Corrosion state of connections

67

gusset-68

plate ranged from 9.3mm to 11.58mm, corresponding to thickness loss ratios from 3.5% to 22.5%. On the flanges of the diagonal members, the inner surfaces were heavily corroded.

The remaining flange thickness of the diagonal members is also given in Table 3-12; it ranged from 6.4mm to 8.6mm, equivalent to a thickness loss ratio ranging from 4.4% to 28.9%.

Table 3-12 Remaining thickness of upper connections

○¹ ○² ○³ ○⁴ ○⁵ ○⁶ ○⁷ ○⁸ ○⁹ ○¹⁰

P51u

Upstream side

Initial thickness, t0 12 12 12 12 12 12 10 12 10 12

Outer surface corrosion depth, drave 0.02 1.33 0.65 0.34 1.12 0.29 0.14 1.02 0 0 Inner surface corrosion depth, dlave 2.1 0 3.9 1.6 0 2.1 0 0 2.9 3.3 Remaining thickness, tRave 9.88 10.67 7.45 10.06 10.88 9.61 9.86 10.98 7.1 8.7 Thickness loss rate (%) 17.7 11.1 38 16.2 9.3 19.9 1.4 8.5 29 27.5

Downstream side

Initial thickness, t0 12 12 12 12 12 12 12 10 12 10

Outer surface corrosion depth, drave 0.21 1.6 0.83 0.47 1.6 0.3 1.68 1.3 0 0 Inner surface corrosion depth, dlave 1.2 0 3.1 2 0 1.9 0 0 0.5 1.6 Remaining thickness, tRave 10.59 10.4 8.07 9.53 10.4 9.8 10.32 8.7 11.5 8.4 Thickness loss rate (%) 11.8 13.3 32.7 20.6 13.3 18.3 14 13 4.2 16

P25d

Upstream side

Initial thickness, t0 12 12 12 12 12 12 10 12 10 12

Outer surface corrosion depth, d_rave 0,54 1,48 1,03 0,92 1,26 0,67 0.94 0,43 0 0 Inner surface corrosion depth, dlave 2,65 0 1,64 1,23 0 0,4 0 0 0,81 2,12 Remaining thickness, t_Rave 8,81 10,52 9,33 9,85 10,74 10,93 9.06 11,57 9,19 9,88 Thickness loss rate (%) 26,6 12,3 22,3 18 10,5 8,9 9.4 3,6 8,1 17,7

Downstream side

Initial thickness, t₀ 12 12 12 12 12 12 12 10 12 10

Outer surface corrosion depth, d_rave 0,32 0,94 0,47 0,13 0,8 0,23 0,64 0,43 0 0 Inner surface corrosion depth, d_lave 1,55 0 1,8 1,95 0 0,77 0 0 1,08 0,76 Remaining thickness, t_Rave 10,13 11,06 9,56 9,92 11,2 11 11,36 9,57 10,92 9,24 Thickness loss rate (%) 15,6 7,83 18,9 17,3 6,7 8,3 5,3 4.,3 9 7,6

P73d

Upstream side

Initial thickness, t₀ 12 12 12 12 12 12 9 9 9 9

Outer surface corrosion depth, d_rave 0.7 2.7 0.6 1.1 0.9 0.9 2.6 1.6 0 0 Inner surface corrosion depth, dlave 0.6 0 0.2 0.83 0 2.1 0 0 0.9 0.5 Remaining thickness, tRave 10.7 9.3 11.2 10.07 11.1 9 6.4 7.4 8.1 8.5 Thickness loss rate (%) 10.8 22.5 6.7 16.1 7.5 25 28.9 17.8 10 5.6

Downstream side

Initial thickness, t0 12 12 12 12 12 12 9 9 9 9

Outer surface corrosion depth, drave 0.24 0.42 0.04 0.6 0.54 0.6 0.4 1.1 0 0 Inner surface corrosion depth, dlave 1.5 0 0.5 0.96 0 1.7 0 0 2.6 0.5 Remaining thickness, tRave 10.26 11.58 11.46 10.44 11.46 9.7 8.6 7.9 6.4 8.5 Thickness loss rate (%) 14.5 3.5 4.5 13 4.5 19.2 4.4 12.2 28.9 5.6

At the rivet region of the upper connections, corrosion was concentrated around the rivet heads in the shape of a doughnut at a radius of about 20-30mm from the centre of the rivet head.

69 (2) The lower connections

In the case of the lower connections (P72d and P72u), corrosion was localized. In the case of P72d, localized corrosion occurred on the surface of the gussets on the upstream and downstream sides and on the flanges of the diagonal members on both inner and outer surfaces. On the outer surfaces, corrosion was concentrated mostly on the tensile diagonal member side of the gusset-plate and on the flange of the compressive diagonal member on the upstream side.

Table 3-13 Remaining thickness of lower connections

○¹ ○² ○³ ○⁴ ○⁵ ○⁶ ○⁷ ○⁸ ○⁹ ○¹⁰

P72d

Upstream side

Initial thickness, t₀ 10 10 10 10 10 10 9 9 9 9

Outer surface corrosion depth, d_rave 0.05 1.23 1.22 0.37 1.12 1.17 0.4 2.1 0 0 Inner surface corrosion depth, dlave 0.14 0 0.29 0.3 0 0.3 0 0 0.9 0.7 Remaining thickness, t_Rave 9.81 8.77 8.49 9.33 8.88 9.53 8.6 6.9 8.1 8.3 Thickness loss rate (%) 11.9 12.3 15.1 6.7 11.2 14.7 4.4 23.3 10 7.8

Downstream side

Initial thickness, t₀ 10 10 10 10 10 10 9 9 9 9

Outer surface corrosion depth, drave 0.05 0.3 0.05 0.22 0.63 0.6 0.4 0.2 0 0 Inner surface corrosion depth, d_lave 0.07 0 0.19 0.07 0 0.4 0 0 0.1 0.5 Remaining thickness, tRave 9.88 9.7 9.76 9.71 9.37 9 8.6 8.8 8.9 8.5 Thickness loss rate (%) 1.2 3 2.4 2.9 6.3 10 4.4 2.2 1.1 5.6

P72u

Upstream side

Initial thickness, t0 9 10 9

Outer surface corrosion depth, d_rave 0.21 0 0.5 Inner surface corrosion depth, dlave 0 0 0 Remaining thickness, t_Rave 8.79 10 8.5 Thickness loss rate (%) 2.3 0 5.6

Downstream side

Initial thickness, t₀ 9 10 9

Outer surface corrosion depth, drave 1.8 0 0.43 Inner surface corrosion depth, d_lave 0 0 0 Remaining thickness, tRave 7.2 10 8.57

Thickness loss rate (%) 20 0 4.8

On the inner surfaces of the gussets, there was severe localized corrosion at the edges of the gusset-plates and in the transition region between the gusset-plate and the diagonal member flanges on the both upstream and downstream sides. In the case of P72u, the corrosion state was unobservable over the whole of the gusset-plate outer surfaces, on both upstream and downstream sides, because of previous strengthening of the original gusset-plates. The strengthening consisted of additional gussets fixed in place with connecting bolts.

The only visible corrosion was concentrated on the flanges of the reinforced diagonal members on both the upstream and downstream sides and on the flanges of the non-reinforced diagonal members on the upstream side. The bolts connecting the original gusset-plates and the ones attached for strengthening prevented plaster specimens being taken, so the inner

70

surfaces were not measured. The corrosion state and corrosion depth of the strengthening plates have been neglected.

As can be seen clearly in Figure 3-56, the gusset-plates of the lower connections were also divided into several parts for measurement (○¹  ○¹⁰ for P72d and ○¹  ○³ for P72u). The remaining thickness is shown in Table 3-13. Connection P72d exhibited less thickness loss to both gusset-plates and diagonal member flanges than the upper connections, with the loss varying from 1.2% to 12.3%. In the case of P72u, there was severe corrosion on the flanges of the reinforced diagonal member, especially on the downstream side, with the average corrosion depth of 1.8mm representing a 20% thickness loss, while on the upstream side the thickness loss was 5.6% or equivalent to 0.5mm of average corrosion depth. On the flanges of the non-reinforced diagonal members, the thickness loss was 4.8% for the downstream side and 2.3% for the upstream side.

(3) Comparison between the upper and the lower gusset-plate connections

This steel truss bridge was constructed mostly with box and I-shaped elements so, in a severe environment with salt spray and airborne salt, it was readily affected by water and sediment pooling, especially on the inner surfaces of gusset-plate connections. Further, in this design of steel truss bridge, the connections to the lower chord members are below the deck surface, making them susceptible to sedimentation and water, particularly on their outer surfaces.

The gusset-plates on the upper connections are more severely affected by corrosion than those on the lower connections, as described in previous sections (1) and (2), respectively.

The location of the bridge near the embouchure, where it is exposed to salinity, sediment accumulation and salty spray, suggests that strong winds intensified the salty airborne environment that plays an important role in corrosion. The accumulation of salt from spray on the inner surfaces of the gusset-plates of the upper connections, where it could not be washed away by rain, led to generally high levels of corrosion, such as at connection P51u and P25d.

Further, outer surfaces on the upstream side of connection P72d were prone to salt accumulation, since the floor beam was perpendicular to the wind direction, leading to severe corrosion on the left side of the gusset-plates.

In order to clarify the relationships between corrosion depths as measured in this work and the actual deposition of airborne salt, it will be necessary in future work to carry out an analysis of airborne salt, wind speed and wind direction.

71

Figure 3-56 Divided parts of connections