The tunnel vision for environmental monitoring12 April 2013
Tunnel clients, designers and contractors are paying increased attention to environmental monitoring. Urban Transport Agenda speaks to some of these parties to gauge their main concerns, and what developments are taking place to ensure a safe tunnel environment.
The original Thames Tunnel project (1825-1843) was undertaken with the first ever tunnelling shield design, and resulted in the deaths of seven men. One of several major floods injured and nearly killed Isambard Kingdom Brunel himself, purportedly as he was saving others.
In addition to the regular water inflows and marsh gas ignitions, the project, "driven from a depth of 60ft by a capacious shield", had cost £120,000, "notwithstanding difficulties and accidents", according to an 1835 report published by the Thames Tunnel Office.
The shield acted, as the report states, as a kind of horizontal coffer-dam, and protected the workers - to an extent.
"Working in the tunnel, the air quality was appalling, with the workers frequently being dragged senseless up to the surface to recover in the fresh air," explains the Brunel Museum's website. "The River Thames was no better than a sewer, and the sewage entering the tunnel gave off methane gas, which was frequently set alight by the candles that provided a dim light for the workers. There were no miner's safety lamps in the Thames Tunnel."
Today, the view is that health and safety concerns are strong enough to ensure a protected and pleasant working environment. Perhaps the occasional catastrophe from the mining sector, or the daily human cost of mining in China, is the exception.
Although the primary causes of accidents in tunnels are trips, falls and spills, and many would have safety measures focus on these, there are passive measures that can be put in place to ensure a non-toxic working environment. As evacuation is always the preferred means of worker protection, monitoring systems are key to protection in the tunnel environment. The factors frequently monitored are gas, fire, temperature, pressure, air velocity and dust.
A spokesman for monitoring-equipment manufacturer Trolex says that, typically, a project will demand a complete system from a manufacturer, rather than cherry-picking from different suppliers. He emphasises that, despite this, flexibility is important to ensure a customer does not have to make any compromises.
Although it is not usually a requirement to provide individual components of a system to a construction project, the best examples of bespoke systems are the permanent solutions. Trolex cites the National Grid's head house gas systems in the UK as an example.
"This incorporated our gas sensors, which went back to a programmable logic controller (PLC) panel that was built by Trolex, and utilised other suppliers parts. This panel also gave digital outputs to the head house main control system, which allows end-customer systems to be integrated into our system; for example, if a gas alarm goes into an alarm state, then the local PLC will also be alerted."
Detection process and reaction
With methane gas detection an example of how a typical monitoring system works when it reaches a predetermined level, the machine's power is switched off and the tunnel evacuated. Often this is achieved with a traffic light-style warning system, and the appropriate responses are taught in regular health and safety meetings.
On certain issues, some people think the industry is not going in the right direction. Dust can come from a number of sources: pneumatic breaking of concrete, shotcreting works and moving spoil, among others. Keith Bowers, engineering professional head for tunnels at London Underground, says dusty worksites are one of his main concerns.
"My perception is that we are going backwards," he says. "When I started as an engineer using dry mix, dust was a real problem. I think people may have forgotten because I am noticing contamination on worksites again.
"It's an ethical question: if people in the street would complain about it, why should it be okay on worksites? But often it is hard to place liability; for example, a plant can neglect to protect the operator, and the manufacturer can claim that it was not intended to be used in a dusty environment. You also hear the argument that a glass cabin gets too dusty to be practical. Well, I wonder what would be happening to a worker's lungs.
"I believe in a degree of autonomy for contractors when it comes to direct works, and when they are constructing something that has the certainty of dust generation designed into it, it is obviously not just their problem."
Roger Bridge, tunnelling manager at Balfour Beatty, disagrees.
"I can only comment on my own worksites, but it is important to make sure severe dust exposure doesn't happen. This is perhaps harder these days, as we have more mechanical breakers and such, but we are more environmentally aware.
"My work at the A3 Hindhead tunnel project [in the UK] is a good example of this. New, stricter requirements were brought into force post-award with EH 40/2005 and HSE regulations. The revised limits brought the maximum number of particles down from something like 0.3mg/m3 to 0.1mg/m3.
"We used dry de-dusters to deal with the particulate matter from the excavation in sandstone. It was a very clean site, though it is the invisible particles that can enter capillaries and cause lung issues. We looked at a drum cutter, but in the end went for a ripper, so we weren't breaking the rock into such fine pieces. Adopting a conveyor belt meant trucks were not kicking up dust or filling the excavation with fumes as well."
Bridge discusses the common practice of personal exposure detectors, but points out that even these and the in-tunnel dust stations require laboratory analysis for measurements to be fully understood. It seems there is no way to get an accurate, instantaneous measure of dust exposure.
Keeping the roof up
In his paper 'Innovations and improvements of technology and data management systems to monitor large-scale tunnelling works', Jon Scott, managing director of the UK-based geotechnical monitoring specialist itmsoil, describes in some detail recent advances in technology that have allowed monitoring to become a critical element of tunnelling projects.
Four main factors have driven down the cost of monitoring. The automation and miniaturisation of sensors makes them quicker to install and far less obtrusive in an environment where space is at a premium; they are also "virtually immune" to temperature and RF interference. In addition, data collection methods are more stable than before, and can be reliable over mobile and Wi-Fi through other portable devices. Software has also advanced, easing data acquisition, database processing, presentation and user access.
With sizes reducing from centimetres and inches to a matter of millimetres (and sometimes a cost-reduction factor of ten), Scott adds that the vast increase in the number of sensors sees data transmission take on a greater significance than before. While a handful of sensors can each be connected to a data logger by individual cables, the number of modern sensors would make this costly, time-consuming and hazardous.
Scott cites a recent project where cable installation and protection took three times as long as sensor system installation, only for the steelworks contractor to destroy the cables. In another project, water poured in and the electrics needed to be repaired. The benefits of modern wireless technology are obvious.
Scott points to energy harvesters. Small devices can harness vibration or thermal energy and convert it into electricity, which is small but usable. All of these advances have brought monitoring from the periphery of a project into the mainstream, Scott concludes.
Bernhard Wimmer, managing director of Austria-based tunnelling and mining equipment specialist Durstmueller, tells of a less typical type of monitoring that has just been introduced into the tunnelling arena during the Crossrail project in London, UK.
The basic idea is to provide bank-card-sized active radio frequency identification (RFID) tags to certain tunnel workers. The tags have a built-in lithium battery that sends out a signal every 1.5 seconds at 433MHz to receivers, wherever they are placed. The maximum range is approximately 100m, depending on the type of antenna invested in. The readers are linked to a computer that processes the data, which can render a real-time visual status display, and can switch components and systems on or off.
The RFID tags can keep tabs on where workers are and how many are in the underground environment, and they can also enable rig proximity detection. The placement of readers in rugged housings at key points on drilling rigs (or any plant) can create an unbroken worker detection zone around the machine. The potential uses of this technology are demonstrated by the Crossrail case.
"We have installed the very first system on a Robodrill drill rig," says Wimmer. "Our system interfaces with the drill rig and we have arranged different 'classes' of users: pilots, co-pilots and (normal) workers.
"In general, the workers should be protected against the dangers of a rock drill under operation. Depending on which class [of worker] a detected transponder is related to, the drilling will be influenced accordingly."
Wimmer adds that for example if a 'normal' worker is detected in the vicinity of the rock drill, drilling will be stopped. If a co-pilot is detected, drilling will be stopped, unless the co-pilot is in the basket and presses a special pedal situated there. The machine will continue drilling, but with a reduced speed. If the co-pilot is in the basket and closes all windows so that the rock drill cannot harm him, drilling can continue at full speed, despite his detection by the system.
"There are several combinations of 'what if' scenarios programmed for the best safety and protection of employees," explains Wimmer. "A very good combination is to add a [complete] personnel (and eventually object) tagging and tracing system. If all the employees on site are equipped with a tag, this is just the logical next step for more safety.
"It can automatically be displayed how many people there are and where they are underground. Even if people are moving inside vehicles like cars, trucks or trains," he adds.
The battery life is approximately four years, while a limitation of the system is if a miner does not wear, or has a defective, tag.
Finally, the refuge chambers of the mining industry are becoming more common in tunnelling. The latest update of BS 6164 in 2011 requires a refuge chamber in tunnelling. Industry standards mean that they are able to withstand pressures in excess of 15psi and can support eight to 24 miners for several days when disconnected from surface power.
Companies such as Trolex and its refuge chamber partner Strata Worldwide were involved, as well as others such as MineARC Systems, with the mining world's requirement for refuge chambers/safe havens. There are systems available to provide readings and data on the internal and external environments. The actual control of oxygen supply and CO2 control is already part of the integral chamber operations, and not part of a separate system.
This article was first published in Urban Transport Agenda's sister publication, Tunnels & Tunnelling International. Urban Transport Agenda would like to thank the Brunel Museum for its contribution.