Preparing for a Safe Return to the Skies is a Safety first article published in June 2020 that already highlighted the increased risk of unreliable airspeed events after aircraft parking or storage. The number of reported occurrences since this article was published is still a reason for concern.

Between January 2020 and March 2021, 55 events of unreliable airspeed indication during takeoff were reported to Airbus.

44 of the 55 reported cases of unreliable airspeed at takeoff were due to obstruction of the Pitot tube. Obstructions can be caused by the presence of insects, sand, dirt, dust or any other foreign materials that could enter the Pitot when protective covers are not fitted to the aircraft when on the ground. In one reported case, the obstruction was because the protective covers were not removed before the flight.

The chart below (fig.1) shows the duration of time an aircraft spent on the ground before the flight when the unreliable airspeed event occurred.

(fig.1) Duration of time an aircraft spent on ground before reported unreliable airspeed event

Also beware during shorter ground stays

Contamination of Pitot probes by insects does not happen only during long periods of parking or storage. Half of all reported Pitot contamination related events occurs when the aircraft is parked for a time period of less than 48 hours. A significant number of reported occurrences of obstructed Pitots were on aircraft in transit and on the ground for less than two hours. Pitot probes are not always protected by covers during short duration transits.

Why Pitots are even more exposed to the risk of contamination during the pandemic

The COVID-19 pandemic had the effect of a significantly reduced number of flights, which means aircraft spent more time on the ground between flights. In cases where the air data probe protective covers are not fitted, the exposure to the risk of Pitot contamination is greatly increased.

In 36 of the 55 reported cases, the flight crew detected the speed discrepancy and rejected the takeoff. For many of the reported rejected takeoffs, the speed discrepancy could have been detected earlier during the takeoff roll, which would have incited the flight crew to reject the takeoff at a lower speed. The following case studies of three events of unreliable airspeed at takeoff highlight the importance of speed monitoring during the takeoff roll.

Flagged speed indications detected during the takeoff roll

An A330 aircraft was lined up for takeoff in night conditions. The first officer was the Pilot Flying (PF). The weather was clear with no wind. The recomputed takeoff decision speed was 150 kt and the rotation speed was 159 kt.

During aircraft acceleration, the speed indications were flagged on both PFDs.

A 100 kt callout and a rotation based on ground speed indication

The Pilot Monitoring (PM) made the 100 kt callout when the ground speed reached 100 kt. The PF then initiated the rotation at 159 kt of ground speed.

Unreliable airspeed indication procedure application

The flight crew applied the FCOM unreliable airspeed procedure when airborne and switched all three ADRs to OFF when the aircraft reached FL 110, activating the BackUp Speed Scale (BUSS) indication (as requested by the procedure when all ADRs are affected below FL 250).

In-flight turnback and overweight landing

The flight crew decided to return to the departure airport and performed an overweight landing.

When the aircraft finally returned to the gate, it was noticed that the protective covers were not removed before the flight and they were still fitted on all three Pitot probes.

Analysis of the recorder data confirmed that the Pitot covers, which were not removed before the flight, were the cause of the unreliable airspeed indication.

Three missed opportunities of detecting the covers

Post event analysis shows that the Pitot protective covers were not seen by the maintenance engineer during the external aircraft inspection, and neither by the captain during the preflight exterior walkaround nor by ground personnel during pushback, as recorded on the airport surveillance videos.

Speed display during takeoff roll

Recorder data also showed that the display of the SPD red flag on both PFDs from 50 kt of ground speed should have made the flight crew aware of the airspeed issue which would have enabled them to reject the takeoff.

① From 0 kt to 50 kt of ground speed: The Indicated Air Speed (IAS) was at the bottom of the speed scale on both PFDs and on the speed scale of the Integrated Standby Instrument System (ISIS). This was because the measured airspeed from all 3 ADRs was below 30 kt.

② From 50 kt to approximately 140 kt of ground speed: The SPD red flag was displayed on the speed scale of both PFDs and the IAS on the ISIS was still at the bottom of the speed scale.

③ From 140 kt of ground speed until rotation and liftoff: The IAS was between 30 kt and 50 kt on at least one of the PFDs. The IAS of the ISIS remained at the bottom of the speed scale.

(fig.2) PFD airspeed indication during the takeoff roll of the event

This sequence is in accordance with the display logic of the IAS on the PFD:

• If the measured airspeed is below 30 kt and ground speed is below 50 kt, then the IAS remains at the bottom of the speed scale
• If the measured airspeed is below 30 kt and the ground speed is above 50 kt, the SPD red flag is displayed on the speed scale
• When the measured airspeed is above 30 kt, it will be displayed on the PFD.

A speed discrepancy at the 100 kt callout

An A330 aircraft was ready for takeoff. The captain, who was the PF, applied takeoff power and the aircraft started to accelerate. The flight crew noticed a discrepancy between the PFD airspeeds at the 100 kt crosscheck. The flight crew continued the takeoff and performed the rotation at 133 kt.

ECAM cautions and level off for troubleshooting

Shortly after liftoff, the NAV IAS DISCREPANCY ECAM caution triggered, followed by the NAV ADR1 FAULT. The flight crew levelled off the aircraft at 3000 ft to perform ECAM actions and troubleshooting. The flight crew set the AIR DATA rotary selector to “CAPT ON 3” and resumed the climb to cruise FL 340.

Overspeed warning while approaching cruise FL

An overspeed warning triggered while the aircraft was passing FL 334. The autopilot remained engaged and the aircraft leveled off at FL 340. The overspeed warning stopped a few seconds later.

In-flight turn back

The flight crew performed an in-flight turnback and landed safely. Maintenance personnel inspected the Pitot on the captain’s side and found that it was obstructed by dust.

Flight data recorder analysis confirmed the effects caused by the obstruction of the captain’s Pitot probe.

A missed opportunity to reject the takeoff

If the airspeed was monitored even more closely by the flight crew during the takeoff roll, they may have identified the speed discrepancy sooner, allowing them to reject the takeoff and bring the aircraft safely to a stop.

Airspeed display during the takeoff roll

① The IAS remained at the bottom of the PFD speed scale on both PFDs, and on the ISIS, at the start of the takeoff roll until the measured speed reached 30 kt (which is its normal behavior). The captain’s IAS went above 30 kt when the first officer and ISIS IAS both indicated approximately 47 kt.

② When the first officer and ISIS IAS indicated approximately 80 kt, the captain’s IAS was only displaying 41 kt.

③ When the first officer and ISIS IAS indicated 100 kt, the captain’s IAS was only displaying approximately 55 kt.

④ The captain rotated the aircraft when 133 kt was indicated on the first officer’s PFD and ISIS, but the captain’s IAS was only displaying 80 kt.

(fig.3) PFD airspeed indication during the takeoff roll of the event

Rolling takeoff

An A320 was cleared for takeoff and the PF, who was the captain, performed a rolling takeoff. The captain performed the 1.05 EPR stabilization step and then applied takeoff thrust.

Rejected takeoff following an airspeed discrepancy at the 100 kt crosscheck

The aircraft accelerated nominally, but the captain identified a speed discrepancy when the PM did the 100 kt callout, and immediately rejected the takeoff.

Maintenance personnel performed troubleshooting when the aircraft returned to the gate and they found small pieces of leaves in the captain’s Pitot probe and its pressure line.

An early speed discrepancy on the captain’s side

① Recorder data showed that obstruction of the captain’s Pitot probe was providing an initial IAS of approximately 45 kt while the aircraft Ground Speed (GS) was 0 kt prior to the aircraft lining up on the runway.

② At the application of takeoff thrust, the first officer and ISIS IAS indicated 35 kt with nominal acceleration shown by the speed trend arrow on the first officer’s PFD. However, the IAS on the captain’s PFD was 49 kt with a very small speed trend arrow.

③ When 100 kt was reached on the first officer’s PFD and ISIS, the captain’s IAS was only 58 kt.

④ When the aircraft safely came to a complete stop after the RTO, the captain’s IAS was still at 55 kt.

(fig.4) PFD airspeed indication during the takeoff roll of the event

A useful 100 kt crosscheck!

During the entire takeoff roll, the captain’s IAS increased slightly but remained below 64 kt. The 100 kt crosscheck enabled the captain to identify the discrepancy and immediately reject the takeoff.

A possible earlier RTO

The Standard Operating Procedure requests monitoring of the PFD speed scale during the entire takeoff roll. Following this recommendation may have made the flight crew aware of the airspeed discrepancy earlier than the 100 kt callout and enabled them to reject the takeoff at lower speed. The first opportunity to detect the speed discrepancy was before lining up for takeoff, when the captain’s airspeed indicated 45 kt while the aircraft was stationary with ground speed at 0 kt. The second opportunity was at the application of takeoff thrust when the IAS on the captain’s PFD was almost steady speed with a very small speed trend arrow.

The three events described above illustrate the importance of closely monitoring the airspeed throughout the takeoff roll. Both the Pilot Flying (PF) and the Pilot Monitoring (PM) have a role to play.

While the PF maintains the aircraft on the centerline using external references, the PM must actively monitor the airspeeds from the start of the takeoff roll. This will allow for the PM to detect any inconsistent airspeed indications between instruments, an abnormal airspeed trend or absence of airspeed indications as early as possible.

The “Role of the Pilot Monitoring during Takeoff” video on the Airbus Worldwide Instructors News (WIN) website illustrates each of the various steps for the PM to perform during the takeoff.

Case study 3 shows us the importance of the 100 kt crosscheck, which is requested in the Standard Operating Procedure. It is the last line of defense in preventing a takeoff with an unreliable airspeed indication. The flight crew should be prepared to reject the takeoff at the time of the 100 kt crosscheck if an airspeed discrepancy is observed.

Takeoff with obstructed Pitot probes can happen for any flight as highlighted by the reported events described in this article. It is evidence of why it is essential to carefully monitor airspeed during every takeoff.