Details of riveted railroad truss bridges

Front matter

KNAP pi

1 )i ;1aUs (|:Riyt;1;g; Railroad : ■. Truss Bridges Ji:;;i;|

OF

Introduction

The history of hridges extends as far "back as that of the ancients whose first attempts were prohalDly limited to the use of a single log or stone used to span a streani or ravine. Real "bridge construction began with the stone arch of the Assyrians, which some historians date hack as far as 4000 B. C. The EgilyTDtians as far back as 2500 B. C. built arches without mortar.

These early people used wooden beams for roofs and bridges as they knew nothing of the truss. The modern truss may be an evolution of the arch, but more likely is a type of its own.

The king post truss was probably the origin of all modem trusses. From this truss the panel system was evolved. Up to the beginning of the 19th. century, allthe early bridges were evolved from the king post truss and were built with a decided camber and a stiffening arch. Very little is now known of the stresses in such bridges, and at that time no attempt was made to determine stresses.

It was in this manner that bridge building went on blindly. If a mistake was made and a bridge fell dov/n, another was built a little stronger. It was not until the beginning of the 19th. century that bridge design become a profession, and not until the middle of the century that it become a science.

engineers were awakened to the fact that stronger bridges must be built. About 1800 Burr and Palmer built wooden truss bridges with camber. In 1820 Town introduced a lattice bridge of plank connected by pins. In 1840 William Howe patented a bridge using wooden chords and wrought iron verticals. The Pratt Truss was originated in 1844,

Up to this time all these bridges were built without any knowledge of the stresses^in them. In 1847 Squire "Whipple become interested and published a book on the determination of stresses. He also designed several bridges of the form known as the Whipple Truss. His treatise was the origin of the mathematical determination of bridge stresses, and he may justly be regarded as the Father of rational bridge design.

We can now see that the bridge of today has been developing since 1850. At present we have two types of truss bridges, pin connected and riveted. Much can be said of the relative merits of each.

Riveted trusses are coming more and more into favor on account of their great rigidity as compared with thr pin connected tj'pe. They are the more expensive of the two, but are likely to continue the favorite form of construction on account of their great rigidity and consequently low cost of maintenance.

It is the purpose of this thesis to present a discussion of the various details which go to make up a riveted span. Por this purpose blue prints and drawings from many different bridge companies, as v/ell as technical journals, have been studied. V/here possible the economical as well as the structural side will be considered.

In discussing this subject two main lines will be followed, namely; The Sections and the Connections. On account of these being inter- related they will both be discussed under one head for each detail.

The plates used give sketches to illustrate the detail. These sketches are not drawn to scale.

The following details are discussed:

1. Stringers,

2. Floor Beams,

3. End Post,

4. Top Chord,

5. Intermediate Post,

6. Bottom Chord,

7. Diagonals,

8. Portal & Sway Bracing,

9. Lateral Systems, .10. Bearings.

Art. 1. Stringers.

In all types of bridges, whether plate girder cr truss construction, the floor system is much the same. In all we find stringers, upon which the ties rest directlj'-, placed lengthwise in the bridge. The stringers are supported at the ends by floor beams. There is not much latitude in the design of stringers. Either standard I-beams or plate girders are used. The length of these stringers vary from a very short span up to 33 feet 3-3/4 inches. The latter is the lenfe,th of the stringers in the Pennsylvania Railroad bridge over the Delaware River, and they are the longest spaned stringers in this country.

The usual arrangment of stringers is that of placing two parallel at a distance of six and one half feet centre to centre. On the Boston and Maine Railroad, four stringers are used. Two large ones are placed directly under the rails and two smaller ones are placed two and one half feet outside. The outside stringers are not stressed except as the main stringer is bent or broken. This seems to be a poor design, and uncertain. A better design would be to use twosmall I-beams of the same sizfi, and placed side by side as Pig. 3. The Northern Pacific Railroad use only two stringers. They use I-beams for short spans and standard built-up sections for long spans. Pig.l. shows the section of a stringer and also a method of fastening it to the floor beam, i

S'ig.l. Fig 2.

This is a very common method for fastening the stringer to the

floor beam and is used in both truss and plate-girder bridges.

The only difference between'ihis connection and that of the I-

A,

I

"beam is that no filler has to be inserted between the flange angles of the floor beam and the connection angles of the stringers.

In case the end shear of the stringer is too great to allow enough rivets to be placed in the distance of its depth, a shelf-angle is placed under the I-beam or plate girder, and this angle takes some of the shear. This shelf -angle is also used to facilitate erection, in which case stiffener angles are not required. See Fig. 2,

j

Fig. 3. Fig. 4.

In a through truss difficulties are found on account of the great spacing of the ties at the end of the stringer. A method used to avoid this and make a continuous stringer flange is to raise the stringer and allow the top flange of the stringerto pass over the flange of the floor beam. Fig. 4 shows this ^ method. i

Ihen the stringer comes to the end of a through bridge it is often necessary to extend the stringer a foot or two past i the last last floor "beam. This is necessary in order to "bring the last tie on the bridge within about six inches of the first tie on the abutment. To accomplish this a bracket is built up and riveted to the floor beams. Fig. 5. shows this detail as used by some companies.

Pig. 5. Fig. 6.

There is an objection to this detail on account of the fact that a load applied near the end causes a tension in the rivets which fasten it to the floor beam. To overcome this fault the web of the floor beam is slotted and a splice plate is allowed to take the tension. Fig. 6 shows the method of making this detail.

In deck bridges the stringers are often placed on the floor beams and fastened to them by rivets through the lower

flange of the stringers. When this system is used, or when the ;

il

span of the stringer is very long, a system of cross bracing is used. Fig. 7 and 8 show two simple methods for bracing the two. |

In the New York, 3Jew Haven & Hartford Railroad the end floor beam is sometimes omitted and the stringer rests upon theabutment. In order to bring the stringer to the right II level a detail similar to Pig. 9 is used. This is fastened to a masonry "by means of "bolts in slots in order to allow a lengthwise movemient ,

Pig. 7. Pig. 8.

o

o

o

o

o

o o

o

o

o

o

oo

o

o

o

o

o

o

o

o

o

o

o o

o

o

o

\o

o

o

o

o o

o

o

o

o o

o

o

o

o o

o

o

o

J

L

Pig. 9.

In most types of bridges where very long stringers are used, a system of "bracing is used independent of the bottom

lateral system of the truss. The most common system is the plain Warren sz'stem shown in Pig. 10. This same system with verticals is also used in some case^^see Pig. 11.

Art. 2. Ploor Beams.

The floor beams of a "bridge are placed perpendicular to the stringers and join the two trusses of the bridge at their panel points. On account of the weight which is "brought on this mem"ber, only "built up sections can "be used; and on a,ccount of the size of these sections, some difficulty is found in fastening the floor "beam to the bridge truss.

The floor beams are used throughout the length of the length of the bridge and may be divided into two general types on account of their position, namely: end-floor beams and intermediate floor-beams.

Much difficulty is found in fastening the end floor- "beam, because of the inclination of the end post. Fig. 12 shows a simple method of fastening the end floor-heam. This particular detail is taken from a "bridge designed "by the Phoenix Bridge Company. Here the weight on the end floor-heam comes directly upon the pin and bearing.

lo o

o

o

o

o

o

o

o

o

o

o

o o

o

o

o

o

o

o

■P- -Q._o_g_ p_ p__o o o o o o o o o o o o o o o o o o o o o o o o o

The sketch shows the position of the^etail and its

relation to the post. The floor-heam consists of a simple

built-up section without any end angles. This floor-beam is

riveted into place and field riveted to the truss. One of the

angles shown is to be field riveted. This makes it necessary to

place a row of field rivets in the vertical and horizontal legs

the

of the angles. In order to accomplish this. box like arrangement is left open until the f loor-l)eam is in place and then a metal plate is holted to the box. The main objection to the detail is the field riveting that is required.

The standard detail of the end floor-beam used on the New York Central and Hudson River Railroad is shown in Pig. 13. On account of the projecting angles of the lower chord, it is necessary to make a notch in the floor-beam in order to extend the web below the chord. As considerable weight and shock comes on the end floor-beam, it is necessary to extend a gusset plate above the intersection of the bottom chord and end post to make a solid connectoin. ITo serious objection can be seen to this connection.

The American Bridge Company use a simple means of fastening the floor-beam. This consists of letting the lower flange of the floor-beam rest on the bearings and rollers. A plate is then riveted to the floor-beam and bent so as to fasten on the cover plate of the end post in order to stiffen the floor-lDeam and keep it from rotating about its lower flange. This detail is shown in Fig, 14.

ooooooooooo ooooooooooo

There is little question but that this detail is the most favora.ble of end floor-beam connections on account of its ease in putting in place and the absence of many field rivets. The appearance may not be as neat as some others but is as effective and is also more economical. Ij

A similar detail to this is where the plate is riveted to the side of the end post instead of the top. It is difficult to rivet this plate to the side on account of the flange and proximity of the floor-beam itself.

A Canadian bridge company used a detail as sho^n in Pig. 15. The upper flange and part of the web of the floor-beam is allowed to extend over the top of the end post. The end post has flanges on either side and in order to place the floor-lDearn in place a portion of each flange is removed. A plate is riveted to the end post and floor-bean to prevent a rotating a'bout the intersection of the web and end post. All the stress is transfered through rivets to the end post.

The principle advantage of this method of connection is the simple method for bracing the floor-beam where it comes above the top of the end post in order to prevent the great tendency to rotate. It is difficult however to determine how the stress will act. Part of the stress is taken up on either side by rivets; but the stress that is taken up loy the outside rivets is carried by the web from the inside of the end post. One reason for the lack of favor in this detail is the amount of shop work required to get it into position.

avoid making a jog past the lower chord, the method shown in Pig. 16 is very effective, the few field rivets required being a ver^ important point in its favor. It willalso be seen that instead of the weight being taken by a row of rivets, a bearing

Whereit is desirable to build a deep floor-beam and surface is had on the cover plate of the end post.

The Northern Pacific Standard plans show a method much the same. Pig. 17 and 18 show two details used. These differ only in the type of connection at the pedestal. In the one the lower chord extends past the pin, and in the other a separate detail is used for the pedestal.

The details shown in Fig. 16, 17 & 18 are all similar and have the same ohjections. In the first place all of them are more elaborate than common practice justifies. The designer of today should try to make a detail as simple as possihle as well as neat in appearance. It is also argued that riveting of these connections to the end post have the effect of weakening the post. Uot withstanding these facts some railroads have adopted these details as standards.

The American Bridge Company has a detail which is used in place of an end floor-beam, Fig. 19 shows one view of this detail and Fig. 20 another.

Fig. IS.

In this detail, which is in reality an end strut ,the stringer to passes over. This avoids using the stringer "bracket. ITo attempt is made to construct a rigid fastening to the end post

for little weight is taken at that point. The top flange of the strut passes over the top of the end post, and is riveted to the cover plate "by means of a bent plate. The only stress that is brought on this connection is weight due to the member itself and any twisting movement that occurs. The argi^ment in favor of this detail is the ease with which the road bed and bridge can be joined on account of the stringer extending past the end of the bridge.

One drawback to this detail is the form of "bridge seat that is required. In order to have a place for the hearing, it is necessary for the centre position of the "bridge seat to be raised some distance.

At one time the end floor-beams were used only in long spans. At present they are used almost entirely, and a number of rail* roads have adopted them as standard construction.

o

o

o

o

o d

|0

o o o o

O 0 o

o o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o

o o

o

|o

O O 0 o

o o o

o

o

o

o

o

o

o

o

o

o

Perhaps a more difficult problem is tha intermediate floor-beam connections. This is made difficult on account of the proximity of the lower chord.

In deck truss bridges the lower chord does not interfere and a detail similar to Pig. 21 will serve. This floor-beam is made up of plates and angles with stiffeners under each stringer. These stiffeners consist of four angles, two on each side placed "back to back. The end angles are field riveted to the intermediate posts. All that is necessary is to have enough space for the required number of rivets.

o

c

o

o o (3V

o

O O 0 o]

o

o

o o o \

O O O Oi

o o

o o o o

o

o

O O O Oi o

o o o o

o

o

o

O O O O 1

o

o

o o o o ol

o

o

o o 0 o 1 o o

C40 o o

OOOOQOOO OQ

OOOOOO^OQO

It is only in the thru truss that difficulty is found

in the intermediate floor-beams. Pig, 22 shows t^e details of a floor-beam used by Boston Bridge V/orks, Here, the lower chord consisted of channel with the flange turned in and the inner side of the intermediate posts and lower chord were in a plane. The splice plate acts as a stiffener and adds strength to the floor-beam. In order to give ample room for riveting^ the floor-beam to the intermediate post it is often necessary to extend a plate and pair of angles some distance above the floor-beam, A similar intermediate floor-beam used by the American Bridge Company is shown in Pig, 23 in which no extension plate Is necessary. The sketch explains itself. The only difficulty is found in the field riveting.

Where the flange on the 'bottom chord extend outward it ip necessary'- to make an offset in the end of the floor-beam similar to that shown in Pig. 24.

This floor-beam is riveted to both the bottom chord and intermediate post. Where it is possible to make a detail like this it is a matter of economy for the lower the floor-beam is the less height is required for the head room thru the bridge. Pig. 25 is a sketch of a detail used where the lower chord is a built-up box and extends out farther than the intermediate post or subvert ical. Apair of angles are riveted on the upper flange to insert a triangular gusset plate to act as a brace and give more rigidity to the truss.

part of the floor-beajaa to the side of the bottom chord as this is a difficult part of the field riveting. The detail shown

Occasionally it is not desirable to rivet the lower in Pig. 26 is used in the Port Perry Bridge over the Monongahela River. A trapizoid weh plate stiffened with angles is riveted to the ■bottom side of the lov/er chord. This construction has the effect of causing a negative moment which reduces the positive "bending moment due to the load. The same thing is true of the detail shown in Fig. 27 and 28.

looooooooo/

o oloo

O OlOO

o o 1 o o

O O lOO

0 O 'OC

X 0 0 lOO

"^^K O O 'OO

^^AsO o loo

^ L

o o o o o o o o o

o o o o

o o o o

o o o o

o o o o

o o o o

o o o o

Beside the advantage had in construction, "both of these

details permit of close inspection of all parts as well as room i|

The detail has no advantage to paint ,^ver that of Pig. 28 except in appearances. The curved

i.

angle gives a pleasing appearance iDut makes a difficult proTalem of design as well as difficult shop worlc.

These details show the general methods used in connecting the floor-lDeam to the truss. In choosing a detail the first requisite is simplicity of design, hoth from a standpoint of economy and ease of erection. An elaborate detail would require more skill to design and construct, hut the final purpose would not be served any "better than by one simpler and cheaper.

As a rule simplicity and cheapness will go together. The detail that will serve the purpose and be economical is the

detail to use. l

ii

Art. 5. End Post. j

l|

The longest compression member of ordinary truss bridge is the end post. On account of this member always taking compression the post is designed as a column. Besides the stresses brought in post by the dead ani live loads, wind loads and ^ eccentric loads add to the direct stress. The wind on the bridge and train cause an overturning effect on the bridge. The stresses due to the wind are then a direct stress and a

I

bending stress at the foot of the portal bracing. The eccentric stresses are caused by a tendency of the end post to bend on account of the load not being applied at the neutral axis of the section. All of these forces add to the stresses in the post, and thus requires the section to be enlarged.

Pig.29. shows the moot conimon detail of a cross section. In this detail two channels are used with a cover plate on the upper side, and lacing on the lower to keep the channels spaced. When the cover plate alone is used the centre of gravity of section is near the plate.

In order to hring the centre of gravity nearer the xxjsdaas JOf 3t±3e centre of the two channels, narrow strips of steel are sometimes added to the lower flanges of the channel to balance the section. These strips are called flats. Vi/here the required section is less than the minimum allowed^ the cover plate is omitted and the post is laced on both its top and bottom sides. In this case the pin is placed in the centre of the web, and therefore no eccentricity results. Channels are used for depths of fifteen inches or less, and built up sections for depths of over fifteen inches.

Two common types of built up sections are shown in Pig. 30 and 31. Pig. 30 may have a flat to help balance the section, or may have lacing on both sides in case th cover plate is not required. The extra angles in Pig. 31 are for the purpose of balancing the section. Sometimes it is necessary to stiffen the post "by means of a plate placed between the flange angles and riveted for the length of the post. This is shown in Pig. 31

Still another section used in riveted bridges is the one shown in Pig. 32.

This consists of two angles and three plates. The design is poor on account of the centre of gravity "being far from the centre of the webs making an unsymetrical section which is to be condemned. The section is not deep but nevertheless the pin at L has to be thrown off the centre of the webs in order to eliminate eccentric stresses. On account of the upper end being riveted the fact that the pin is off centre makes a difficult problem in exactly computing stresses. The lov/er part of the post can not be held rigid unless lacing is used. A heavy section sometimes used is shown in Pig. 33. This section is not in common use.

The method of lacing might be taken up here. There are two general kinds of lacing, namely; single and double lacing. Since the function of lacing is to keep the two sides of the post apart and cause equality of stresses in the two sides of the member the double lacing is to be prefered except in light sections.

The form of lacing shown in Pig. 34 is the most common. . In this the post is held solid and the stresses equally distributed "between the two channels.

There are two common forms of single lacing. These are shown in Pig. 25 and 26. The first requires more lacing and unless the lattice bars are of such a thickness to make a riyet to long, this form is prefered to the second. The form of Pig. 36 requires almost twice as many rivets as Pig. 35 and is no more effective. In fact the appearance is not as desirable.

Por medium length bridges, posts with sections such as are shown in Pig. 29 and 30 are the best and the lacing of Pig, 35 is prefered.

Two details of heavy end post sections are shown in Pig. 37. These post sections are only economical in very large bridges, on account of the difficulty found in riveting the inside members.

There are three general methods of fastening the end post to the lower chord, vis., where the end post is placed out side of the lower chord and is riveted directly to it; or wiiere the lower chord is outside and the end post is riveted on the inside; or both are riveted to a gusset plate.

The first two methods are object ionahle on account of the additional details required at -the top of the pedestal. The latter detail is a stronger and simpler detail and makes a more rigid connection.

A method where the gusset plate is used is shown in Pig,

38. The bottom chord and end post are the same width and are

connected by means of two plates riveted inside of the members.

This detail is neat and economical. One feature to be noted is

plates

the extension of the two^below the lower chord. This extension provides a connection to the pedestal, thus making the e.ad post and lower chord fasten directly and rigidly to the pedestal. The end post is field riveted to the plate. This detail is good.

Another riveted plate connection is shown in Pig. 39. This detail was designed for the Vera Cruz and Pacific Railway in Mexico. There may have "been some object in view for this design "but it seems to be a poor one. One objection is the position of the pin. In place of lacing in this bridge plates 2-1/2" by '7/I6" were riveted at intervals along the lower side of the end post. This design is simple but the appearance is anything but pleasing.

The Phoenix Bridge Company have designed a detail for the connection of the end post to bottom chord. Pig. 40 shows this. The construction of this detail is simple enough. The two plates which hold the end post and the top chord together also support the end floor-beam. A serious objection to the design is found in the fact that the pin is not in a line with the centre of gravity of the end post. As far as the construction is concerned this detail is good and makes a satisfactory connection.

Pig. 41 illustrates another neat detail employing a gusset plate. This detail is good in that the pin is in line with the centre of end post and bottom chord, and no bending moment can result in the connection. This detail is neat and evolves no difficulty of erection or design. The only draw back to this is the high pedestal that is required.

Pig. 42 shows a detail by a Montreal Bridge Company and was designed for the Canadian Pacific Railroad. The peculiarity in this design is the pedestal which consists of a disk. A plate is riveted to the bottom of the connection and a diaphragm riveted in the position shown by the dotted angles. The design of the connection is good as far as the end post and bottom chord is concerned.

Pig. 43 shows a common detail used by the American Bridge Company. This approaches the direct riveted connection spoken of in the beginning of this article. This detail is very good in that the end post and bottom chord are rigidly held together by a gusset plate which is riveted to both. This is also good in that the gusset plate and end post extend far enough below the lower chord to make a clear connection for the pedestal .

A similar detail to this is used as a northern Pacific Railroad standard. This is shown in Fig. 44. The only difference iDetween this and the former detail is that insteg,d of a pedestal connection "being made with the gusset plate a separate pedestal connection is riveted and "bolted to the intersection of the two. This has an advantage over Pig. 43 in appearance if in nothing else. It is also a little simpler in construction hut perhaps a little more expensive.

In all the details of the riveted "bridges studied, there seemed to "be an almost universal use of the gusset-plate connection in preference to the direct connection. Hot a single direct-connection detail was found. The pro"bable reason for the

plate connection is first, that more rivet area is required

second

than could otherwise "be obtained, and^that the gusset-plate connection confomed more nearly to the rest of the design.

In all of the plate connections, the general opinion of bridge companies and designers seems to be in favor of the pedestal being connected below the intersections of the centre of gravities of U-j^Lq and I^q-'^I'

In the details shown, those illustrated by Pig. 43 and 44 are the most economical and neat in appearance. Pig. 41 might be excepted, but outside of these three details all the rest are perhaps to be ranked as possibilities of constructive skill and workmanship. These three might well be chosen as standard details, fulfilling as they do most, if not all, of the requirements in an economical manner.

; 1