The point is checking the glideslope.
LEM is correct.
Consider the case of a near CFIT caused by flying a G/S that was left in the “test position” in the edited quotes in the following report (link to full report supplied by Shore Guy above).
CAA NZ
AIRCRAFT SERIOUS INCIDENT REPORT
OCCURRENCE NUMBER 00/2518
B767-319ER
ZK-NCJ
Synopsis
Early on the morning of Sunday 30th July 2000 the Air New Zealand Duty Line Manager was notified of a suspected false glideslope capture experienced by Air New Zealand flight NZ 60during approach to Faleolo International Airport, Apia, Western Samoa.
NZ 60 had been cleared to Faleolo via a FALE arrival for an ILS runway 08. The approach was planned to be an auto coupled ILS, using a low drag approach profile. During descent the aircraft was established on the 15 nm arc as per the STAR procedure.
Approaching the localizer course at 2800 ft LOC was armed, and the auto flight system subsequently captured the localizer inbound course. During the turn on to the localizer the aircraft was decelerated and configured to Flap 1. APP was armed after localizer capture and the auto flight system captured the glideslope shortly after. The crew reported a rapid energy increase, with speed increasing to near the flap 5 limit speed. To assist with energy control, while continuing to configure the aircraft for landing, the crew used speedbrakes and landing gear. The flight instrumentation glideslope deviation indicators displayed ‘on glideslope’ throughout the approach. Shortly after landing flap selection the PF (Pilot Flying) noted an anomaly in DME versus altitude. Around the same time the PNF (Pilot Not Flying), while trying to establish visual contact with the airfield and runway, became aware that visual cues did not correspond with what was expected. The SP (Supplementary Pilot) also became aware of an anomaly in aircraft position at approximately the same time as the two other crew members.
A go-around was commanded, initially climbing straight ahead followed by a climbing left turn, to pick up the 340o radial FA VOR to rejoin the 12 nm arc for a subsequent approach.
This second approach was flown with careful attention to distance and altitude, using the
published DME recommended altitudes as per the LOC (GS out) table on the approach plate for glide path management. The glideslope deviation indicator also indicated on glideslope throughout the second approach. The glideslope indications were ignored and the approach continued to a successful landing.
Subsequent analysis of the FDR information established that the aircraft had descended on a glide path of approximately 3.5° to a point approximately 5½ miles short of the runway with ‘normal’ localizer and glideslope indications displayed on the flight instrumentation.
It was later established that the ILS glideslope transmitter had inadvertently been left in
control (monitor) bypass mode, with the unserviceable transmitter selected. In the bypass
mode, the glide path transmitter executive monitor was unable to shut down the faulty
transmitter or to transfer to the serviceable transmitter. The result was the radiation of invalid glideslope information consisting solely of the carrier plus side bands (CSB) signal component. The side bands only (SBO) signal component was missing from the glideslope transmission.
The distinction between a ‘false’ glideslope and an ‘erroneous’ glideslope must be
appreciated:
__ A false glideslope is a recognised phenomenon and is a normal by-product of the ILS
transmission. A false glideslope provides a distinct, but incorrect, path to the origin of
the glideslope.
__ An erroneous glideslope, however, does not provide a defined path. Whilst continuously indicating to the crew that the aircraft is on slope irrespective of its position in space, with no warning flags visible, very little or no guidance is being given to the aircraft. A crew using an erroneous glideslope is utilizing a system that has an error present, for example; a glideslope transmission that has a component of the signal missing or the components being radiated in the incorrect phase relationship. An
erroneous transmission may occur intentionally during maintenance or testing, or inadvertently due to maintenance error.
Findings
Aircraft Equipment
3.2.8 The absence of the SBO signal from the transmitted glideslope signal leaves a balanced CSB [only] signal that may not be immediately identified by the crew as an invalid “on slope” signal. The CSB [only] signal is interpreted by the airborne ILS receiver as a glideslope with infinitely wide limits. The result
may be the course deviation indicators remaining centered on the display with no warning apparent to the crew. Shifting the phase relationship between the CSB and SBO signals may have a similar effect.
Ground Equipment
3.4.4 Pilots unknowingly using an ILS transmitter, while the transmitter is on maintenance or test, may use erroneous information because there may be no flight deck warning.
5. Safety actions taken or agreed
5.1 Air New Zealand
As per the provisions of NZCAR 12.59 (Investigation and Reporting), NZCAR 119.79 (Internal Quality Assurance) and AC 12-2 (Incident Investigation), Air New Zealand has raised corrective and preventive actions to address the issues
raised by this report. In addition to actions aimed specifically at system improvement, the relevant activity is as follows:
__ Increasing flight crew knowledge and awareness regarding glideslope altimeter checks.
ICAO State Letter AN 7/5-01/52 and Attachments
1. I have the honour to draw your attention to a number of incidents which have occurred in recent years resulting from the operational use of instrument landing system (ILS) signals being radiated during ILS testing and maintenance procedures, specifically for phasing and modulation balance testing. Such signals may be perceived onboard aircraft as “on-course” and/or “on-glide-path” indications regardless of the actual position of an aircraft within the ILS coverage and with no flag or alarm indication in the cockpit. The use of ILS localizer and/or glide path signals for approach guidance during these testing and maintenance procedures can therefore result in false indications to the flight crew and has the potential to cause a controlled flight into terrain (CFIT) accident.
US Department of Transportation
Federal Aviation Administration
National Airspace System Operations Program
Evaluation of Glideslope Operations and Maintenance
With Respect to Incidents of
Near-Controlled Flight into Terrain
Area Investigated
Description of Area
The group evaluated reports of two incidents of near controlled flight into terrain. Part of this evaluation included flight data recordings that were conducted by an aircraft using a GS facility that was not radiating sideband energy. This resulted in a
constant on-glide path indication to the cockpit, though it did not exactly duplicate the radiated signal of a GS facility that is radiating a signal in quadrature as is the case during phasing. The group also evaluated applicable publications that govern
facility maintenance and aircraft operations.
Conclusions
Certain changes can be implemented that will enhance the integrity of terminal flight operations. Though there have been no reported incidents of CFIT in the NAS, there are unanswered questions about the specific events, which led to the international
incidences of near-CFIT. The possibility of a CFIT incident occurring in the future is a concern and the probability of such an event is uncertain. Given the two events to date, and the fact that calibration of the system produces a hazardously misleading
GS signal, it appears improvement in GS system integrity is necessary.
Recommendations
Actions taken
Operations Manual. The Boeing Company initiated a change to their operations manual. This change: requires aircrews to cross-check their altitude at the final approach fix (FAF). Though they consider this to be basic airmanship, they felt the added emphasis would enhance aircrew situation awareness. Refer to Appendix 4, Option I.
To those of us who are a little older, it is rather chilling in the light of the above report to reflect on those previously “unexplainable” CFIT events during ILS approaches that have happened over the years (and possibly LOC-only approaches when a “good” G/S was radiating in the test mode).