In a tunnel, air flow is a safety system.
Measure it like one.
Ultrasonic wind sensors for tunnel ventilation control, at the portals, in the ducting and along the bore, with a tunnel mode built for measuring flow along a single axis.
The ventilation system acts on the number, so the number has to be right
A tunnel ventilation system does three jobs. It keeps air quality inside the bore acceptable during normal running, it holds temperature within limits, and in a fire it controls where the smoke goes. All three are driven by a measurement of air flow, and the control system acts on that measurement without questioning it.
That is what makes a failed sensor dangerous rather than inconvenient. A mechanical anemometer that has seized does not report a fault. It reports still air, and still air is a plausible reading inside a tunnel. The fans respond to a number that is no longer true.
The environment does the seizing. Tunnel air carries exhaust particulates, brake and tyre dust, salt dragged in from the road, and a constant wash of moisture, and the bore is one of the least convenient places on an asset to reach for maintenance. A sensor in a duct or on a tunnel ceiling is a lane closure, a permit and a night shift.
Every infrastructure decision starts at a measurement point
A transport corridor measures across the whole asset. Wind at the bridge deck and the tower, at the crane jib and the exposed viaduct. Road-level conditions, lineside exposure, portal pressure and the water level in the drainage below.
Bridge deck wind monitoring
A bridge closure is a regional traffic event, so the wind speed that triggers it gets scrutinised afterwards by people who were not there.
- WindObserver 90 0 to 90 m/s, 316 stainless, no moving parts
- WindObserver 75 open head holds accuracy as the flow angle increases
- WindSonic 75 0 to 75 m/s for additional deck positions
Pylon and tower top
The tower sees a different wind from the deck, and access means a rope team and a road closure.
- WindObserver 90 heated, IP66, icing tested to MIL-STD-810F
- WindMaster Pro 3-axis to 32 Hz for buffeting and vortex studies
- WindObserver 70 heated head for a second reference height
Bearings and expansion joints
Thermal movement at a joint is measured in millimetres, over decades, on a component that cannot be taken out of service to check.
- Blade position sensors non-contact inductive, no wearing parts
- Blade60 contactless linear displacement feedback
- MetConnect THP structure temperature via a four-wire Pt100 input
Tower crane jib anemometer
A crane stops at a defined wind speed. Everything downstream of that decision, the programme, the claim and the argument, refers back to one reading at the jib.
- WindSonic 75 0 to 75 m/s, 0.5 kg, full-range accuracy published
- WindObserver 90 0 to 90 m/s for exposed or coastal sites
- WindSonic 60 0 to 60 m/s for lower-rise lifting
Site compound and mobile plant
Mobile and crawler cranes move between pick points, so the wind reference has to move with them and still be defensible.
- MaxiMet GMX200 wind only, 0.5 kg, for a temporary works mast
- WindSonic 60 straightforward to mount and configure with MetSet
- MaxiMet GMX500 wind, temperature, humidity and pressure in one unit
Motorway gantry and road weather
Variable speed limits and warnings are set from a roadside measurement, and the gantry is the only place on the network you can reach without a lane closure.
- MaxiMet GMX551 wind, T, H, P, solar and rainfall in one housing
- MetConnect One Modbus and SDI-12 into an existing roadside outstation
- WindSonic 60 a compact wind reference on the gantry leg
Road surface conditions and ice prediction
Gritting decisions are made hours ahead on a forecast that is only as good as the road-level measurement feeding it.
- MetConnect THP temperature 0.2 degrees C, humidity 1.5 percent RH, Pt100 input
- MaxiMet GMX400 temperature, humidity, pressure and rainfall
- TruMet PW100 optical rainfall, no bucket to block with grit and debris
High-sided vehicle and crosswind warning
Exposed viaducts and embankments generate crosswinds that matter to a curtain-sided lorry and to nothing else on the road.
- WindObserver 75 open head, accurate across a wide flow angle
- WindSonic 75 0 to 75 m/s at an exposed roadside position
- MaxiMet GMX200 wind only, 0.5 kg, for a verge-mounted mast
Tunnel portal and ventilation
Portal wind and pressure change how a tunnel ventilates, which matters most in the incident nobody has had yet.
- WindSonic 60 0 to 60 m/s, compact enough for a portal mounting
- MetConnect THP pressure 300 to 1100 hPa, accuracy 0.5 hPa
- MaxiMet GMX300 temperature, humidity and pressure at the portal
Overhead line and catenary
Overhead line dewirement takes a route out for a day, and wind and temperature are what set the tension limits it runs to.
- WindObserver 70 heated 316 stainless head at lineside exposure
- MetConnect THP temperature and humidity for tension monitoring
- WindSonic M heated aluminium body, operating to -40 degrees C
Barriers, gates and signalling equipment
A barrier that fails to prove its position stops trains, and the sensor doing the proving cycles thousands of times a month.
- Blade position sensors non-contact inductive, unlimited-life position feedback
- Blade360 contactless rotary position for barrier arms
- Blade Dual Output a second independent channel where redundancy is required
Drainage, culverts and flood risk
Highway and rail flooding starts at a culvert nobody inspected, and level is the measurement that gives you warning rather than hindsight.
- LevelPro capacitive level sensing with no moving parts
- LevelLite level measurement for smaller chambers
- TruMet PW100 rainfall intensity to 0 to 200 mm/hr alongside level
Plant and machinery condition
Site plant fails where it is working rather than where it is serviced, and the wear shows up in the oil first.
- WearDetect oil debris sensing for ferrous wear particles
- LevelPro hydraulic and lubrication level with no moving parts
- Blade position sensors non-contact actuator and boom position feedback
Precipitation and driving conditions
Rainfall intensity drives aquaplaning risk and speed-limit setting, and a gauge that blocks stops reporting exactly when the reporting matters.
- TruMet PW100 optical, no moving parts, 0.1 mm resolution
- MaxiMet GMX603 integrated optical rainfall, 5 percent or better
- MaxiMet GMX551 rainfall alongside wind, T, H, P and solar
Structural monitoring and the long record
Structural health monitoring is a twenty-year series, and an instrument change part way through becomes a caveat on everything after it.
- WindObserver 75 no moving parts, IP66, built for unattended service
- WindMaster Pro 3-axis at 32 Hz for dynamic response work
- MetConnect THP analogue and Pt100 inputs for strain and temperature kit
Calibration and verification
When a closure or a stopped lift is costed, the wind reading behind it becomes evidence. Traceable calibration is what makes it hold.
- UKAS-accredited air velocity calibration traceable certification for anemometers
- Multi-parameter calibration six parameters, equipment from any manufacturer
- Recalibration scheduling intervals, reminders and turnaround
Flow along the bore, not wind across a field
Measuring inside a tunnel is a different problem from measuring in the open. The air is constrained to one axis, it is pushed by traffic as much as by the fans, and what the control system needs is a signed value: how fast, and which way along the bore.
That is why the sensors used here run in a tunnel mode. Instead of reporting a compass direction, the sensor reports flow along one axis with a sign, so a single value tells the control system both the speed and whether the air is moving towards the north portal or the south. On a single channel 4-20mA output, 12mA is set as zero flow, with the range below it representing flow one way and the range above it flow the other.
The practical consequence is that an ultrasonic sensor can take over from an existing propeller anemometer without the control system being rewritten. Many tunnels were commissioned with propeller units feeding exactly that signal, and the replacement can present the same signal on the same wiring.
Measurement also happens in more than one place. Portals carry the exposed cross-wind problem for high-sided vehicles approaching at speed. Ducting carries the extraction flow. The bore itself carries the flow the smoke control strategy depends on.
Three places, three different questions
At the portal, the question is cross-wind on vehicles entering and leaving, which is a two-axis measurement in open air and the same problem a bridge or an exposed approach carries.
In the ducting and along the bore, the question is flow rate and direction on one axis, mounted to a side wall or a ceiling where a conventional head would foul on the structure and where nobody wants to return.
Portal, duct and bore are separate specifications
View wind sensorsThe measurement that matters most is the one taken during an incident
In a fire, the strategy is to control where the smoke travels. In larger tunnels that means driving the flow towards zero in the zone around the incident so that smoke does not spread along the bore ahead of evacuating people.
Reaching and holding near-zero flow is a closed loop, and it depends on a sensor that is still reporting accurately at low flow rates and in hot, smoke-laden air. This is the case that should set the specification, not normal running.
Specify for the incident, not the average day
Discuss your requirementsA seized anemometer does not report a fault. It reports calm.
Every argument for solid-state measurement in a tunnel comes back to that sentence. A cup or propeller anemometer degrades quietly: bearings stiffen, the starting threshold creeps up, and the readings drift low long before anything stops. Inside a bore, nobody is watching it to notice.
An ultrasonic sensor has nothing to seize. It times sound pulses between fixed transducers, so there is no starting threshold, no bearing wear and no routine part replacement. In an asset where access is a night closure, that is a maintenance argument as much as a measurement one.
What a tunnel installation uses
Two ultrasonic ranges cover the positions a tunnel needs, with accredited calibration behind the measurement.
Wind and air flow measurement
WindSonic
2-axis ultrasonic, no moving parts
The sensor most tunnel installations are built on. Low cost enough to place at several points along an asset, with a tunnel mode for single-axis flow and a single channel 4-20mA option for replacing an existing propeller unit.
WindObserver
Precision 2-axis ultrasonic, stainless steel head
Where the duty is harder than a standard installation: a stainless steel open head that clears itself of debris, an optional de-icing system for exposed portals, and a hazardous-area variant where the tunnel carries a classified zone.
Calibration and verification
Where the measurement is taken
Ventilation and pollutant control
Holding air quality inside the bore during normal running, and driving fans to clear carbon monoxide so that maintenance crews can work in the tunnel safely.
Piston effect and extraction
Trains push air ahead of them through a confined bore. Measurement separates traffic-driven flow from the flow the ventilation system is producing.
Cross-wind on high-sided vehicles
The exposed approach either side of a tunnel is where a high-sided vehicle meets a sudden change in wind loading. That is a two-axis open-air measurement.
The output has to suit the system already in the tunnel
Tunnel control systems are long-lived and were commissioned around whatever the sensors of the day provided. A replacement that forces a change to the SCADA or the fan control logic is a much larger project than a sensor swap.
The ultrasonic ranges provide analogue and digital outputs, including the single channel 4-20mA tunnel mode that mirrors a propeller anemometer signal, so an upgrade can stay a sensor upgrade.
Match the signal, keep the control system
Read the application noteWhat the specification looks like in practice
Wind sensors have been supplied into tunnel applications worldwide for many years, in the portals, in the air ducting and on tunnel side walls and ceilings, which is where the tunnel mode came from in the first place.
Solid-state measurement removes the failure mode that matters in a confined asset: there is no bearing to stiffen and no starting threshold to creep, so a sensor that is working reports flow and a sensor that is not reports a fault rather than calm.
The sensors are built in the UK, and the measurement behind them can be verified against an independent, UKAS-accredited calibration for air velocity, which is the evidence a tunnel operator needs when the ventilation strategy has to be defended.
Questions about tunnel ventilation measurement
Tunnel mode makes the sensor report air flow along a single axis with a sign, rather than reporting a compass direction. Inside a bore the air is constrained to one axis anyway, and what the control system needs to know is how fast it is moving and which way along the tunnel.
Yes, where the existing unit feeds a single channel 4-20mA signal. That signal sets a centre point, usually 12mA, as zero flow, with the range below representing flow one way along the bore and the range above flow the other way. An ultrasonic sensor can be configured to present the same signal, so the control system sees what it has always seen.
Three positions are typical and they answer different questions. Portals measure cross-wind on vehicles entering and leaving, in open air and on two axes. Ducting measures the extraction flow. The bore itself measures the flow the ventilation and smoke control strategy acts on, usually from a side wall or ceiling mount.
Because the strategy is to control where the smoke goes. In larger tunnels the extraction fans are driven to bring flow towards zero in the zone around the incident, so that smoke does not travel along the bore ahead of people evacuating. That is a closed loop around a flow measurement, and it needs a sensor that is still accurate at low flow rates.
In road tunnels the flow is deliberately altered so that maintenance crews can work in the bore without being exposed to carbon monoxide from vehicle exhaust. The ventilation system is clearing the working area, and the measurement confirms that it has.
Because the risk sits at the portal rather than inside. A high-sided vehicle leaving a tunnel moves from still, sheltered air into an exposed approach in a very short distance, and the change in wind loading is abrupt. Portal measurement supports the warning systems and speed restrictions that manage it.
Tunnel air carries exhaust particulates, brake and tyre dust, road salt and persistent moisture. On a mechanical anemometer that combination attacks the bearings, which raises the starting threshold and biases readings low. An ultrasonic sensor has no bearings and no routine part replacement, which is why it is the usual choice where access is difficult.
WindSonic covers most installations, including multiple measuring points along one asset, and carries the tunnel mode and the 4-20mA retrofit option. WindObserver suits harder duty: a stainless steel open head for heavily contaminated air, optional de-icing for exposed portals, and an intrinsically safe variant where a classified zone is involved.
Where the ventilation strategy has to be evidenced, yes. Air velocity is one of the core parameters covered by independent, UKAS-accredited calibration, and that accreditation applies to instrumentation from any manufacturer, not only to the sensors described here.
Speak to our team
Tell us what you need to measure, where and under what conditions. We’ll help you work out the appropriate approach.
We aim to respond in one working day.