Civil Engineering Reference
In-Depth Information
the leading face; a practical distance with respect to buildability is 1 to
4 m in the top heading (as part of a larger tunnel) and 4 to 10 m in a com-
plete cross section. After construction, the invert is covered with muck and
backfill. This protects the invert against damage and provides a track for
the heavy plant while the excavation continues. Depending on the shape
of the tunnel and the space required for manoeuvring the plant, the thickness
of this cover is approximately 1 to 3 m. However, this makes the invert
invisible and it is impossible either to inspect its integrity or to install any
of the common optical monitoring systems.
These days, the state-of-the-art in monitoring the invert is by doing it
indirectly and involves interpreting the displacement measurements of the
vault (i.e. crown or crown and bench). However, the wide-spread experience
on many construction sites is that a broken invert can only be detected in
a very progressive state of damage - if at all. At least the Eggemouse can
tell if the invert is broken, which is of paramount importance. However,
it is still not known exactly how to recognize the beginning of an invert
failure by interpreting vault monitoring data. Any conclusions drawn from
interpreting the monitoring data of the vault with respect to the integrity
of the invert cannot be proven as the invert is not visible. The behaviour
of the invert is therefore open to speculation.
Based on these experiences some essential questions kept recurring:
1
Is it generally possible to get early signs of the reduced bearing capacity
of the invert by interpreting the displacement measurements of the
crown?
2
Also, is it possible to assess the residual bearing capacity of the broken
sprayed concrete lining?
In order to answer these questions a comprehensive research project was
undertaken at Hanover University, Germany (Stärk 2002). At the beginning
of the research, measured data from tunnels with broken inverts were
analysed. However, the data, even at the Eggetunnel, gave no indication of
any problem with the invert, and since the moment of the collapse of the
invert was never known, this approach was not successful. Calculations were
also not helpful, as the theoretical model could not be verified due to the
lack of measured data in the invert. The research was therefore advanced
by using model tests (Figure 8.3). Two different shapes were used: a full
cross section with a deep invert vault, and a crown section with a temporary
invert. The model tunnels were made from gypsum and embedded in clay.
The load was applied by horizontal and vertical hydraulic jacks, inde-
pendently controlled to achieve different coefficients of lateral earth pressure.
Depending on the geometry, 12 or 14 monitoring points were distributed
over the complete cross section including, of course, the invert. The location
of the monitoring points in the crown and bench corresponded to the
traditional monitoring positions in a real tunnel.
