Studying portal braces

Warning: The following discussion is extremely geeky and not suitable for some audiences.

We all know the situation. You find a truss bridge that is well-preserved, except for one little detail: the plaques are missing, leaving only dangling mounting brackets to taunt you. Meanwhile, county records are spotty or non-existent, destroyed by the Great Courthouse Fire of Aught-Something.

How do you track down the history of the bridge? If you're lucky, the bridge has a distinctive feature that ties it to a particular company (like the 'X' scrollwork used by the Canton Bridge Co. above the portal). Otherwise, the design of the bridge will likely give a vague idea about the time frame, but probably not much else.

For through truss bridges, however, the arrangement of the portal bracing can yield important clues about the bridge's origin and builder.

Take, for example, the Silver Bridge in Arkansas. County records show that the bridge was built in 1929-30, which seems wrong for a pin-connected truss.

It's not immediately obvious, but one of the Silver Bridge's three spans is different from the rest. Not only is it slightly longer, but it has more elaborate portal bracing featuring a lattice on top of curved knee braces.

(Photo by AHTD)

The other two spans have a simple "A-frame" brace style.

(Photo by AHTD)

It seems clear that the two sets of spans have different origins and were built at different times. Research by Randall Houp reveals that the bridge was first built in 1912 with three spans. Flooding in 1927 washed away two of the spans, which were then replaced by new pinned through trusses in 1929 by the Lakeside Bridge Company. It's not clear why Lakeside chose to use pin-connected spans (technology which had been obsolete since at least 1920). Perhaps the spans were surplus from elsewhere? I can just imagine the bridge company salesman telling the county officials, "We've got a great clearance price on these discontinued models. Everything must go!"

The different portal braces on this bridge tell the story. Even without confirmation from the county records, it's clear that this bridge had a complex history.

With this in mind, I would like to find a way to classify and study the different styles of portal braces used on metal bridges. Hopefully, by collecting information on the more common configurations, it will be possible to date (to some extent) those bridges that don't have plaques or records of any kind. For instance, it would be extremely helpful to find the oldest examples of A-frame portals so that we could make a statement like, "A-frame portals were never built before 1892" (or whatever the case may be).

A 1904 book, A text-book on roofs and bridges, is one of the few sources I've been able to find about portal bracing.

It explains, "The corresponding end posts of the two vertical trusses are connected by overhead bracing... The function of this portal bracing, as it is called, is both to stiffen the end posts against vibrations and to assist them in transferring the wind pressure from the upper lateral system to the supports of the bridge. Its connections with the end posts are extended as low as the headroom will allow, and the form of the bracing is determined to some extent by the depth available below the portal strut..."

So, let's take a look at some of the portal configurations. To keep my sanity, I've limited my searching to just Missouri bridges. Other states may have wildly different designs, but the basic styles can be found nationwide.


This straightforward design seems to be the most common among one-lane through trusses, especially those built during the 1900s and 1910s.

The 1904 textbook describes the A-frame as "the simplest and most economical of bracing" when enough headroom is available.

Here's an example of a footbridge where the A-frame brace serves as the only connection between the upper chords:


A variation on the A-frame is to add additional members to create the shape of an 'M'. The zigzag diagonal members resemble a Warren truss. In some ways we could say that the portal bracing is a truss-within-a-truss.

(Photo by David Smothers)


For wider bridges, especially those on state highways, a simple 'W' pattern was employed.

The standard style used by the Missouri highway department featured V-laced members arranged in a Warren truss-like form. You could say that this was a truss-within-a-truss-within-a-truss.

Simple beam

Thanks to their low clearance, the 'W' portal braces were frequently damaged by oversized trucks. In these cases, Missouri would sometimes retrofit the bridge with a simple beam allowing for more headroom. That's what happened to this bridge on Old US 40:

Missouri continued to build through truss bridges well into the 1950s, but these later bridges used a single beam for the portal bracing.

Double-intersection Warren

The W-style portals could be taken to the next level by adding additional members. The double-intersection Warren style is probably the second most common configuration for one-lane trusses, behind the A-frame configuration.

It could have curved knee braces:

Or stright knee braces at a 45 angle:

Or knee braces built with a solid, rounded plate:

Or knee braces that are incorpoated into the lattice:

More complicated lattices

From a double-intersection lattice, it's only a short leap to reach a "triple-intersection" design, such as this Wrought Iron Bridge Co. creation:

The next step is a quadruple-intersection lattice.

At this point it doesn't really make sense to keep counting the intersections, as there really is no limit to how many members can be incorporated into a lattice. Here's a particularly elaborate portal for a rail bridge over the Missouri River:

(HAER photo)

High trusses

Before the widespread adoption of Parker and Pennsylvania trusses, it was necessary to build really tall Pratt or Whipple trusses to span great distances. These tall trusses required significant portal braces, typically featuring a stack of different lattices, beams, and X-braces.

Here's an example from 1885:

This 1886 bridge by the King Bridge Company featured a lattice, X-brace, another lattice, and fancy knee braces.

You've probably seen this portal brace before: the silhouette is at the bottom of every page on this website.


We've seen A, M, and W designs, but there's one letter of the alphabet left: the 'X'. A single X-frame is a typical design for long-span truss bridges.

One 'X' might be stacked on top of another.

Modern welded trusses often incorporate an 'X' design, but without the lower strut to allow for more headroom.

Here are two cantilevered trusses side-by-side, built at different times periods. Notice how the portal design evolved between the two.


Here are some other interesting portal designs from Missouri:

Among all U.S. bridges, the Wells Street Bridge in Fort Wayne, Indiana, has one of the most elaborate and ornate portals. The details are amazing:

(Photo by Anthony Dillon)


The study of portal braces may seem trivial -- and perhaps it is -- but this aspect of bridge design warrants more research to help fill in the gaps for those bridges that are otherwise undocumented.

Comments  (7)

Studying portal braces
Posted February 10, 2010, by James Baughn (webmaster [at] bridgehunter [dot] com)

Jason Smith submitted this study of portal braces, complete with hand-drawn illustrations. See the attached image or go here for the PDF:

Uploaded file: JPEG image data, JFIF standard 1.01, 93377 bytes

Studying portal braces
Posted February 10, 2010, by Wes KInsler (wkinsler [at] wkinsler [dot] com)

Very nice article, James. I did notice, however, that your "W" portal looks like it may have started life as an "M" style portal. Portals loosing struts and knee braces as traffic got larger and more frequent seems to be a nation-wide occurance. And it seems, more than any of the other smaller bridge details, the shape and style of the portal goes a long way to giving a bridge its "personality", and alterations to a portal all contribute to that.

Studying portal braces
Posted February 10, 2010, by Anthony Dillon (spansaver [at] hotmail [dot] com)

Being the bridge loving geek that I am, I found this to be anything but trivial. Like the Happy Pontist stated, the portal bracing does assist in carrying the compressing force from the upper chord down through the endposts to the abutments and into the ground. The portal bracing works with the struts (sway bracing on larger spans), lateral bracing, and floor beams to keep the trusses aligned with one another. This system is absolutely critical to the integrity of the span.

Decoration of the portals and portal bracing helped not only to dress up the span, but could indeed help to promote future business. Even to those "non-believers" that don't find the charm in metal bridges that is given to their covered counterparts, it is hard to dispute the beauty in a decorated span like the Wells Street Bridge.

Of course, there are those of us who can find some form of beauty in them all.

Studying portal braces
Posted February 10, 2010, by Nathan Holth (form3 [at] historicbridges [dot] org)

Forgot to comment that another design that is uncommon, yet pops up enough to be categorized which is the pedimented portal brace. Exact design varies from bridge to bridge, builder to builder. Classic example:

Studying portal braces
Posted February 10, 2010, by Robert Elder (robertelder1 [at] gmail [dot] com)

This is a very interesting post. Almost all portal bracing configurations that I have seen appear to be variations that are based on the designs listed.

There is only one major category that I might add. Some railroad bridges use a portal bracing system that appears to consist of three solid metal panels. Not to promote another bridge in my backyard, but this is the first example I could think of:

There is also a variation of this configuration, only with large holes punched through the otherwise solid panels.

Uploaded file: JPEG image data, JFIF standard 1.01, 77236 bytes

Studying portal braces
Posted February 10, 2010, by The Happy Pontist (happypontist [at] googlemail [dot] com)

Am amazing survey!

It doesn't particularly assist you, but it's worth noting that the portal is frequently doing a lot more than just minimising vibration or helping to resist wind. For any truss where the upper chord is in compression (such as any single span truss), it is also helping to prevent that chord from buckling, by helping the truss end members carry the buckling restraint forces to ground. From that perspective, the portal frame can be of any shape which is well-proportioned to prevent lateral sway, and I'd be surprised if you couldn't find some examples of K-frame bracing to add to your W, X etc examples.

Studying portal braces
Posted February 9, 2010, by Nathan Holth (form3 [at] historicbridges [dot] org)

Certainly a challenging task you have undertaken. I wish you the best of luck.

Some random thoughts:

Builder and bridge size are the two main determinants on portal design. Builder could be a bridge company or a state developing standard plans.

I could write a whole essay on state standard truss bridges in Pennsylvania with their widely varied portal bracing designs.

Agree with comments on A-Frame. It was a simple, utilitarian design resulting from the standardization and simplification of truss bridges during the late 1800s.

Portal bracing is just one part of the bridge which can be "read" to identify a builder and age for an unknown bridge. Often the more subtle details can be equally informative, such as the hanger design, use of recessed nuts, or even patented rolled sections such of Wrought Iron Bridge Company's patented rolled T-sections.

Also, you touched on a side topic which would also be a great discussion in its own right which is truss connection types. Pinned connections in the 1920s were indeed extremely rare and well past their time except for large or continuous trusses. However there are isolated incidents. I don't know why. Maybe it was due to some old engineers who could not be taught new tricks. Thomas Gilkey in Pennsylvania is a great example. He built pinned truss bridges into the 1920s. Example: