Rivet at the 200-foot mark of the Thames Tunnel shield

The rivet at the 200-foot mark of the Thames Tunnel shield frame measured exactly one inch when the calipers were set at noon. By six in the evening, after the tide had pressed the riverbed down upon the brickwork and the heat of the men’s breathing had saturated the iron, the same rivet measured one and one-sixteenth inches. The specification for the shield demands a rigid interface between the timber poling boards and the iron staves. Any deviation in the measurement of the frame components introduces a gap. A gap of one-sixteenth of an inch allows the saturated silt of the Thames to migrate from the face into the working chamber. The failure mode is a catastrophic influx of water that drowns the miners and stalls the enterprise for months.

Measuring the same iron member twice reveals that the object is not a constant; it is a function of its environment. The first measurement represents the design state in a vacuum of stress. The second measurement represents the truth of the load. The iron expands under the thermal energy of the worksite and compresses under the hydraulic weight of the river. The discrepancy between the two readings is the precise location where the engineering model breaks. If the gap is not accounted for in the initial tolerance, the bolts will shear when the tide rises.

Redundancy must be built into the measurement itself. The surveyor must calibrate the instrument against a fixed datum on the shore, then again against the shifting masonry of the shaft, and finally against the iron itself. The load path is never static. The difference between the morning’s datum and the evening’s datum is the measure of the earth’s intent to crush the work. This variance is the most critical piece of data in the logbook. I have ordered the foremen to record the temperature alongside every linear measurement taken after sunset. The next ring of brickwork will be laid with a compensated mortar joint to absorb the movement of the shield.