the number of effective stringers from Table 33. If considering both oneway and twoway traffic, compute twice, using the appropriate values for each way. To classify builtup stringers, use the dimensions of the beam at the center of the span (the point of maximum positive moment).
DECK CLASSIFICATION
365. Decks distribute the live load to the stringers. Decks do not contribute to the moment capacity of the steel stringers unless composite construction is used. Reinforced concrete and steelgrid decks are used in civilian construction. Both decks are seldom critical in bridge classification. Classify timber decks as outlined in paragraph 357.
WIDTH AND FINAL CLASSIFICATIONS
366. Check the width restrictions (paragraph 344 and Table 34). The lowest of the moment, deck, or twolane width classifications is the bridge's final classification.
COMPOSITESTRINGER BRIDGES
367. Compositestringer bridges are difficult to recognize or distinguish from noncomposite steelstringer bridges. If in doubt, classify the bridge as a noncomposite steelstringer bridge, which results in a more conservative classification. Appendix F contains classification examples.
COMPOSITE BEAM ACTION
368. When structurally connected, the concrete deck and steel stringer form a composite beam. Figure 316 shows a typical composite beam using a standard rolled beam and cover plate. The top flange takes maximum advantage of the compressive strength of the concrete deck. The bottom flange and cover plate are efficiently used in tension. Shear connectors or studs connect the concrete to the steel stringer and resist the horizontal shear forces between the concrete and the top flange.
Figure 316. CompositeBeam Section
SUPPORTED AND UNSUPPORTED CONSTRUCTION
369. During construction, steel beams are placed on the supports with cranes. The concretedeck formwork is then constructed on top of these beams and the concrete deck is poured. During the deck placement, the steel beams may or may not have beam shoring along their length (supported or unsupported). If the beams are shored (supported) until the concrete of the deck cures, the resulting composite beam will be effective for the entire dead load of both the beam and slab, as well as live loads. If the beams are unshored (unsupported) during construction, then the steel beam by itself must support its own dead load, and the composite beam section will only be effective for the dead load of the deck and live loads. Normally, the cost of shoring is not practical when compared with the small increase in material costs required for unsupported construction. Unless the method of construction is definitely known, assume that unsupported construction methods were used. The analytical procedure shown below makes this assumption.
MOMENTCLASSIFICATION PROCEDURE
370. The total moment that the composite beam resists, assuming unsupported construction, has two parts. They are the deadload moment (that the steel beam alone resists) and the liveload moment (that the composite beam resists).
DeadLoad Moment
371. Determine the deadload moment per stringer (paragraph 335). Note that the stringer must carry its own weight plus the weight of a portion of the concrete deck.
Stringer Section Modulus
372. Determine Ssteel, which is the section modulus for the steel stringer by itself. Refer to Table D2.
Effective ConcreteFlange Width
373. The effective width of the concrete flange is the lesser value of the following:
 Onefourth the span length, in inches (reduce the span length by 20 percent for continuous spans).
 Twelve times the concreteflange thickness, in inches.
 The centertocenter stringer spacing, in inches.
Equivalent SteelFlange Width
374. The concreteflange width is represented as an equivalent steelflange width for sectionmodulus calculations. Compute the equivalent steelflange width as follows:


Section Modulus of Composite Section
375. Compute Scomposite, which is the section modulus of the entire composite beam (including the concrete deck) with respect to the bottom of the stringer. Use the methods described in paragraph D14 to compute the section modulus or use the value from Table D8.
Stress Due to Dead Load
376. Compute the stress induced in the steel by the deadload moment as follows:


LiveLoad Moment per Stringer
377. Compute the liveload moment per stringer as follows:


FinalMoment Determination
378. Use the procedure shown in paragraph 334 to compute the remainder of the moment classification.
WIDTH AND FINAL CLASSIFICATIONS
379. Check the width restrictions (paragraph 344 and Table 34). The lowest of the moment, deck, or twolane width classifications is the bridge's final classification.
STEELGIRDER BRIDGES
380. A steelgirder bridge consists of two main flexural members (girders) that are built from steel plates. In older bridges, the members are built up with riveted plates and angles and a floor beam supports the stringers and the deck. In some cases, stringers are omitted and the floor beams alone transmit deck loads to the girders. For loadclassification purposes, check the capacities of the deck, the stringers, the floor beams, and the girders. The limiting values will determine the load classification. Figure 317 shows a girder bridge. Figure 318 shows a through and a deckgirder bridge. Figure 319 shows the main components of a girder bridge. Appendix F contains classification examples.
Figure 317. Girder Bridge
Figure 318. Girder Bridges
Figure 319. GirderBridge Components
YIELD AND ALLOWABLE STRESSES
381. Determine the yield and allowable stresses as shown in Table 36. This applies to all girderbridge components discussed in the paragraphs below.
GIRDER CLASSIFICATION
382. Using the values below, determine the girder moment classification. If the girders are of composite construction, use the procedure in paragraph 370. Compute the yield and the allowable stresses from Table 36. Because there are no other checks required for steel girders (for example, shear), the resulting moment classification will be the only one.
Effective Number of Girders for OneLane Traffic
383. Compute the maximum effective number of girders for onelane traffic as follows:


Effective Number of Girders for TwoLane Traffic
384. Figure 320 shows assumed loading conditions for normal twolane bridges. Compute the number of girders for twolane traffic using equations 316 and 317 .




Figure 320. Assumed Loading Conditions for Maximum Moment in Floor Beams
385. For bridges with more than two lanes, the value of the minimum spacing between vehicles in adjacent lanes will generally be too conservative. In these cases, an engineer should determine the spacing requirement based on the—
 Actual curbtocurb width.
 Expected travel lanes for the convoys.
 Presence of median strips and convoy speed.
 Degree of traffic control.
Total Moment Capacity per Girder
386. Compute the total moment capacity of one girder as follows:


Dead Load per Girder
387. Use equations 319 through 321 to compute the various components of the uniform dead load per girder and use equation 322 to compute the total dead load per girder:








DeadLoad Moment per Girder
388. Compute the deadload moment per girder as follows:


LiveLoad Moment per Girder
389. Compute the liveload moment per girder as follows:


Total LiveLoad Moment per Lane
390. Compute the total liveload moment per lane as follows:


Moment Classification
391. Determine the moment classification for both wheeled and tracked vehicles and for one and twolane traffic. Use the hypothetical vehiclemoment tables or curves in Table B2 or Figures B1 or B2.
STRINGER MOMENT CLASSIFICATION
392. The stringers of a girder bridge act as a subspan that span the gap between the floor beams. The stringers are assumed to be simply supported with a span length equal to the centertocenter floorbeam spacing. Determine the stringer moment classification using the procedure found in paragraph 364 for noncomposite construction and paragraph 370 for composite construction. Compute the uniform dead load per stringer as follows:


STRINGER SHEAR CLASSIFICATION
393. Shear will seldom be critical if the stringers sit on top of the floor beams. However, if the stringers are connected to the floor beams by bolts or rivets, check for shear as follows:
 Shear capacity per stringer.
 Deadload shear per stringer.
 Liveload shear capacity per stringer.
 Total liveload shear, one and two lanes.
 Shear determination. Use Table B3 or Figures B3 or B4 for classification.
FLOORBEAM MOMENT CLASSIFICATION
394. Floor beams run perpendicular to the line of traffic (Figure 319). Do not confuse these beams with stringers. The loading transmitted to a floor beam is a function of the span length (girder spacing), the floorbeam spacing, the dead load, and the weight and dimensions of the vehicles.
Total Moment Capacity of the Floor Beam
395. Compute the total moment capacity of the floor beam as follows:


Dead Load and DeadLoad Moment
396. The floor beam spans transversely across the bridge and supports the longitudinally spanning stringers and the deck above. Thus, in addition to its own selfweight, the floor beam must also support that part of the dead weight of the deck and stringers. Thus, compute the uniform dead load on the floor beam as follows:


and—


LiveLoad Moment
397. Compute the liveload moment per component as follows:


Maximum Allowable FloorBeam Reactions
398. Compute the maximum allowable floorbeam reactions as follows (also see Figure 320):
Moment Classification
399. To determine the moment classification, refer to the floorbeam reaction Figures 321 through 324. Find the vehicle MLC (wheeled or tracked) which produces a maximum allowable floorbeam reaction which is equal to or less than the computed values from equations 335 and 336.
Figure 321. Maximum Wheel Line Reactions for Wheeled Vehicles (W4W30)
Figure 322. Maximum Wheel Line Reactions for Wheeled Vehicles (W40W150)
Figure 323. Maximum Wheel Line Reactions for Tracked Vehicles (T4T40)
Figure 324. Maximum Wheel Line Reactions for Tracked Vehicles (T50T150)
FLOORBEAM SHEAR CLASSIFICATION
3100. Shear will seldom be critical if the floor beams bear directly on the supporting girders (not connected via bolts or rivets). However, if the floor beams are connected to the girders by bolts or rivets, check for shear as described below.
Shear Capacity per Floor Beam
3101. Compute the shear capacity per floor beam as follows:


DeadLoad Shear per Floor Beam
3102. Compute the deadload shear per floor beam as follows:


LiveLoad Shear Capacity per Floor Beam
3103. Compute the liveload shear capacity per floor beam as follows:


Maximum Allowable FloorBeam Reactions
3104. Compute the maximum allowable floorbeam reactions for oneway and twoway traffic as follows:
Special Allowance for Caution Crossing
3105. If the load rating for shear needs to be higher than the rating that was computed from equations 340 and 341, compute a special cautioncrossing allowance using the equations below. These equations will provide the highest possible rating for floorbeam shear. However, for these equations to be effective, the convoys must be carefully monitored on the bridge. The drivers must drive as close to the center of their respective lanes as possible. If they cannot do this, do not use these equations.
Shear Classification
3106. Using the floorbeam reaction curves in Figures 321 through 324, find the MLC (wheeled or tracked) which produces a maximum allowable floorbeam reaction that is equal to or less than the computed values from equations 340 and 341. For caution situations, use equations 342 and 343.
DECK, WIDTH, AND FINAL CLASSIFICATIONS
3107. Consider the following when determining deck, width, and final classifications:
 Decks distribute the live load to the stringers. Decks do not contribute to the moment capacity of the steel stringers unless composite construction is used. Reinforced concrete and steelgrid decks are used in civilian construction and are seldom critical in bridge classification. Classify timber decks as outlined in paragraph 357.
 Paragraph 344 and Table 34 apply to width restrictions.
 The final classification is the lowest of the girder, floorbeam, stringer, deck (if checked), or twolanewidth classification.
TRUSS BRIDGES
3108. A truss (Figure 325) is a structure composed of straight members joined at their ends to form a system of triangles. It has the same function as the beams and girders and carries loads that produce bending moment in the structure as a whole. Many different types of trusses and truss combinations are used in long spans where beams and girders are not economical. These spans vary from 150 feet to over 1,000 feet. Some light truss bridges have simple spans as short as 60 feet.
3109. Bending is resisted by the top chords in compression and the bottom chords in tension. Diagonals act as a web and resist shear. The end connections can be pinned, riveted, welded, or bolted. Pinned and riveted connections appear in the older structures, while the postWorld War II structures have shopwelded and fieldbolted connections.
Figure 325. Truss Nomenclature
3110. The floor system of a truss bridge (Figure 326) has floor beams that are connected at the panel points (intersection of diagonal truss members). The floor beams support floor stringers, which span between the floor beams and carry the load the same as those of a stringer bridge. For a proper analysis, the following types of truss and span configurations must be understood:
 Pony truss. A pony truss is a halfthrough truss that does not have an overhead bracing system and is normally used on relatively short spans (Figure 327).
 Through truss. A through truss is used for longer spans and has an overhead bracing system. Traffic passes through the truss (Figure 328).
Figure 326. Truss Floor System
Figure 327. PonyTruss Bridge
Figure 328. ThroughTruss Bridge
 Deck truss. A deck truss is used for longer spans and carries the traffic on the top chord (the truss system is below the bridge deck) (Figure 329).
Figure 329. DeckTruss Bridge
TRUSS SPANS
3111. Trusses can be continuous over their interior supports (Figure 330). Refer to paragraph 342 to determine the equivalent span length. Classify a continuoustruss bridge using the end span or the longest interior span, whichever controls.
Figure 330. ConstantSection ContinuousTruss Bridge
3112. Most of the longspan truss bridges use cantilevered construction (Figure 331), which consists of two end spans and a suspended span. The end spans are anchored to the abutment by an anchor arm, and a cantilever arm projects from the pier. A suspended span is hinged to the two ends of and supported between the cantilever arms. This type of bridge should be classified using the suspended span, assuming that the suspended span is simply supported between the supporting hinges. In some cases, hinges are not used but the suspended span can be identified by the geometry of the bridge. The suspended span extends between 3/8 and 1/2 of the clear span between the piers.
Figure 331. Suspended Truss Span
EXPEDIENT CLASSIFICATION
3113. A complete reconnaissance of a truss bridge can be very timeconsuming. If time or access to the bridge is limited, a reasonable expedient classification can be achieved by considering only the floor system (stringers and floor beams [Figure 326]). The floor system of a truss bridge is the same as that of a steelgirder bridge. Refer to paragraphs 392 through 3106 to rate the stringers and floor beams. If a quicker classification is needed, use the correlationcurve method (pay close attention to the spanlength requirements of paragraphs 313 and 314).
ANALYTICAL CLASSIFICATION PROCEDURE
3114. The analytical classification of a truss bridge consists of classifying the truss on the basis of positive moment capacity of the truss and checking the capacity of the floor beam and floor stringers.
Total Dead Load
3115. Deadload computations on a truss bridge can be rather lengthy, because it has many differently sized members. If possible, determine the actual dead load of the bridge by its individual components. Do this by—
 Analyzing each panel.
 Adding all of the weights of the components for the total deadload weight.
 Dividing the total deadload weight by the length of each panel.
3116. If the above method cannot be used, use one of the following equations and Figure 332, to compute the total dead load of the bridge:
Figure 332. Truss Dimensions
Dead Load per Truss
3117. Compute the deadload weight per truss as follows:


DeadLoad Moment per Truss
3118. Compute the deadload moment per truss as follows:


Tensile Force in Bottom Chord
3119. Compute the maximum allowable tensile force in the bottom chord as follows:


Compressive Force in Top Chord
3120. Check the two different buckling coefficients as follows:



3121. Table 39 gives the allowable compressive strength in truss members (in pounds per square inch [psi]). Divide the answer by 1,000 to get the ksi for use in equation 351. Use the larger KL/r coefficient from equation 349 or 350 and compare it to the appropriate buckling coefficient (denoted by C_{c}) in Table 39. Note that the value of C_{c} depends on the steel type and the yield stress. Use the appropriate equation from Table 39 to compute the allowable compressive stress (denoted by F_{c}).

Moment Capacity per Truss
3122. Compute the total moment capacity as follows:


LiveLoad Moment
3123. Compute the liveload moment per truss as follows:


Effective Number of Trusses
3124. Compute the maximum effective number of trusses for onelane and twolane traffic as follows:
 Onelane traffic.
 Twolane traffic. First compute—
 Bridges with more than two lanes. The value of the minimum spacing between vehicles in adjacent lanes will generally be too conservative. An engineer can determine the spacing based on the actual curbtocurb width, expected travel lanes for the convoys, the presence of median strips, the convoy speed, and the degree of traffic control.
Total LiveLoad Moment per Lane
3125. Compute the total liveload moment per lane as follows:

Truss Moment Classification
3126. Determine the truss classification, based on bending moment, for both wheeled and tracked vehicles and for one and twolane traffic. Use Table B2, or Figures B1 or B2. For simply supported trusses, use the actual span length. For continuous or cantilever trusses, use the span length as discussed in paragraph 3111.
Stringers and Floor Beams
3127. Determine the classifications of the stringers and floor beams using the same procedure as that for girder bridges in paragraphs 392 through 3106. Check the bending moment and the shear capacity.
Deck Classification
3128. Decks distribute the live load to the stringers. They do not contribute to the moment capacity of the steel stringers unless composite construction is used. Reinforced concrete and steelgrid decks are used in civilian construction. Both decks are seldom critical in bridge classification. Classify timber decks as outlined in paragraph 357.
Width and Clearance Restrictions and Final Classification
3129. Check width restrictions as discussed in paragraph 344 and Table 34. Throughtruss bridges have overhead bracing and require overhead clearance consideration. The final classification is the lowest of the girder, floor beam, stringer, deck (if checked), or twolanewidth classification.
REINFORCED CONCRETE SLAB BRIDGES
3130. Concrete is very strong in compression and is a very efficient structural material. However, concrete is very weak in tension. Therefore, any concrete areas that may be subject to tensile stresses must be reinforced with steel reinforcing bars. Reinforced concrete slabs are often used for shortspan bridges (Figure 333). Because of the highly varied dimensions and reinforcing details associated with reinforced concrete bridges, accurate analytical classification (as discussed later in this chapter) will be impossible without design or asbuilt details. If these are not available, use the other methods discussed in Sections II and III of this chapter.
Figure 333. Typical Reinforced Concrete Slab Bridge
ASSUMPTIONS
3131. The analytical method (described later in this chapter) only applies to slab bridges with the main reinforcement running parallel to the direction of traffic. The slab acts as a oneway slab in the direction of traffic (Figure 334A). Assume that the area above the neutral axis acts in compression and that the reinforcing steel in the bottom of the slab carries all of the tension and the concrete carries no tension. The assumed stress distribution is shown in Figure 334B. Only the moment capacity is determined for the slab since shear generally will not control in thin, reinforcedconcrete members. Only a onefootwide strip of slab at the midspan should be considered. For continuous spans, convert the span length to an equivalent span length as outlined in paragraph 341.
Figure 334. Details for a Reinforced Concrete Slab Bridge
CONCRETE STRENGTH
3132. Try to obtain the ultimate strength of the inplace concrete from asbuilt drawings (listed on drawings as 28day strength) or from concrete core tests. If this is not possible and the concrete is in satisfactory condition, refer to Table 310. If the year the bridge was built is unknown, use 2.5 ksi for the concrete strength.
REINFORCING STEEL STRENGTH
3133. Try to obtain the yield strength of the reinforcing steel from asbuilt drawings. If this is not possible, use Table 311. If Table 311 does not show the needed yield strength, use 40 ksi for a bridge that appears relatively new and 33 ksi for a bridge that appears very old or deteriorated.
REINFORCING STEEL RATIO
3134. Compute the reinforcing steel ratio as follows:




COMPRESSIVESTRESSBLOCK DEPTH
3135. Compute the compressivestressblock depth as follows:


SLAB MOMENT CAPACITY
3136. Compute the moment capacity per foot width of slab as follows:


DEADLOAD MOMENT
3137. Assume that the total dead load of the bridge, including the roadway and the curbs, is distributed over the full width of the slab. As shown in Figure 334A, the slab width may or may not equal the roadway width. Compute the deadload moment as follows:


ALLOWABLE LIVELOAD MOMENT
3138. For normal operating conditions, compute the allowable liveload moment per foot width of slab as follows:

For emergency conditions where a higher allowable loading is required, compute the liveload moment as follows:


EFFECTIVE SLAB WIDTH
3139. Compute the effective slab width as follows:


TOTAL LIVELOAD MOMENT
3140. Compute the total liveload moment for the entire slab as follows:


MOMENT CLASSIFICATION
3141. Use the total liveload moment from equation 366 and the span length (adjusted for continuous span if necessary) with the moment values from Table B2 or Figures B1 or B2 to determine the moment classification. The total liveload moment is the same for both one and twoway traffic with this type of bridge. Therefore, twoway traffic will only be limited by the lanewidth restrictions shown in Table 33.
REINFORCED CONCRETE TBEAM BRIDGES
3142. Figure 335 shows a typical reinforced concrete Tbeam bridge. Tbeams are used to obtain longer span lengths than those allowed by slab bridges (paragraph 3130). The deck of the bridge acts integrally with and forms the top portion of the Tbeam. The vertical leg of the Tbeam (the stem) serves to position the reinforcing steel at a greater distance from the neutral axis. Because of the highly varied dimensions and reinforcing details associated with reinforced concrete bridges, accurate analytical classification (discussed below) will be impossible without design or asbuilt details. If these are not available, use the other methods discussed in Sections II and III of this chapter.
Figure 335. Reinforced Concrete TBeam Bridge
ASSUMPTIONS
3143. Figure 336 shows the assumed stress distribution of the Tbeam. An analysis should be based on a typical interior Tbeam. The exterior beams are assumed to have equal or greater capacity than the interior beams. As with the slab bridge, the Tbeam bridge is analyzed only on the basis of moment capacity, which means shear will generally not control the rating. The deck is also assumed to have sufficient thickness that it will not control the rating and is thus not rated.
Figure 336. Assumed Stress Distribution in a TBeam
CONCRETE AND REINFORCING STEEL STRENGTHS
3144. For concrete and reinforcing steel strengths, refer to paragraphs 3132 and 3133.
EFFECTIVE FLANGE WIDTH
3145. The deck width that carries the compressive stresses for an individual Tbeam is the effective flange width and is the lesser of the following:
 Onefourth the span length, in inches (do not modify the span length for continuous spans).
 Twelve times the concreteslab thickness plus the stem width, in inches.
 Centertocenter Tbeam spacing, in inches.
TENSILE FORCE IN REINFORCING STEEL
3146. Compute the tensile force in the reinforcing steel. First compute—



COMPRESSIVESTRESSBLOCK DEPTH
3147. Compute the compressivestressblock depth as follows:


MOMENT CAPACITY OF TBEAM
3148. If the compressivestressblock depth is less than or equal to the concretedeck thickness, compute the moment capacity as follows:


3149. If the compressivestressblock depth is greater than or equal to the concretedeck thickness, determine the moment capacity. First compute—






DEADLOAD MOMENT
3150. Assume the total dead load is distributed equally to each Tbeam. The deadload moment per Tbeam would be as follows:


ALLOWABLE LIVELOAD MOMENT
3151. For normal operating conditions, compute the allowable liveload moment on a single Tbeam as follows:

For emergency conditions where a higher allowable loading is required, compute the liveload moment as follows:

TOTAL ALLOWABLE LIVELOAD MOMENT
3152. Multiply the allowable liveload moment by the effective number of Tbeams for one and twolane traffic to obtain the allowable liveload moment as follows:

MOMENT CLASSIFICATION
3153. Use Table B2 or the moment curves in Figures B1 or B2 to determine the moment classification. Compare the values of the total liveload moment and the span length (or equivalent span length if the span is continuous).
WIDTH AND FINAL CLASSIFICATIONS
3154. Check the width restrictions in paragraph 344 and Table 34. The lowest of the moment or twolane width classifications is the final bridge classification.
REINFORCED CONCRETE BOXGIRDER BRIDGES
3155. Reinforced concrete boxgirder bridges (Figure 337) are used to acquire even greater span lengths than Tbeam bridges. The deck of a boxgirder bridge acts integrally with and forms the top portion of the thinwebbed Ishaped girders.
Figure 337. Typical BoxGirder Bridge
ASSUMPTIONS
3156. The boxgirder bridge is analyzed as a series of connected concrete Ibeams, with flange widths equal to the spacing between the webs of the Ibeams. As with Tbeams, the exterior beams are assumed to have an equal or greater capacity than the interior beams. The boxgirder bridge is analyzed only on the basis of moment capacity (shear will generally not control the rating). The deck is also assumed to have a sufficient thickness and a short enough span length so that it will not control the rating and is therefore not rated.
PROCEDURE
3157. Use the same procedure as outlined for concrete Tbeams in paragraph 3143 to determine the classification of concrete box girders. Use the following equation to compute the effective flange width:


PRESTRESSED CONCRETE BRIDGES
3158. Structural prestressing is placing a member in compression before it is loaded (Figure 338). This action is an improvement to conventional reinforced concrete. Normal concrete is very weak in tension and is very prone to cracking. Water penetrates the concrete and the concrete deteriorates. Prestressing prevents cracking under normal loads by placing the member's entire cross section in compression.
Figure 338. Conventional Reinforced Concrete Compared to Prestressed Concrete
RECOGNITION
3159. There are many different forms of prestressed concrete beams, and they are sometimes difficult to distinguish from conventional reinforced beams. Compared to conventional beams, prestressed beams are usually precast and much more shapely than conventional pouredinplace beams. The most common form in the short to mediumspan ranges is the standard Igirder with a castinplace composite deck slab (Figure 339). Precast, pretensioned solid or voided slabs are used for shorter spans. Many longspan boxgirder and Tbeam bridges are also prestressed. Some bridges may be of segmental posttensioned construction (another form of prestressing).
Figure 339. Prestressed Concrete Bridges
COMPOSITE CONSTRUCTION
3160. Prestressed beams are generally made composite with a concrete deck. This allows the deck to form a large part of the top flange of the beam. The roughened concrete surface and steelshear reinforcement provide resistance to horizontalshear forces between the deck and the precast beams. While it is very difficult to distinguish (visually) between composite and noncomposite construction, most prestressed construction is composite. Therefore, the analytical procedure discussed in this chapter assumes composite construction.
SUPPORTED AND UNSUPPORTED CONSTRUCTION
3161. Precast, prestressed beams are placed on their supports with cranes during construction. The concretedeck formwork is constructed on top of these beams and the concrete deck is poured. During the deck placement, the prestressed beams may or may not have beam shoring (supported or unsupported) along their length. If the beams are shored (supported) until the concrete of the deck cures, the resulting composite beam will be effective for the entire dead load of the beam and the slab, as well as live loads. If the beams are unshored (unsupported) during construction, the precast beam alone must support its own dead load and the composite beam will only be effective for the dead load of the deck and the live loads. Normally, the cost of shoring is not practical when compared with the small increase in material costs required for unsupported construction. Unless the method of construction is definitely known, assume unsupported construction. Therefore, the analytical procedure discussed in this chapter assumes unsupported construction.
LOADCLASSIFICATION METHODS
3162. The analytical loadclassification method gives the most accurate load classification. However, its use depends on complete details of the internal prestressing, which are generally not available without the original design drawings. If they are not available, use the other methods discussed in Sections II and III of this chapter.
ANALYTICAL CLASSIFICATION
3163. If the interior and exterior beams are different, base the analysis on a typical interior beam. The exterior beams are assumed to have equal or greater capacity than the interior beams. Prestressed beams are analyzed on the basis of ultimate moment capacity, since shear will generally not control the classification. Also, the deck is assumed to have sufficient thickness that it will not control the classification and is thus not considered. The assumed stress distribution for prestressed beams is shown in Figure 340.
Figure 340. Assumed Stress Distribution in a Prestressed Beam
Concrete and Steel Strengths
3164. Prestressed beams may have a combination of prestressed and conventional nonprestressed steels. Obtain conventional reinforcing steel strengths as described in paragraph 3133. Strength properties of prestressing can be obtained from the design drawings or from Table 312. Concrete strengths can be obtained from the design drawings or, if they are not known, use 4,000 psi.
Effective Flange Width
3165. For Tbeams or precast beams that are composite with the slab, the width of the deck that helps carry the compressive stresses for a beam is the effective flange width (Figure 340). The effective flange width is the lesser of the following:
 Onefourth of the span length, in inches (do not modify the span length for continuous spans).
 Twelve times the concreteslab thickness plus the stem width, in inches.
 The centertocenter beam spacing, in inches.
Steel Reinforcement Ratios
3166. Compute the steel reinforcement ratio (Figure 340). First compute—



Reinforcement Index
3167. Compute the reinforcement index as follows:

Average Stress in Prestressed Steel
3168. Compute the average stress in prestressed steel at maximum load as follows:

Maximum Tensile Force
3169. Compute the maximum tensile force developed by a beam as follows:

Area of Concrete Resisting Compression
3170. Compute the steel reinforcement ( Figure 340 ) as follows:

Area of the Concrete Flange
3171. Compute the area of the concrete flange available to resist compression as follows:


Moment Capacity
3172. Below are four equations used to compute the beam moment capacity. To choose the proper equation, compare the reinforcement index, the area of the concrete flange available to resist compression, and the area of the concrete resisting compression and then choose the equation for which all the comparisons are true.
 If R_{r} < 0.3 and A_{f} > A_{c}, then—
 If R_{r} < 0.3 and A_{f} < A_{c}, then—
 If R_{r} > 0.3 and A_{f} > A_{c}, then—
 If R_{r} > 0.3 and A_{f} < A_{c}, then—
DeadLoad Moment
3173. Assume the total dead load is distributed equally to each prestressed beam. The deadload moment per beam would be computed as follows:

Allowable LiveLoad Moment
3174. Compute the allowable liveload moment as follows:
 For normal operating conditions, compute the allowable liveload moment on a single prestressed beam as follows:
 For emergency conditions where a higher allowable loading is required, compute the liveload moment as follows:
LiveLoad Moment per Lane
3175. Compute the liveload moment per lane as follows:

Moment Classification
3176. Determine the moment classification using the values of MLL and the span length (or an equivalent span length if the span is continuous). Refer to Table B2 or Figures B1 or B2.
WIDTH AND FINAL CLASSIFICATIONS
3177. Check the width restrictions in paragraph 344 and Table 34. The lowest of the moment or the twolane width classification is the final bridge classification.
ARCH BRIDGES
3178. An arch is one of the most efficient structural shapes and one of the oldest methods of building relatively long spans. Some masonryarch bridges that were built by the Roman legions are still in use today. Modern arch bridges are constructed with reinforced concrete and steel. Figure 341 shows general types of arch bridges. The masonry arch is a form of the deck arch. Except for the masonry arch, all arches have a floor system just like a truss or girder bridge. The floor beams are connected to the arch at support points. The support points usually have vertical, columntype members that carry the floorbeam loads to the main arch members.
Figure 341. Types of Arch Bridges
ModernArch Bridge
3179. The analytical classification procedure for masonryarch bridges is presented below. A complete reconnaissance of other arch bridges can be very timeconsuming, and an exact analysis of these bridges is very tedious and timeconsuming. A reasonable classification can be achieved by only classifying the stringers and floor beams of the floor system (the same as for a girder bridge [paragraphs 392 through 3106]). If an even more expedient classification is required, use the classificationbycorrelation procedure discussed in Section III of this chapter (see paragraph 313 for spanlength requirements).
MasonryArch Bridge
3180. A masonryarch bridge is very difficult to analyze accurately. An empirical formula that is based solely on the bridge's dimensions is provided below.
Provisional Load Classification
3181. Measure the critical dimensions (arch span length and total crown thickness) of the bridge (Figure 342). Plot these dimensions on the nomograph in Figure 343 to determine the provisional load class (PLC). With a span length of 30 feet and a total crown thickness of 42 inches, use the nomograph as follows:
 Find the arch span length (denoted by L) in Column A.
 Find the total crown thickness (t + d) in Column B.
 Draw a straight line through the points in Columns A and B. The point at which the line intersects Column C is the PLC. For this example, the PLC is 96.
Figure 342. Critical Dimensions of a MasonryArch Bridge
Figure 343. Provisional Load Carrier for a MasonryArch Bridge
Military Load Classification
3182. Obtain the onelane, tracked MLC. Multiply the PLC by the appropriate bridge factors (Tables 313 through 319 and Figure 344) as follows:

Figure 344. Profile Factors for Arch Bridges
3183. Apply the E, F, and G factors with discretion. For example, if an arch is deformed and cracked because of abutment faults, do not downgrade the bridge for all three factors. In such cases, apply the E, F, and G individually to the PLC, modified by A through D. Use the lowest of the three results as the onelane MLC. If the bridge is wide enough for two lanes of traffic, multiply the onelane MLC by 0.9 to get the twolane MLC. Use Figure 345 to equate one and twolane, tracked classifications to respective wheeled classifications.
Figure 345. Bridge Classification Correlations
MOVABLE BRIDGES
3184. When a highway crosses a navigable waterway with light boat traffic, movable bridges are often constructed as a costsaving measure (Figure 346). The three general types of movable bridges are swing, bascule, and verticallift. Construction is usually a truss or girder system with machinery to move the bridge away from the navigation channel. Determine the MLC the same as for fixed bridges of the same type (girder, truss, and so on). The machinery/gearing for moving the bridge should have no effect on the MLC. Special warning signs should be posted indicating the presence of a movable bridge.
Figure 346. Typical Movable Bridges
SUSPENSION BRIDGES
3185. A suspension bridge (Figure 347) is used mainly for long spans where support from below is impracticable (for example, when a water current is too swift or when the gap to be bridged is too deep). Most spans over 2,000 feet are of suspension construction. Like truss and girder bridges, all suspension bridges have a floor system consisting of stringers and floor beams. The floor beams are connected to the suspension cables at hanger points.
Figure 347. Typical Suspension Bridge
3186. The load capacity of a suspension bridge may be based on many limiting elements, such as the support towers and the suspension cables, anchorages, and hangers. The reconnaissance and analysis effort for all of these elements would be very time consuming. A reasonable classification can be achieved by only considering the floor system (such as the stringers and floor beams that are suspended between the hangers [the vertical cables hung from the suspension cables]). The floor system of a suspension bridge is the same as that of a steelgirder bridge. Use paragraphs 392 through 3106 to rate the stringers and floor beams. If a more expedient classification is required, use the classification by correlation procedure discussed in Section III of this chapter (pay careful attention to the span length requirements of paragraph 313).
OTHER BRIDGES
3187. This section may not have covered all bridge types found in the TO (especially in foreign countries). Local civilian authorities are the best source for obtaining a reasonable MLC on these bridges; otherwise, an analysis of the superstructure will usually suffice.
NEWSLETTER

Join the GlobalSecurity.org mailing list 