S-76D Loss of Control on Approach to an Indian Drilling Rig

Posted by on Feb 8, 2025 in Accidents & Incidents, Crises / Emergency Response / SAR, Design & Certification, Helicopters, Human Factors / Performance, Offshore, Oil & Gas / IOGP / Energy, Regulation, Safety Culture, Safety Management, Survivability / Ditching | 0 comments

Pawan Hans Sikorsky S-76D Loss of Control on Approach to an Indian Drilling Rig (VT-PWI)

On 28 June 2022 Pawan Hans Sikorsky S-76D VT-PWI, contracted by ONGC, impacted the sea on approach to the Sagar Kiran jack-up drilling rig in the Arabian Sea. 

The helicopter capsized and while all 9 occupants egressed the helicopter, four passengers died in the water before they were rescued.  Two passengers suffered serious injuries.  Another passenger and the two flight crew escaped with minor injuries.

Salvage of Pawan Hans Sikorsky S-76D VT-PWI (Credit: AAIB)

The Indian Aircraft Accident Investigation Bureau (AAIB) published their safety investigation report in August 2023.

ONGC, Pawan Hans and the Introduction of the S-76D in India

State owned ONGC is the largest oil and gas company in India.  As is also common in Mexico and Brazil, the state oil company is the main customer for offshore helicopters in their domestic market.

Pawan Hans (PHL) was formed in 1985.  ONGC owns 49% of the company, with the Government of India holding the majority stake. The Government has made 4 attempts to sell off their stake in the last 8 years.  Privately owned competitors to Pawan Hans include Global Vectra and Heligo.

Pawan Hans operated 44 helicopters of 7 helicopter types at the time of the accident.  Offshore operations were their largest operating sector.  For many years the Airbus AS365 family had been their prime offshore type, with 31 in their fleet in 2022.  

According to the in the AAIB report:

ONGC…had put the criteria that aircraft being made available for operations should not be more than 7 years old [presumably for the commencement of new tenders]. Pawan Hans was not in a position to meet the customer demand without induction of newer aircraft in its fleet. Thereby, PHL’s top management decided to go for induction of new helicopters through leasing option.

An attempt to lease S-76D and S-76C++ helicopters from a leasing company was made earlier but could not fructify. Hence, alternate leasing options were explored and PHL processed for leasing seven S-76D helicopters with planned induction in 2019-2020.

These aircraft, formerly operated by Thai Aviation Services, were being replaced by Leonardo AW139s.

After disruptions and delays owing to Covid-19, a master lease agreement was signed with lessor in October 2020.

Pawan Hans was required to comply with provisions contained in the Air Operators Certification Procedure CAP 3400 for induction of new helicopter in its fleet.  Pawan Hans submitted request for pre-application meeting to [Directorate General of Civil Aviation] DGCA on 10 Nov 2020.

Pawan Hans thus became the only operator of the S-76D in India.  The aircraft was leased by Milestone Aviation company Vertical Aviation No1 Ltd to Pawan Hans though to 17 February 2028.

The aircraft arrived at Pawan Hans on 26 Aug 2021 in disassembled condition. Aircraft had 4636:42 hrs at the time of delivery to Pawan Hans. Aircraft was assembled and ground run was carried out on 23 Oct 2021.  The test flight for issue of ARC was carried out on 11 Jan 2022, and Certificate of Airworthiness and ARC was issued by DGCA on 28 Mar 2022.

Pawan Hans has a Safety Management System (SMS), accepted by DGAC.  Pawan Hans had
identified various risk mitigations for the S-76D introduction, including:

  • Experienced offshore S-76D, TREs/TRIs to be recruited
  • S-76D offshore experienced PIC will be hired/deployed in initial stages
  • Existing Pawan Hans offshore PICs to be trained on S-76D and utilised as co-pilots 

However, regulatory approvals were delayed because no DGAC Inspector had been type trained on the S-76D (though this is a mere recommendation in ICAO Doc 8335).  As an possible alternative:

DGCA could have opted for short term secondment of type qualified pilots…

This may not have been easy to do as training was also delayed because of the difficultly Pawan Hans had getting type experienced ex-pat instructors in-country during a pandemic and DGAC approved and the plan to have an initial cadre of PICs with S-76D experience did not materialise.

Ironically, if the more modern Leonardo AW139 had been selected some of the difficulties would have been lessened as the type was in service in India.

Pawan Hans did have a Flight Data Monitoring (FDM) programme, encouragingly a regulatory requirement in India for offshore operations.  They used an outsourced analytics provider but as only a limited set of extreme alert levels had been set initially, no exceedances had been analysed and debriefed prior to the accident.

The Accident Crew

The Pilot In Command (PIC) had 6763 flying hours of experience, 94 on type.  The PIC was one of 8 initially shortlisted by Pawan Hans for conversion to S-76D.  The PIC underwent simulator and ground training in June-July 2021 but returned to continue flying the AS365N3, awaiting deliveries of the S-76D. He started flying the S-76D offshore in May 2022, as a commander rather than initially as a co-pilot, as originally planned.

The co-pilot had 5281 flying hours of experience, 73 on type.  He underwent S-76D simulator and ground training in August 2021, again flying AS365N3s until May 2022.

The Accident Flight

The helicopter took off from Juhu airport, outside of Mumbai at 0538 UTC with an ETA at
Sagar Kiran of 0610 UTC with two crew and 7 passengers. The helicopter had flown a further 153:37 hours since import at that point.

Due to the presence of cumulonimbus clouds the helicopter re routed after getting clearance from the ATC.

Approximately…16 NM from destination helideck, descent was commenced… At…6-7 NM, the helicopter was at 500 feet AMSL. At about 5 NM [the] crew carried out pre-landing checks. [the] PIC who was Pilot Flying [PF] instructed the Co-Pilot, who was Pilot Monitoring [PM] to keep his side weather radar ON, due to poor visibility. Thereafter, [Flight Management System] FMS was programmed to position the aircraft on final approach track of 290° at 02 NM.

At c 06:07 UTC, the Sagar Kiran Radio Operator reported to the crew that “the prevailing winds were varying between 3-5 knots with bearing 340° and temperature of 26° with light rain/ precipitation”.  The installation also had recorded the following:

At 1.5 NM from the Rig, the helicopter was aligned on finals…at a height of about 500 feet and maintaining speed of about 70 kts.

Sagar Kiran Mobil Drilling Unit (MODU): (Credit: via AAIB)

Hereafter, PIC handed over controls to the Co-Pilot. Autopilot was decoupled at about 1 NM from destination helideck and visual approach was continued. PIC who was now Pilot Monitoring carried out checks on Finals.

At about 300 feet altitude PIC called SPEED and went on to switch off the Co-Pilot side Weather Radar which had been left ON. 

The investigators comment that a lack of familiarity with the S-76D resulted in the PIC using a sub-optimal method of doing this in the circumstances (reaching across to the Co-Pilot’s Control Cursor Device [CCD] to use the weather radar’s Virtual Control Panel [VCP], rather than pressing the WX RADAR button on the overhead panel).  This lengthened the distraction, for the benefit of avoiding a 2 minute warm up period when reselecting the radar.

As such, the additional layer of safety monitoring by the PM [now the PIC] was not available during crucial few seconds [estimated as 10-14 seconds] before the helicopter deviated from the approach path and the significant deviation of helicopter from the stable flight path could not be identified and prevented.

While the PIC was switching off the Weather Radar on Co-Pilot side, his attention was directed to EGPWS warning and an alarmed reaction from Co-Pilot with helicopter sinking at a high rate of descent.

PIC took over control and tried to take recovery action. The helicopter, however, continued to descent and impacted the sea at 06:11:51 UTC.

The impact was at a rate of decent of 2500 fpm and 6.8G.  

Recovered Wreckage of Pawan Hans Sikorsky S-76D VT-PWI (Credit: AAIB)

The CVR transcript for the later portion of the flight was as follows:

Note the EGPWS gave only 7 seconds of warning (which we discuss later).

The Safety Investigation into the Loss of Control

Examination of CVFDR showed that at 06:11:07 UTC, 44 seconds before impact, and at 706 ft the autopilot was decoupled (indicated by the vertical line below): 

AAIB note that:

Pressing the cyclic TRIM REL switch removes all artificial spring feel forces on the cyclic, and because the cyclic becomes completely free, this can cause overcorrection. It is not a recommended method to frequently press the TRIM REL switch to make changes in attitude or bank except during last stages of approach very close to the landing area necessitating larger attitude changes.  Pressing the cyclic TRIM REL switch is not a standard practice, but appears to be a habit resultant from military training and flying on ‘Mi’ helicopter variants amongst Pawan Hans pilots as is evident from the remark ‘Recommended to get more familiar with a cyclic trim release’ by TRE in the training report of the PIC.

Initially:

There was no significant flight path deviation immediately after the aircraft decoupled indicating that the AFCS was functioning normally. The PM continued to give speed and height callouts to the PF.

The aircraft was descending at 34 fpm 1.7º nose up. Ground Speed and IAS were 47 kts and 56 Kts respectively.

While the altitude was 750 ft, the helicopter began to decelerate. At about 06:11:31 the pitch was increased nose-up and the pilot continuously continued reducing the torque. The pitch angle increased to 7.1 Nose up and while Ground Speed and IAS were recorded as 46 Kt and 58 kt respectively. In another 3 seconds, while the Engine 1 and 2 torque setting was at 15% and altitude was 752 ft, Pitch angle increased to 12.1 Nose up. The ground speed and IAS were 39 kts and 54 kts respectively and rate of descent increased to 364 fpm.

Pawan Hans Sikorsky S-76D VT-PWI Approach Path (Credit: AAIB)

Progressive and excessive raising of the nose and simultaneous reduction in the collective pitch put the helicopter into a situation where available power was insufficient to maintain the altitude.

At 06:11:41 the maximum pitch up angle of about 23.5° was recorded and this more or less coincides with the groundspeed reducing to almost zero. By this time the aircraft had entered an energy deficient state, due to the low torque value and near zero speed.

The rate of descent increased to 1136 fpm. High nose-up attitude combined with reducing air speed at low altitude made it practically difficult to recover the helicopter from impact with the water.

The onset of uncontrolled flight occurred when the aircraft was pitched to approximately 23° nose up while simultaneously lowering the collective to the full down position.

At 06:11:46, few seconds prior to impact the PIC initiated a full collective pull. At 06:11:48, the collective reached its maximum upward limit, resulting in engine torque increasing to 130% (combined) and demanding maximum available torque. At this point the aircraft was approximately 153 ft above the surface of the water. This resulted in Nr dropping to below 90% Nr and a rapid left yaw of the aircraft. However, by this time due to a very low height above water, the recovery action did not yield the desired results and the helicopter impacted with the water at 06:11:51 UTC at vertical speed of -2,500 fpm (3.8G).

Spectral analysis of the CVR detected a rotor speed droop which initiated c5.4s prior to impact.

The droop initiated at a rate of 3.9% per second to 99.5% NR and then increased to a droop rate of 15.5% per second to 77.8% NR. Rotor speed then began to recover at a rate of 10.9% per second to 99.6%NR before impacting the water. 

The events between CVR elapsed time 02:01:31 hrs and time of impact are correlated with FDR data here:

From the CVR data, the initiating event for the onset of the [main] rotor speed droop appears to be in reaction to the first EGPWS alert to “PULL UP”. This is corroborated by the FDR data which shows that just following the alert there is a rapid Collective up movement and an accompanying rise in Engine Torque. Further, the increased rate of rotor speed droop appears to be in reaction to the second EGPWS alert to “PULL UP”. This is also corroborated by the FDR data which shows another rapid rise of the Collective and a demand in Engine Torque to greater than 125%.

AAIB do not discuss what EGPWS training was given.  AAIB do however opine that:

An experienced co-pilot (offshore Captain on earlier type (Dauphin or other)) and an experienced expat S-76D pilot as PIC, as was initially planned would have handled the situation in a different manner and a safe landing could have been achieved.

AAIB discuss the method the crew used to switch off the radar and suggest it indicated a lack of familiarity with the type and was a distraction at a critical moment.

An experienced (expat) pilot as initially planned by Pawan Hans, would have coached the co-pilot to gain experience and be fully conversant with the cockpit instrumentation and controls.

The Sagar Kiran did have a VHF radio recording capability but it was found by investigators to be unserviceable and not recorded as such in the rig’s daily serviceability reports.  No data was forthcoming from other installations in the area either.

Post Impact Survival & Rescue

AAIB say that the Emergency Flotation System (EFS), which is installed in the nose and main landing gear bays, “got deployed on impact”.  They unfortunately do not specify if that activation was by the crew, the Automatic Float Deployment System (AFDS) or both.  Despite being older, the S-76 AFDS is however a more capable design with greater redundancy than the S-92A AFDS. 

Despite the EFS deployment the helicopter however capsized, though it did initially remain floating on the surface inverted.  Rescuers later reported the wave height was 2-3 meters.

The helicopter was equipped with belly mounted external liferafts from Dart Aerospace. 

Sikorsky S-76D Dart Aerospace External Life Rafts (Credit: via AAIB)

While one of the two cockpit handles was found to have been activated, the rafts did not deploy.  “The OEM” reportedly stated to the AAIB that the “Life Raft Pod Assemblies are designed to sustain a maximum vertical entry speed of 300 fpm”. While far less than the actual rate of descent (>2000 fpm) this is consistent with a concept of use after a controlled ditching rather than a high energy water impact (or ‘crash’).  Investigators found the two inflation hose connections broken and disconnected.  Neither belly mounted pod was recovered, preventing further examination.

Pawan Hans Sikorsky S-76D VT-PWI Aircraft side pneumatic connection to Life Raft Pods (left side and right side) (Credit: AAIB)

The Radio Operator on Sagar Kiran witnessed the water impact.

Emergency Response was triggered by Sagar Kiran for “helicopter ditching near the rig” and a rescue boat was launched with a rescue team.

At 0615 UTC the Radio Operator “asked standby OSV Malviya 16 [sic] [to] approach” that location.  The vessel, which appears to be a modified supply vessel not a specially equipped standby vessel, gave a 45 minutes estimate for its arrival, even though AAIB imply it was no more than 6 nm away.  

At that point the helicopter was visible at a distance of about 1 NM from the rig. The rescue team observed two persons floating in the sea, motionless.

One minute later the rig launched a boat itself.  However this was a Totally Enclosed Motor Propelled Survival Craft (TEMPSC) not a Fast Rescue Craft (FRC) and so unsuited to the role.  The rig’s FRC had been unserviceable for over a year.

Sagar Kiran: Lifeboat (TEMPSC) and Unserviceable Fast Rescue Craft (FRC)(Credit: AAIB)

One person was observed swimming and calling for help and was rescued by the rescue team.  The rescue boat after rescuing one survivor headed to the Rig.

There were still five persons on the upturned fuselage of the S-76D at this point.

Wreckage of Pawan Hans Sikorsky S-76D VT-PWI Afloat (Credit: via AAIB)

The rescue boat did not attempt to retrieve motionless passengers who were observed floating in the sea or offer any assistance to those holding on to the floating helicopter keeping safety of rescue team in mind.

The boats departure…

…caused panic amongst the survivors holding on to the helicopter.  The rescue boat did not go back to the helicopter as they feared that debris from helicopter may damage the rescue boat. 

In practice an ungainly TEMSEC was more likely to have further damaged and potentially sunk the helicopter.

Meanwhile, one of the passengers who had initially held on to the floating helicopter lost his grip and also drifted into the sea. After arrival of Malviya-16 at around 0655 UTC [40 minutes after they were tasked], all four persons holding on to the floating helicopter were rescued and brought aboard.

The last was recued at 0733 UTC.  As the vessel had not launched a boat, the survivors had to board via rope ladders or scramble nets, which proved difficult (the rope ladders were 4 m too short!).

Malviya-16 Scramble Net (Credit: via AAIB)

Thereafter, the helicopter sank into the sea.

Navy and Coast Guard Search and Rescue (SAR) aircraft were tasked to the scene and recovered the 4 bodies.  Asphyxia due to drowning was the recorded cause of death.

The flight crew reported no problems with their lifejackets.  The passengers were equipped with LRS001 lifejackets with a hybrid rebreather.  

LRS001 Lifejacket (Credit: via AAIB)

While the lifejacket must be activated manually with a single ‘beaded toggle’, the charging of the rebreather with air can be activated manually with a ‘hammer toggle’ or by automatically on contact with water. 

LRS001 Lifejacket Rebreather Hammer Toggle (Credit: via AAIB)

Three of the four passengers who that drowned and one survivor had uninflated lifejackets.  The five lifejackets (of the four deceased and the one survivor) were examined:  

  • One had its inflation toggle successfully activated and its rebreather cylinder had discharged automatically.
  • Three had their rebreather ‘hammer toggles’ ripped off but the inflation toggle had not ben pulled.
  • On one jacket neither had been pulled.

The four uninflated jackets all successfully inflated when then tried.

Investigators examined the passenger safety briefing video and found the ‘Instructions for Inflating the Life Jacket’ were incorrect.  They explain:

While audio narrative [in English] was “to inflate the life jacket pull the toggle sharply after jumping into water. This will inflate your lifejacket automatically”, the video pointed at the Hammer Toggle.

The Inflation toggle, which is meant to inflate the life jacket was nowhere identified in the Passenger Safety Briefing video.

We believe these comments are misleading as the lifejacket shown in the video was according to the AAIB a different one, with a ‘hammer toggle’ for inflation.

The Hindi narration differed from the English and said to pull toggles “on both sides” to inflate.

The Passenger Briefing Card also showed the older lifejacket.  This was spotted by an Indian DGAC Airworthiness Inspector but Pawan Hans staff just temporarily fitted the older lifejackets to get an Airworthiness Review Certificate issued.  

The old jackets were used when a emergency egress demonstration for DGAC was conducted, but the newer LRS001 was used for the DGAC proving flights.  This difference was not noticed by DGAC (or by ONGC in their review).

AAIB Probable Cause of the Accident

The accident was caused as helicopter entered into uncontrolled flight during its final approach, because of undesired aggressive nose up maneuver coupled with full lowering of the collective by Pilot Flying, which resulted in steep autorotative descent at near zero speed leading to impact with sea.

AAIB identified the following contributory factors in the accident:

  • Lack of familiarization and proficiency on use of the aircraft’s AFCS, which led to inappropriate handling of controls; and
  • Absence of monitoring by Pilot Monitoring due to involvement with switching-off of weather radar during crucial phase of flight leading to delayed take-over in emergency.

AAIB raised 19 safety recommendations to address issues discussed in our summary and other, generally more incidental issues:

  • Four were directed to Pawan Hans: one on management suitability (discussed in ‘Our Safety Observations’ below), two on flight ops training standards and one on FDM.
  • Nine to DGAC: on guidance to operators, regulations on pilots qualifications for a new type, increased regulator oversight, CAP 3400 updates and increasing flexibility when inspectors haven’t been type trained.
  • Five to ONGC: on lifejacket and briefing card standards, offshore equipment serviceability and emergency response capability.
  • One to Sikorsky: to revise FDR Data Frame Document SER-76040335 in relation to correcting radio altitude data

Our Safety Observations on this Accident

The change in the plan on crewing resulted in a big risk increase.  While there is often a focus on the type experience of individual pilots, in this case the type experience across the organisation and of introducing new types in particular was low.  Ironically, if the more modern and more widespread AW139 had been selected by Pawan Hans some of the entry into service difficulties would have been lessened as that type was in service in India already so there would have been more experienced personnel available both to the operator and the regulator. 

Covid was clearly a further a challenge.  While the elongated eight month period after the import of the aircraft did give time to reconsider the risk assessment and refine the mitigations, it may have reduced the tolerance to risk due to the increased urgency of bringing these leased aircraft online and earning. AAIB notes that:

Since Pawan Hans had not inducted new aircraft in its fleet in recent years, it was facing difficulty in renewing existing contracts or bidding for new contracts. The induction of Sikorsky S-76D was a major event for Pawan Hans and crucial to its profitability and market share.

The S-76Ds were being phased out of use in Thailand in favour of AW139 and they migrated to a region where they were still seen as an advancement on the aircraft in service locally.

The EGPWS Terrain Avoidance and Warning System (TAWS) gave just 7 seconds of warning.  Again EGPWS, as currently implemented, proved ineffective as it is optimised for fixed wing operations.  We understand that based on the flight data presented in the AAIB report, with the new offshore envelopes developed by UK CAA research (CAP1519: Offshore Helicopter TAWS Alert Envelopes), c 17 seconds warning would have been given (a 243% improvement).  However a possibility that is not explored by AAIB, is whether the EGPWS “PULL UP” audio warnings, which the pilot’s would not have encountered on their prior types, may have prompted incorrectly pulling up the collective and causing a droop in main rotor speed (though by that point impact was near inevitable).  

DGCA’s reluctance to engage with Pawan Hans until type training had been provided to their inspector will have undermined DGCA’s oversight and prevented prescient early consideration of higher risk elements of the entry into service or detection. 

While ONGC’s AS4 aviation standard is discussed by AAIB, the nature and effectiveness of their customer audits and oversight isn’t examined.  Issues with safety equipment, videos and briefings should all have been detectable.

There are surprisingly few international industry recommended practices in the oil and gas industry for effective response to helicopter accidents adjacent to installations.

Our Safety Observations on Future Introductions of New Types

This case study is particularly relevant as in December 2024 Pawan Hans announced that ONGC were contracting them to provide 4 HAL Dhruv NG helicopters, a civilian version of the Advanced Light Helicopter (ALH) Mk III, for offshore operations from 2025 on a 10 year contract.  

This will be the first use of the 5.3 t, 15.9 D-value Dhruv in the offshore energy sector.

The type’s development suffered various problems (including with a novel main gear box design that features just two stages of reduction with dual first stage inputs, and runs the flying controls inside the main rotor shaft for ballistic protection). It first flew in 1992 and some 400, mostly military, have now been built (UPDATE 11 February 2025: and accumulated c 400,000 hours). Aviation Safety Network record 13 fatal accidents and 18 aircraft written off in accidents.  The cause of recent accidents included airworthiness issues with Dhruv’s flying control rods necessitating a material change.

The DGCA India ‘reference date’ for determining the applicable certification standards for the Dhruv was 8 July 2000.  The DGAC adopted US Federal Aviation Administration (FAA) requirements as follows:  FAR-29 Amendment 1 through 47 as on 9 May 2001.

ONGC is a now member of IOGP.  IOGP have a recommended practice (IOGP 690/5-1) on certification standards for offshore rotorcraft, which includes in this case using those certified to US FAR-29 at Amendment 45 (issued 25 October 1999) or later.  Strictly the Dhruv meets that requirement, however DGCA India’s experience certifying domestically designed aircraft is limited.  The Dhruv certification was validated by the EU Aviation Safety Agency (EASA) in June 2023.  The resulting type certificate is however restricted to visual flight rules (VFR) operations, maximum fuel quantity of 50%, a minimum crew of two pilots and no passenger or cargo transport.  This suggests a number of concerns and/or open items.

Its most recent fatal accident occurred on 5 January 2025.  According to the Hindustan Times:

A detailed analysis by the Council of Scientific and Industrial Research-National Aerospace Laboratories (CSIR-NAL), Bengaluru, found that the device that malfunctioned compromised the ability of the pilots to control the helicopter’s motion, said the first official. It is most likely the swashplate assembly in the ALH’s transmission system, HT has learnt. 

UPDATE 12 February 2025: Transmission snag led to ALH crash: HAL chief

Hindustan Aeronautics Limited (HAL) chief DK Sunil on Tuesday said that a “swashplate fracture” caused the crash of a coast guard Dhruv advanced light helicopter (ALH)…

Operations of the type are yet to resume.

Safety Resources

The European Safety Promotion Network Rotorcraft (ESPN-R) has a helicopter safety discussion group on LinkedIn.  You may also find these Aerossurance articles of interest:

Aerossurance has extensive air safety, flight operations, survivability, SAR, airworthiness & certification, human factors, helidecks, aviation regulation and safety analysis experience.  For practical aviation advice you can trust, contact us at: enquiries@aerossurance.com

Follow us on LinkedIn, Blue Sky @aerossurance.bsky.social and Twitter @Aerossurance for our latest updates.