KODA read an article in the New York Times last week (“In Seattle, a Sinking Feeling About a Troubled Tunnel” http://nyti.ms/134X4Z2) about Seattle’s stranded TBM ‘Bertha’ – more specifically, the problems some locals are having with property damage as a result of subsidence around the shaft being used to rescue the giant machine.
The project is reportedly undertaking massive amounts of dewatering during construction of the shaft and it prompted memories of a project in which interesting subsidence data was observed. In this small scale scenario a 4m diameter TBM was excavating in solid sandstone about 20-30m deep. Expectations for subsidence were very small, especially those related to convergence of the tunnel – especially as minor convergence was measured in the Roadheader excavated launch cavern.
Half the Story
Regardless of that fact, a rigorous subsidence monitoring regime was implemented. Excellent far field reference points were installed and a good measurement schedule was established. Daily digital level runs connected to reference points about 100m away from the tunnel. Fortnightly checks were made to reference points even further away – and it was on one of those fortnightly checks that it was discovered that the measured/reported subsidence was in fact only about half of the total. This was caused by the subsidence of the inner reference points.
In that particular zone of tunneling a good amount of water was entering the tunnel and that water loss from the ground caused the subsidence. (I remember struggling on a few occasions to find a dry spot on the wall to mount the total station for the TBM guidance system…even water-logging a hammer drill while doing so). Evidently water was being pulled in from >100m either side of the excavation – much further than convergence related deformation would have ever caused.
Ultimately the subsidence was small and no damage was ever likely to be caused, but at the time it really highlighted how easily individuals, and teams, can make assumptions about the appropriateness of engineering measures.
We Don’t Like an Inch
Before finishing up, one comment in the article is worth discussing: “A whole block went down an inch. We don’t like an inch”. That deserves to be put into context i.e. why don’t you like an inch? An inch by itself is not going to cause property damage, but the inconsistency of that inch will – the tilt, strain and shear which may result from subsidence are the drivers of concern. Other reasons for ‘not liking’ that inch are more likely related to it being unexpected and seemingly outside the control or expectations of the project or public.
The Final Word
So, whether it’s subsidence monitoring, testing for a broken sensor, or exporting a new design to a project – sit back and think:
- What assumptions am I making?
- How do I know what is correct, and if it is not, how might that impact the project and me professionally?
- What do I expect to happen?
- How can I check, test and verify what I am saying is correct?
- What else could be happening that could explain this?
Please share and we look forward to comments.
Until next time, KODA