Lufthansa A321 at Hamburg on Jan 2nd 2015, odour in cockpit

Last Update: June 22, 2018 / 16:55:31 GMT/Zulu time

Bookmark this article

Photo of Lufthansa D-AIRM, Airbus A321 — Lufthansa D-AIRM, Airbus A321 (Photo credit: ATom.UK / Flickr / License: CC by-sa)

Incident Facts

Date of incident
Jan 2, 2015

Classification
Incident

Airline
Lufthansa

Flight number
LH-2

Departure
Frankfurt/Main, Germany

Destination
Hamburg, Germany

Aircraft Registration
D-AIRM

Aircraft Type
Airbus A321

ICAO Type Designator
A321

A Lufthansa Airbus A321-100, registration D-AIRM performing flight LH-2 from Frankfurt/Main to Hamburg (Germany) with 84 people on board, encountered an odour in the cockpit prompting the crew to don their oxygen masks at some stage during the flight. The aircraft continued for a safe landing in Hamburg.

The occurrence aircraft remained on the ground in Hamburg for about 53 hours, then positioned and returned to service about 56 hours after landing.

The French BEA reported in their weekly bulletin that the occurrence was rated a serious incident and is being investigated by Germany's BFU.

On Jun 15th 2015 Germany's BFU reported in their January bulletin that cabin crew had noticed a nasty smell in the aft cabin during climb out of Frankfurt, which disappeared after a short time. On approach, while at about 5000 feet, an increasingly pungent and nasty smell developed in cockpit and cabin, the crew switched bleed air system #1 off and activated cross bleed, both flight crew felt suddenly unwell with their fingers prickling and donned their oxygen masks about 90 seconds after the smell was noticed, ATC was advised of the smell and oxygen masks however without declaring PAN or emergency. After a few breathes through the oxygen mask the pilots recovered. The aircraft landed safely and taxied to the gate, pilots still wearing their oxygen masks, where passengers disembarked normally.

The BFU reported that a subsequent measurement with an aerotracer did not reveal any results that could explain the odour.

On Jun 22nd 2018 the BFU released their final report concluding the probable cause of the incident was:

During the flight a smell event and symptoms occurred, which caused the flight crew to don their oxygen masks.

It could not be conclusively clarified if and what contaminations caused the smell and which connection there might be with the symptoms.

The BFU released following safety recommendation:

Safety Recommendation No. 04/2018

The Bundesverband der Deutschen Luftverkehrswirtschaft e.V. (BDL, German
Aviation Association) should coordinate the activities of the operators organised in the association that the crews are provided with relevant information regarding the existence, possible causes, and effects of fume events in aircraft cabins and are trained accordingly.

The BDL should organise an exchange and coordination process of the operators organised in the association regarding the content and course of action in order to achieve harmonised implementation.

Secured findings from research and development are to be used. The ICAO document "Circular 344-AN/202, Guidelines on Education, Training and Reporting Practices related to Fume Events" should serve as the basic principle.

The Directorate General Mobility and Transport of the European Commission, in cooperation with EASA, has initiated a project investigating cabin air quality (FACTS). The BFU therefore refrains from issuing further safety recommendations.

The BFU described the events:

The airplane departed Frankfurt Main Airport at 07:43 hrs for a flight to Hamburg. The Pilot in Command (PIC) was the Pilot Flying (PF), the co-pilot the Pilot Monitoring (PM). During climb the senior cabin crew member informed the PIC that at the doors unpleasant smell was noticeable. The smell was described as "scharfer Geruch (fierce smell)" like "verbrannter Staub (burnt dust)". But it did not smell like "Käsefüße (smelly feet)". The PIC justified the smell with the long time period the airplane had had between flights and the "erhöhte[n] Luftfeuchtigkeit (the high humidity)". The PIC mentioned then that he too had noticed the smell as well and described it as "Öl (oil)" and "TCP". The Cockpit Voice Recorder (CVR) did not record any other reports concerning smell.

At 0823:50 hrs during the approach to Hamburg the PIC said: "Jetzt riecht‘s wieder (now it smells again)". He made the connection with the airplane just having entered clouds. The co-pilot described the smell "mehr wie so Käsefüße (more like smelly feet)", whereas the PIC used the term "oil smell". The PIC switched off "das einser Bleed (the one bleed) and "crossbleed on". Approximately 30 seconds after the smell had started the two pilots began to complain about its intensification. They expressed this with drastic words. Now the smell was described as "beißend (caustic)". Since the switching off of the bleed air of Engine #1 did not reduce the intensity of the smell, at 0824:50 hrs the bleed air system was changed so that the cabin air was supplied by engine No 1 only. This configuration did not reduce the smell either. At 0825:16 hrs the co-pilot suggested they don their oxygen masks. The PIC confirmed this and two seconds later the two pilots donned their oxygen masks.

The CVR recorded several deep breaths followed by a brief conversation about the quality of the communication while using oxygen masks and the PICs instructions to configure the airplane for landing. Then the co-pilot informed the air traffic control unit that smell had occurred and the pilots donned their oxygen masks and therefore unusual "sound" would be heard. He also stated that they would not require any further assistance and it was a normal approach. Then the procedures for landing and communications with the air traffic control unit were conducted.

At 0828:40 hrs the co-pilot asked the PIC how he was doing. The PIC answered: "Also mir geht’s sehr, sehr gut, hab keine Problem im Moment (I feel very, very good, I do not have any problems right now)". He also said that prior to touch-down the smell had been caustic and he had had a short moment of light-headedness. The copilot replied that he had felt prickling.

When the PIC informed the air traffic control unit about the number of occupants on board the airplane he made an error in summation but recognised it right away and corrected the number.

When passing 1,000 ft AGL during the approach the FDR recorded the following parameters, among others.

Deviation LOC8: 0.00039 ddm9 (equals 0.00 DOT10)
Deviation G/S11: -0.00391 ddm (equals -0.01 DOT)
Rate of descent: 720 ft/min
Airspeed: 151 CAS12

Landing gears were extended and the flaps lever was in position FULL. The function Autothrust was in Speed-Mach-Mode and deactivated at this altitude. During the approach the landing checklist was completed and the wind conditions discussed. At 0831 hrs the airplane landed on runway 23. The FDR recordings showed that touchdown occurred with a speed of 131 CAS. Hereby at most the following accelerations were reached: ny: 0.07/-0.12 g, nz: 1.29 g.

When the co-pilot read back the APRON13 frequency he mixed up two digits. The air traffic control unit pointed it out to the co-pilot whereupon he read back the correct frequency.

The fire brigade accompanied the airplane until it reached Gate 15. Shortly before Gate 15 was reached the Auxiliary Power Unit (APU) was switched on. The CVR recordings showed that when the flight crew arrived at the gate they made jokes about their appearances wearing oxygen masks. After they had switched off the engines they removed the oxygen masks.

After the landing the pilots talked intensely about the prevailing wind conditions and the therefore necessary challenging control technique. The PIC expressed how impressed he had been how the changing wind conditions had forced him to react quickly with his control inputs.

To the fire brigade the PIC said that they had donned their oxygen masks as precautionary action.

The pilots stated that they had not completed any checklist to identify the source of the smell because they wanted to focus on the landing. There was no information that passengers had noticed the smell.

The BFU reported that no leakages or contaminations were found on the aircraft. An Aerotracer was used during a high power engine run up before the filters were replaced. During the measurement an indication of "GLYCOL OLD" occurred. The raw data of the measurement were analysed by the manufacturer of the Aerotracer, who confirmed the test results and added that Acetone and Ammonia were detected during the test too.

The BFU charged the Institut für Arbeits-, Sozial- und Umweltmedizin der
Universitätsmedizin Göttingen (ASUM-UMG, University Medical Centre Göttingen) to assess the course of the flight and the impairments the pilots had described and summarized Göttingen's findings:

The symptoms described in the recorded conversations are consistent with those symptomatologies of other cabin air incidents [Abou-Donia et al., 2013; Heutelbeck et al., 2016]. They can be explained, among other things, by the occurrence of Volatile Organic Compounds (VOC) in the cockpit / cabin air.

VOCs comprise seven compound classes: Alkane, aromatics, terpene, chlorinated hydrocarbon, ester, aldehyde, and ketone, and others [Seifert, 1999]. Many of these compounds are ingredients in lubricants or hydraulic fluids (before or after pyrolysis), kerosene, and de-icing fluids used in aviation [Ritchie, 2003; Mair et al., 2015, Rosenberger et aI., 2016]. […]

The described symptomatology is consistent with the impact of VOC. Because of the described clouding of consciousness18 and the appearance of "verwaschene Sprache (slurred speech)" of a pilot here the neurotoxic effect of VOC is paramount.

The observed fast effect, e.g. on the central nervous system, and the quick disappearance as soon as the exposure ended or was interrupted, is characteristic for VOC. Limitation of performance under exposure is surely present.

Regarding the provided conversation transcriptions and the occurrence of exposure the pilots do not report any of the indicator symptoms of an organophosphate intoxication. As explained in the toxicological statement the symptomatologies described in both cases are consistent with the impact of VOC. In the present cases, the impact of organophosphate cannot safely be ruled out, but due to the currently available information it is rather unlikely.

In addition the BFU asked the the Kriminaltechnische Institut des Bundeskriminalamtes (BKA, Federal Criminal Police Office) to assess the CVR recordings regarding any signs of possible intoxication or stress present in the voices. The BFU wrote: "The result showed that the persons involved did not suffer from impaired consciousness during the occurrence. There are no irregularities regarding voice, language, and manner of speaking, which would indicate abnormal stress in the speaker."

The BFU analysed:

The CVR recordings show that the smell was immediately associated with engine lubricant. It is possible that this is connected with the assumption expressed during climb that it must be "Öl (oil)" or "TCP".

The PIC then initiated actions to clear the smell by changing the bleed air supply. Since the flight crew assumed bleed air supply contamination these actions are considered correct.

Because the smell could not be eliminated and the pilots associated their health impairments with it, donning their oxygen masks was a logical action. Immediately after donning their oxygen masks the pilot had perceived improvement of their symptoms. The BFU is of the opinion that the flight crew had taken the correct action to eliminate the symptoms. This is also true considering the fact that flying under oxygen masks causes impairments for the conduct of the flight (e.g. impeded communication).

Then the flight crew completed the landing checklist. Based on the landing mass, the reported wind conditions and the configuration of the aircraft target approach speed (IAS-TARGET)20 was 153 kt at a flight altitude of 1,000 ft AGL21. While passing this altitude the speed was adhered to with the tolerances permitted in the OM-A. Neither the vertical and lateral ILS deviations nor the allowable maximum sink rate were exceeded. Thus, the criteria for a stabilised approach were adhered to. The analysis of the airspeed shows that the pilot was able to fly the aircraft during the approach within the tolerances.

...

The fact that the flight crew associated the smells with the entry in to clouds, but also with the odourless TCP shows that they had an unclear and biased conception about their causes. The BFU is of the opinion that it is necessary that crews are trained regarding the many causes of smells, their classification, and the possibly resulting hazards.

With respect to the technical examination the BFU analysed:

The examination of the airplane did not reveal any findings, which would have explained the smell event. However, the technical examination was limited to the assumption that the smell had been caused by contaminated bleed air. Other possibilities were not taken into account. The BFU is of the opinion that there are different sources for smell development in airplanes. These are not solely limited to contaminated bleed air. These can also be substances, which are already on board and released during normal operation or malfunction (BFU Study 803.1-14, pages 49 ff and ICAO Circular 344-AN/202, Item 2.3.2).

With respect to the Aerotracer results the BFU analysed:

The BFU cannot fully explain why the air quality measurement device has generated the indication "GLYCOL OLD/?", even though on that day the airplane had not been de-iced. There is a possibility that de-icing fluid residue from previous de-icing procedures was present in the ECS. The technical examination of the airplane did not reveal any liquid residue in the examined part of the bleed air system. It is possible that very small amounts had remained undetected in the ECS, which then caused the indication. However, it is then not understandable why the smell did not occur during the previous flights ad why small amounts should have triggered these symptoms. It can also not be explained why the smell only developed during certain flight phases.

There is also no explanation as to why the indications Acetone and Ammonia occurred. Due to the description of the smell and the measured extremely low concentration it is unlikely that the substances were involved in the smell event. It was also not possible to draw conclusions as to the concentration of these substances at the time of the smell event during the flight.

With respect to the aircraft manufacturer statements the BFU analysed:

The aircraft manufacturer told the BFU that VOC occur with various products, which are used in aircraft production. They also confirmed that VOC can occur in cabin air if the bleed air is contaminated with them. According to the results of the studies the concentration of VOC, created through thermal degradation, in the cabin air is so low that the target values for air quality of interior rooms were not exceeded. Based on these statements the BFU does not understand, how a contamination of the cabin air with VOC via the ECS could have occurred, which then caused the symptoms of the crew.

With respect to the toxicological assessement the BFU analysed:

Even though it cannot be ruled out that VOC caused the observed symptoms, it is not plausible in regard to the necessary concentrations in the cabin air. The toxicological statement of the ASUM-UMG assessed the presence of organophosphates (e.g. TCP) as cause for the symptoms as unlikely. It was also not possible to determine other contaminations as cause. The diversity of the described symptoms can physiologically not be explained by a single underlying biological pathway. It is therefore not possible to state a cause for the symptoms of the crew.

Taking into consideration the statement of the aircraft manufacturer, the EASA study, and the measurement results of cabin air during other fume events, the BFU cannot comprehend that contaminated cabin air could trigger the kind of symptoms the pilots noticed.

Hence, after the smell occurred other, possibly psychological reactions, must have triggered the symptoms or acted as catalyst.