Civil Engineering Reference
In-Depth Information
Practical experience has shown that embedded connections may fail.
After about 10 years this had occurred in the CP system on a bridge
from 1987 mentioned above (Schuten et al., 2007). In another case,
connections between a series of precast façade elements were embedded.
A significant fraction of the connections failed and had to be replaced. In
other cases, embedded connections have functioned well for more than
10 years. Apparently the quality of the isolation, both from the 'design'
of the connection and as executed on site, is important. In any case,
embedded connections have a larger-than-zero probability of failure and
consequently having as few as possible is beneficial. Similar experience
exists with connections in junction boxes. In some cases, they function
satisfactorily over many years; in other cases, leakage occurred and
replacement was necessary. Locating junction boxes in sheltered positions
near the façade (below gallery slabs) is beneficial rather than the exposed
heads of cantilever beams.
From a reliability point of view, it is better to have multiple connections
in parallel, rather than in series, where each individual connection failure
may disconnect a large part of the protected area. Ideally, multiple
connections should form a redundant network. This requires a large number
of connections and cables, which is economically disadvantageous. From
the inventory of Dutch CP systems it appeared that failure of connections
occurred in about 2% of the systems at 3 years of age, increasing to 10% of
cases aged 10 years or more.
Figures 6.6
and
6.7
illustrate 'good' and 'bad'
connection setup and positioning.
From the point of view of the service life of the CP system, avoiding
corrosion of connections by careful design and execution, based on
experience in the field, seems the best strategy. A test method for newly
made embedded connections should be conceived, e.g. based on the electrical
resistance between the anode cable and the concrete pore system.
6.5 Time-dependent failure of reference electrodes
Early papers on CP of concrete report frequent failures of embedded
reference electrodes (REs). These cases probably reflect the growing pains of
a new technology. From about 1990, several special types for concrete were
well established, both silver/silver chloride and manganese dioxide types.
However, due to their relatively high price, companies developed cheaper
cells based on carbon or activated titanium. Again early failures occurred.
Nowadays, cheap and reliable REs seem to be available. Causes of failure
are known, such as shrinkage of embedding mortar and drying out of liquid-
filled cells, causing loss of electrical contact.
From the TNO inventory, only one case out of 52 documented CP projects
reported failure of REs after about 10 years, for cells dating back to 1987
(Schuten et al., 2007). However, more failures may have actually occurred.
Cells may have been replaced as an unreported part of maintenance, or a
