On our bridge tour today, the tour guide pointed out a low slung greenish bridge and began expressing his dislike of its “ugly” appearance. This made me feel like the Cannon Street Railway Bridge deserves some love so I was determined to analyze. The visible brute strength of the thick columns and flat spans caught my attention as it is clearly designed for massive railway loads rather than car or pedestrian passengers like other London bridges.

#### Figure 1 – Cannon Street Railway Bridge from South Bank

## Structure Information

The Cannon Street Railway Bridge was built from 1863-1866 by the South Eastern Railway Company (SERC) and designed by Sir Joh Hawkshaw, a famous English Civil Engineer. The SERC was in control of the rail lines to the London Bridge Station and wanted to expand the lines over the Thames in order to reach the Northern half of London, so they initiated the design for this bridge to the Cannon Street Station [1]. In order to carry a greater capacity of trains, the bridge was widened from 1886-1893 by adding a column on each side for a new total of six (from an original of four) and more decking on top. The bridge was strengthened from 1910-1913 in order to bear heavier trains and strengthened again in 1983 to account for both greater loads and general wear and tear [2].

## Historical Significance

The simply supported beam bridge is the oldest and simplest type of bridge, and although the Cannon Street Railway Bridge doesn’t innovate in form it is unique in its massive appearance. With three spans of 45m, numerous thick columns, and 2.6m deep iron girders the bridge clearly looks capable of carrying the 10 rail tracks laid on top of it [2]. It was one of the first rail bridges over the Thames and the only one located in the Eastern part of London and as such a very important connection point in Victorian London as well as today, but it is not a model for modern railway bridges, nor the best example of a period bridge.

## Cultural Significance

The bridge remains a very important railway link over the river Thames and helped bring development to the East side of the city when it was built, as this area was mostly still low income housing with little value. As it is almost entirely used for rail traffic and is not a very beautiful bridge it has never become a cultural symbol or tourist attraction. In fact many citizens of London actively criticize its appearance, and our tour guide today related the story of how it was originally going to be named after Queen Victoria, but she disliked its appearance and had them drop that name. In August 1989 two boats collided right next to the bridge in a tragic accident that caused 51 deaths and led to new lifeboat stations at the sides of the bridge [3]. After expansion and two strengthenings over the last 150 years, the bridge is still used today to transport massive amounts of commuters and tourists around London.

**Figure 2: Cannon Street Bridge Looking Eastward to Tower Bridge**

## Structural Art

The three principles of structural art are: Economy, Efficiency, and Elegance. This bridge is most certainly not elegant as Billington would define it. Although the load paths are as simple as possible and clearly visible (bridge girders to columns to ground), the structure could never be seen as “light” or “graceful”. To me the bridge does not fulfill the category of elegance, even though I personally like the stolid appearance and appreciate the unique appearance of the thick repeated columns. When it comes to being efficient, it is difficult to create a beam bridge with massive railway loads that uses a minimum of material when using iron girders and concrete foundations. An arched bridge could have reduced the material usage and added elegance, so I don’t believe that the Cannon Street Railway Bridge fulfills the principle of efficiency either. For economy however, the materials (concrete and iron) were cheap and easily procurable as well as being inexpensive to build with at the time. The wrought iron girders in the bridge deck are all the same size which made production cheaper, these factors together allow the bridge to fulfill the economic principle of structural art. Overall, by hitting only one of the three principles, the Cannon Street Railway Bridge is certainly not structural art, but that doesn’t mean it is not a useful and functional bridge, and I personally like it a great deal.

(The following paragraph written after doing Structural Analysis, skip ahead then come back so it makes sense…)

After calculating just how massive the loads that each beam and column can take, I’ve reconsidered the efficiency of the bridge. This bridge is so redundant and overengineered that I can no longer think of it as efficient. The only possibility is that its so strong to deal with scour, but even that seems unlikely to me. This bridge is definitely NOT structural art!

## Structural Analysis

The major construction materials used are iron in the girders (wrought) and the exterior of the columns (cast), and concrete in both the columns and the beam right above lateral beam right above them. There is also masonry and brickwork as the bottom foundation, and construction was began by digging small caissons to the bedrock under the Thames and building these foundations. The original four concrete columns were then poured to about the water level, where the fluted cast iron exteriors were placed and then filled with concrete. Originally the bridge did not have the tapered concrete rings at the top of the columns, these, along with one column on each side (new total of six) were added in later retrofittings. The long wrought iron girders were then transported to the site (most likely by barge) and lifted by crane onto the supports, then riveted together. Finally decking, railings, and rail lines were added to the top to complete the bridge. The bridge has three major spans of 45m and two approaches of 39.6m, the plate girders are 2.6m deep and thick I-beam shaped. The concrete columns, assuming same river depth as at London Bridge, are about 13m tall, and from visual comparisons with the known 2.6m depth of the girders, are around 3m in diameter.

**Figure 3: Looking Up at 18 Iron Girders**

The Cannon Street Railway Bridge is a simply supported beam bridge without any fancy arches or trusses, all strength and stiffening are directly from the girders in the deck. Load from self weight and live load from trains is transferred directly from the beams into the huge columns. There are no wind airfoils on the side of the bridge, but it has a small cross section, very stiff deck, and extremely strong supports without a long span so wind load has very little effect on overall design or performance. The bridge faces scour from the Thames, but there are no angled diversion bases around the column bases as they are curved and heavy enough to withstand massive lateral forces. This bridge looks so massive that it would take gigantic live loads to cause collapse, but the three possible sources of failure are: tensile yielding in the girders due to bending in the middle of the spans, bearing stress crumbling at the contact point between deck and columns, and buckling in the columns.

To calculate the max bending stress I first found the volume of a single beam over the longest span (45m) assuming a symmetrical cross section with a flange depth of 0.1m, flange width of 0.5m, and web width of 0.2m. The cross sectional area with these measurements is 0.62m^2, and the total volume per central span per beam is 27.9m^3. Using a density of 7.7g/cm^3 for wrought iron from the internet, the weight of a single girder is 214,830 kg and the self load de to weight is 4,774 kg/m. These are very conservative (high) estimates for the weight per beam. Next I used the formula of stress = My/I, where M is max moment, y is max distance to the edge from the centroid, and I is the moment of inertia. Y is 1.3 meters assuming a symmetric girder, and I calculated the moment of inertia to be 0.386m^4. Then to calculate maximum moment, I know that the max bending moment of a simply supported beam with uniform distributed load is w(L^2)/8. I decided that the max bending moment would be the tensile strength of wrought iron which is 160MPa, so I needed to calculate the max train load that the span could support per beam (of which there are 18). So my equation was: 160MPa = (4,774 + T)*(L^2)/8 using superposition where T is the distributed train load on 1 beam. Using this equation the max train load comes out to be almost exactly 183,000 kg/m which is ludicrously high for passenger trains, and shows the deck to be in no danger of collapse due to bending stress.

**Figure 4: Distributed Train Load Calculations and Bending Stress Diagram**

To calculate bearing stress I found the surface area of the columns, which have a diameter of approximately 4.5m at contact. This gave a bearing area of 15.9m^2, and as there are 18 beams, approximately three of these would be on each column, and I’ve already estimated the weight of a 45m section of beam to be 214,830kg. This means that each column will need to bear the self weight of the deck of 644,490kg along with additional live load from the trains. T find the max live load of the trains before crumbling I used stress = F/A where stress is the compressive strength of concrete which I found online as 25MPa. This gives the equation: 25*10^6 = (644,490 + T)/15.9 where T is now the total train weight on the tributary area (rather than the distributed train load). This gives a final answer of over 395 million Newtons which is PER COLUMN. That’s actually crazy high and basically unachievable for such a small space (the heaviest train ever was three times the necessary force, but over almost 10km) so the bridge is safe from crumbling.

**Figure 5: Load Paths all Channel into Columns**

Finally to calculate the critical buckling stress of the columns I used the equation: P= (pi^2)*E*I/(L^2). To calculate E (modulus of elasticity) I used the ACI code formula: E = 4700* sqrt(C) where C is the compressive strength of concrete (I used 25MPa just as before) and found E=23500MPa. I calculated the moment of inertia of the column using D=1.5m (ignoring the upper concrete collar) to find I=3.98m^4 and used the length of 13m as stated before. This gives another comically massive final answer of 8793.8 MegaNewtons/m^2 PER COLUMN!!!!!! No train could fit that into the 45m long tributary area, this bridge is crazy overengineered.

Honestly I have no idea how they pitched such a ridiculously strong bridge to the shareholders, I think they just asked for a permit to build any bridge from Parliament and then Hawkshaw went wild. Crazy stuff, maybe my assumptions of size of members was off, but I’m pretty sure they were close….

## Personal Response

The first time I saw the Cannon Street Railway Bridge I just walked past it because it looked so lackluster compared to the other bridges (same for London Bridge to be honest). Once we looked at it on the tour I wanted to give it some love because it seemed so neglected, but now that I have analyzed it my opinion has changed. At first I appreciated the bridge because of its strong and thick appearance, but once I realized just how wasteful it is I no longer appreciate it as much. It seems almost sacrilegious (maybe too strong of a word) to make something so bulky and inefficient when there are structural art options available.

### References

- http://www.engineering-timelines.com/scripts/engineeringItem.asp?id=665
- https://www.bristol.ac.uk/civilengineering/bridges/Pages/NotableBridges/CannonStreet.html
- http://www.bbc.co.uk/news/uk-england-london-28839099

I was really interested in this bridge after the tour because it was so bulky and seems so out of place compared to all of the other much more beautiful bridges in London. I thought that I would like it a lot more and it would have a more interesting story because it was so different. After reading your blog I realize my initial reaction in thinking that it was interesting was off base and the waste of material that went in to the construction of this bridge is really frustrating after learning about the criteria of structural art.

It’s great to see a bridge that is not the aestheically pleasing to look at it and get a different perspective on it. The structural analysis is simple to follow and easy to understand! I do think that the structure is a slight bit heavy according to today’s standards a=but it was built within a different time period. It still does have a clear and easy to follow load path so I would still consider it some form of structural art.

I would have never thought to have done a blog post on this since it is probably my least favite bridge that I have seen so far. Your diagrams were super helpful to see the load paths. It also crazy that for an old bridge like this that it is still one of the most used in london