Ice and Rain Protection — Page 247, Lesson 305

Let's start with the three main types of ice detector in current use. I want you to hold onto two key words: accretion and inferential. The first type is the ice detector head, which works on the accretion principle. That means it physically lets ice build up on it, and then detects that build-up. The second is the mechanical ice detector, also on the accretion principle. The third is the element ice sensing unit, which works on the inferential principle — meaning it doesn't wait for ice to build up; it infers the presence of icing from some other measurable change, like a change in electrical resistance or frequency. Let's look at the two ice detector heads. First, the Teddington Ice Detector. This is an aerofoil-shaped mast that protrudes into the airflow and is visible from the cockpit. It has two important features: a heater element, and a light to illuminate the mast at night. In flight, with the heater power supply switched off, ice accumulates on the mast, giving you a direct visual indication of ice accretion. When you want to clear it, you switch the heater on to dissipate the accumulated ice. So this is a purely visual, pilot-observed detector. Now the Smiths Ice Detector. This one is a hollow tube attached to the aircraft by one end, with holes drilled in the leading and trailing faces — four holes in the leading edge and two in the trailing edge. In normal flight, there's a pressure build-up inside the probe, sensed by a relay unit at the open base of the tube. In icing conditions, the leading-edge holes become blocked by ice, creating a negative pressure in the hollow tube, which causes the relay unit to give a warning. A heater element is fitted around the tube to dissipate accumulated ice. So here, the warning is automatic, not just visual. Now the mechanical detectors. The Rotary (Napier) Ice Detector has a serrated rotor shaft continuously driven by an electric motor. The shaft rotates adjacent to a fixed knife-edge cutter, with a clearance of less than 0.002 inches between them. The unit is mounted on the fuselage with the rotor axis at right angles to the airflow, and the cutter in the lee of the shaft — meaning the cutter is on the downstream, sheltered side. Under normal conditions, little torque is needed to drive the rotor. But in icing conditions, ice builds up on the rotor and is shaved off by the cutter. That requires greater rotational torque, which causes the motor to rotate slightly in its flexible mountings. This movement operates a microswitch, giving an ice warning or automatically initiating the anti-icing sequence. The warning remains as long as ice continues to foul the cutter blade. Finally, the Rosemount (Vibrating Rod) Ice Detector. This has a short cylindrical probe mounted on a vibrator housing, which vibrates the probe axially at about 35 kHz. If ice builds up on the probe, the added mass reduces the resonant frequencies. When the frequency falls to a predetermined level, an ice warning is given. The warning signal also operates a built-in heater element in the probe to shed the accumulated ice. After six seconds, the heater switches off and the icing cycle recommences. And here's a clever bit — the frequency of the cycle can be measured to give an indication of the ice accretion rate. So to tie it together: the Teddington is visual, the Smiths is pressure-based, the Napier is torque-based, and the Rosemount is frequency-based. Each gives you a different way to detect the same hazard.