bmi E145 at Norrkoping on Oct 17th 2018, captured false localizer signal, loss of separation

Last Update: October 8, 2019 / 15:13:39 GMT/Zulu time

Bookmark this article

Photo of BMI Regional G-RJXF, Embraer ERJ-145 — BMI Regional G-RJXF, Embraer ERJ-145 (Photo credit: cvtperson / Flickr / License: CC by-sa)

Incident Facts

Date of incident
Oct 17, 2018

Classification
Report

Airline
BMI Regional

Flight number
BM-1753

Departure
Munich, Germany

Destination
Norrkoping, Sweden

Aircraft Registration
G-RJXF

Aircraft Type
Embraer ERJ-145

ICAO Type Designator
E145

A bmi Regional Embraer ERJ-145, registration G-RJXF performing flight BM-1753 from Munich (Germany) to Norrkoping (Sweden) with 18 passegers and 3 crew, was self-positioning for an ILS approach to Norrkoping's runway 27. The aircraft had been cleared to descend to 2100 feet and was subsequently cleared for the approach to runway 27, the crew armed the approach mode. At the same time an ultralight aircraft was operating below the controlled airspace at 1400 feet in uncontrolled airspace performing a VFR flight, the two aircraft were on different radio frequencies. While still positioning to intercept the localizer the aircraft automatically turned west and began to descend, one pilot believed they were capturing the glideslope, the other pilot believed they were on localizer and above glide. The crew changed from automatic glideslope coupling to vertical speed and increased the rate of descent to re-capture the glideslope. Air traffic control noticed the premature turn to the west and began to develop concerns. The crew subsequently queried about the target on their TCAS screen (the ultralight) and queried the altitude of the Embraer, the crew responded they could not see the other traffic, the distance had reduced to one nautical mile horizontally and 200 feet vertically. The Embraer's TCAS issued a resolution advisory "Monitor vertical speed", which prompted the crew to initiate a go around, the crew subsequently told ATC they were now clear of traffic. The aircraft positioned for another approach and landed safely about 10 minutes later.

Sweden's Haverikommission (SHK) released their final report concluding the probable cause of the serious incident was:

The serious incident was caused by the fact that planning and followup of the approach were not carried out in an appropriate manner.

A contributing factor has been lack of knowledge of false ILS signals.

The SHK analysed:

The crew requested and received clearance from the controller for selfpositioning for an ILS approach to runway 27. However, the crew did not specify in their request which point on the final they intended to navigate towards. This meant that the controller was not sure how Midland intended to perform the navigation.

When the aircraft was 8–9 NM from the centre line of runway 27 Midland was cleared down to 2 100 feet and cleared for approach. In connection with the clearance, the pilots armed the approach mode when the track was more than 90 degrees from the approach direction, i.e. outbound.

The approach clearance probably contributed to the pilots arming the ILS system. In self-positioning to a point on the approach line, the air traffic controller is expected to monitor the flight and conclude it by issuing a turn instruction from which the aircraft can connect to a final approach and then receive the final approach clearance. An approach clearance issued at a later stage and within the coverage area of the localizer reduces the risk of the crew arming the ILS system to early and thus also the risk of the system catching false lateral signals.

When Midland was about 7 NM south of the final, the aircraft started a left turn due to a "false" localizer signal at +42° from the inbound course, which activated the ILS and gave command to the autopilot. The crew did not identify this, but instead believed that they were on the correct approach line, i.e. 7 NM further north than the aircraft's actual position.

During the approach, both pilots had the approach line presented on their navigation screens (MFDs), which enabled the pilots to notice that the aircraft began to swing at an early stage. This is also supported by the reference flight SHK carried out in the simulator (see section 1.16).

The fact that the pilots did not notice this is probably due to their attention being on the conflicting traffic displayed on the TCAS with subsequent visual scanning outside of the cockpit.

The distraction resulted in the crew's situational awareness of their correct position being reduced, as none of the pilots were paying attention to the upper part of the aircraft's MFD.

The crew was not aware of the risk that a false localizer signal could be captured by the aircraft's ILS. SHK returns to the question of training in section 2.2.

The controller initially believed that there was nothing abnormal about the turn, that it was only an adjustment of the course towards the approach point and that Midland was probably visual with the airport. Against that background, the controller did not see any reason to intervene.

When Midland passed 270 degrees during a left turn and descended below the lowest radar vectoring altitude, and simultaneously approached a conflict situation with SE-VKA, there were sufficient indications that something was not right and that an immediate intervention by the air traffic controller had been justified. At that stage, however, there was communication between the air traffic controller and Midland regarding the conflicting VFR traffic. The communication was certainly an interfering element, which contributed to the failure to make a correction.

According to the crew, the TCAS was activated while the aircraft turned and descended, resulting in the RA command “MONITOR VERTICAL SPEED” which means that the crew must follow a green area on the vertical speed indicator to avoid collision and at the same time make the call "TCAS RA" to the flight controller. No such call was made. In most cases, a TA occurs before an RA, but in this case, the RA was given immediately according to the crew. In connection with pilots being trained and tested in the simulator, a TA is given first, which makes the pilots monitor and prepare for a possible RA command.

Since, according to the crew, there was no TA, which is thus reasonably expected by pilot, this may have been the reason why no call (TCAS RA) was made regarding the collision incident.