Category: Aviation

The Evolution of Air Traffic Control

Until civil air traffic started to grow the need for its control wasn’t the number one consideration. The pilot was the master of the skies. A basic “see and avoid” approach was taken. See another aircraft and avoid it at all costs. Note, I am talking about the early 1920s.

If you want a nice exploration of how it all started keep an eye on the site of the Croydon Airport Visitor Centre[1]. The first London airport was not Heathrow or Gatwick. No, there’s a stretch of grass, a hotel, industrial units and out of town shopping standing on the site in Croydon of the first London airport. 

Firstly, we can thank Marconi for the first radiotelephony. Providing a means for pilots to speak to airports enabled the development of Air Traffic Control (ATC)[2]. It got going out of necessity because there was limited space on the ground and many aircraft wanted to take-off and land.

Aerial navigation took off in the 1920s. A hundred years ago. WWII drove advancement in every aspect of technology. After WWII, the basic having been established, an international body was established to set standards for international flying. That’s where today’s ICAO originated.

Radar and VHF radio transmissions were the cutting-edge technology that enabled air traffic to grow. Radio navigation aids developed as did automatic landing systems. So, by the time the jet-age started there was a whole selection of technology available to manage air traffic. Not only that but the standards required for these systems to interoperate around the globe were put down on paper.

That legacy has served aviation remarkably well. Incremental changes have been made as new capabilities have been developed. Most notable of that evolution is to return elements of control to the cockpit. A traffic alert and collision avoidance system (TCAS) does just that. It provides a safety net.

What we have available to manage dense airspace and busy airports is a complex, highly interconnected, interdependent set of systems of systems and procedures that is not easy to unravel. Each part, in each phase of flight, plays its role in assuring safe operations.

News and rumours are that quick fixes are being demanded in the US. Responding to recent accidents and a perception that all the above in antiquated, a well know tech guru has been thrown at the “problem”. I shouldn’t be a cynic, as having a fresh pair of eyes looking at the next steps in the development of air traffic management should be good – shouldn’t it?

It’s my observation, as an engineer who knows a thing or two about these things, is that any simple solution means that the parties have not thought long enough about the problem. In this case there are no quick fixes. However, there’s likely to be incremental improvements and they will not come cheap. 

[1] https://www.historiccroydonairport.org.uk/opening-hours/

[2] https://www.historiccroydonairport.org.uk/interesting-topics/air-traffic-control/

Challenges Facing Supersonic Flight

Congratulations go to “Boom” for their supersonic jet flight[1]. Civil aerospace hasn’t ventured into this space for some time. Breaking the sound barrier is not an everyday occurrence in the civil world. There may be an international market for such new aircraft as much as there’s a market for fast cars and expensive boats.

However, I do not think a supersonic flight is the future of civil aerospace. It’s not mainstream. The environmental objectives for the future of aviation are ambitious. Generally, that means getting people from A to B in as clean and efficient a manner as is feasible. That does not include going ever faster and faster.

This new aircraft type is likely to be solely made in America. So, it does fit with the current political direction of the administration in the US. A triumph of technology. President Trump’s instinct to get rid of rules and regulations may work in the favour of Boom. However, in the end, the deciding factor will be – will the international marketplace want such a new aircraft type?

I certainly recall amazing ambition of the people who brought us the Eclipse aircraft[2]. Small light jets were going to be everywhere. Like a Silicon Valley revolution for the aerospace industries. That didn’t happen as predicted because the economics didn’t stack up. I don’t recall rules and regulations being the problem.

Even so, BOOM technology will have a hard job meeting international safety and environmental standards. I seem to remember that’s not new for supersonic flight. Even if the advancements made improve noise performance, there’s emissions and contrails to ponder.

There is another consideration too. It’s the problem Advanced Air Mobility (AAM) is facing now. To capitalise on their capabilities, these aircraft technologies require the reorganisation (modernisation) of national airspace. Plus, agreement at international level[3].

Supersonic flight over the world’s oceans may get agreement. Supersonic flight over national territory is a much harder sell. Some fliers may pay to slashing their travel times on-route. Going round and round in a stack, waiting to land, with conventional aircraft all around, will soon dispel any excitement.

Good luck to Boom. If civil use is minimal, no doubt defence applications will be numerous.

[1] https://boomsupersonic.com/

[2] https://www.eclipse.aero/about/

[3] https://www.icao.int/environmental-protection/Pages/default.aspx

About Animals and Flying

Pigs do fly[1]. But only the more privileged ones. Yes, animals that fly are not restricted to those with their own wings. It’s true that the animal kingdom has been showing us how to fly long before powered flight took-off. Nothing more graceful than a bird of pray swooping and diving. We (humans) can’t match much of what they do with our flying machines however hard we try.

Birds long inspired great thinkers. They opened the prospect of human flight. If they can do it – why can’t we? Surely the right combination of aerodynamic structures and a source of power would solve the problem. Shocking, in a way, that it wasn’t until a couple of keen bicycle repair men and a smart mechanic persisted until they had a working machine. That was only just over a hundred years back.

So, today’s novelty News item[2] of a cat that didn’t want to leave an aircraft puts a smile on my morning face. For all the farm cats I have known, the story doesn’t surprise me at all. It’s the sort of situation where humans are almost powerless in the face of the preferences of a feline.

Naturally, the engineering staff of an airline will have a good look at where the cat has been in its wanderings. There’s always the remote chance for a rogue moggy to play with something they shouldn’t ought to play with. Even on a modern Boeing 737.

I used the word “remote” but there are definite cases of loose animals causing air safety hazards. Looking this one up, because it sits vaguely in my memory, I do recall a dog that crewed through electrical cables after it got free in a cargo hold. Now, however lovable and cuddly a dog maybe that’s a place that no one wants to be in.

Back in 2002, American Airlines Flight 282 approached New York’s JFK. It was a Boeing 757 that landed with chewed-up electrical cables. Crew members heard noises coming from the cargo hold and found that some aircraft radio and navigational equipment wasn’t working. A dog had chewed its way through a cargo bulkhead and attacked wires in an electronics compartment. 

A quick search reveals that there are more cases of incidents caused by loose animals than might first be thought. Animals are potentially hazardous cargo. Sadly, often these flight incidents are not good for the animals concerned.

One thing to remember is that a large aircraft, at flight altitude, is pressurised. That’s not at the air pressure on the ground (unless an airport is a long way up a mountain range). A dog with breathing difficulties is going to find an aircraft environment distressing. Dogs can be skillful escape artists. Myself, I’m not keen to share a flight with them.

[1] https://intradco-global.com/livestock-transport/

[2] https://www.thesun.co.uk/news/33273791/cat-causes-chaos-ryanair-plane-rome/

The Swiss Cheese Model in Aviation Safety

Models in safety thinking take different shapes and forms. A conversation might start – what’s a model? Why are they useful?

Here’s a go at an answer. It’s always risky to explain why something works. It can be like a dry analysis of the particulars of a good joke. That kills the essence. As the words attributed to Albert Einstein say: if you can’t explain it simply then you don’t understand it well enough. Even if that’s not literally a quote it sums-up the need for simplicity.

Aviation is a highly complex, interconnected, socio-technical system with a legacy that coexists with rapid advancement. There are few parts of the globe that are not touched by aviation in some way or another. Getting to and from Arctic wastes, commuting between vast cities or traversing the widest oceans. Aviation touches all of them every day.

There is no piece of paper big enough to write a detailed description of every part of the worldwide aviation system. Even the most extensive computer simulations just take on a small part of the whole. I often use this phrase – “it’s more than a head full”. What I mean is that however smart we might think we are, the normal person can only comprehend a slice of what’s happening. A slice frozen in time.

We get over our limitations in perception and understanding but approximating. That is to carve out a “model” of what’s happening and how parts of a complex system interact. That sounds easy enough to construct. It’s a lot harder than first might be thought.

For one, a model needs to be sufficiently universal to capture an underlying reality or theme.

Next, a model needs to be useful. It has utility. It’s proven to work. To produce useful outcomes.

Thirdly, a model needs to communicate a message across cultures, beliefs and disciplines.

A model that meets all the needs described above can be as big an advancement as any hard technology. I guess it’s not surprising that a professor of psychology comes up with one that has been used and reused successfully over decades.

This week has seen the passing of Professor James T. Reason. He’s left us with a legacy that’s almost incomparable. His Swiss cheese model[1] has become a basic part of every aviation safety professional’s training.

I’ve debated and discussed accident causation a lot. The Swiss cheese model[2] is not the only way of thinking about how accidents happen, but it is an extremely good one. It promotes a way of thinking about how to defend against accidents. That’s powerful.

Like all models it’s a simplification of a highly complex system. Its great strength is that this model allows us to see through the mist. To see part of what is obscured by complexity. That is immensely valuable.

Thank you, Professor Reason. 

NOTE: An IFA Video with Professor Reason Every Day – 20 min film – International Federation of Airworthiness.

[1] https://en.wikipedia.org/wiki/Swiss_cheese_model

[2] https://www.eurocontrol.int/sites/default/files/library/017_Swiss_Cheese_Model.pdf

Future of Single Pilot Operations in Aviation

Flying embraces automation. Now, there’s a statement that didn’t ought to be controversial, but it can be. Even before we became engulfed by the modern digital age, analogue autopilots could assist in the task of flying. Some early ones were mechanical.

The need for full-time hands-on piloting of the physical controls that linked a human and an aircraft’s control surfaces is not fundamental. Large transport aircraft have stepped further, somewhat mimicking what their military counterparts did, and fly-by-wire systems have become commonplace.

As far as technological evolution is concerned, we remain in a transitionary phase. Commercial aircraft that fly overhead are a mixed community. Some, like the Boeing 737 series continue to have cables and pulleys that link aircraft systems and controls. Others, like the Airbus A320 series are the fly-by-wire digital aircraft types in regular service.

Between the pilots in the cockpit and the motion of an aircraft there is a computer. In fact, several computers arranged in a manner so that they continue to work even when subject to failures. A great deal of thought and effort has gone into designing aircraft systems that will be reliable in-service.

Looking at the safety numbers, starting in the 1980s when fly-by-wire was introduced, the overall service experience is extremely good. The practice of system safety assessment has delivered dependable and robust aircraft. Rigorous certification processes are applied. 

Through the technical developments that marched on from the 1980s one requirement has remained. That is that two pilots are needed in the aircraft cockpit. Granted there are exceptions to this rule for smaller transport aircraft. Single pilot operations are not new. For example, in many countries, the Cessna Caravan[1] is approved for a single pilot.

It’s 2025. It’s difficult not to notice the debate around Single Pilot Operations (SPO). That is to open large transport aircraft operations to a new rule. Lower operating costs may be achievable by making a change. It’s even said that this move is a way of continuing aviation’s growth as it becomes more and more difficult worldwide to increase the number of qualified pilots.

It’s good to see this subject being taken up in a forthcoming conference.

RAeS Flight Operations Conference 2025: Single Pilot Operations – Logical Progression or a Step Too Far?[2] 19 March 2025 – 20 March 2025. Royal Aeronautical Society Headquarters in London.

SPO may be enabled by use of complex systems to help make mission-critical decisions. The next step maybe with real-time “artificial” copilots and intelligent monitoring. Will this move the aviation industry toward safer and more efficient aircraft operations? That is the question.

[1] https://cessna.txtav.com/en/turboprop/caravan

[2] https://www.aerosociety.com/events-calendar/raes-flight-operations-conference-2025-single-pilot-operations-logical-progression-or-a-step-too-far

Investigating the Black Hawk and American Eagle Collision

What’s mysterious about the recent tragic collision between a US Army Sikorsky UH-60 Black Hawk helicopter and the American Eagle Flight 5342, was the failure of the normal procedure of “see and avoid” and the lack of an avoiding manoeuvre from the helicopter[1].

Taking the timings from reports of the investigators’ work so far, the air traffic controller’s instruction to the military helicopter to pass behind the commercial jet was seventeen seconds before the catastrophic collision impact. Given the trajectory of the commercial jet, as the pilots were focused on a landing, they had little possibility for an evasive manoeuvre other than a go-around. I imagine the commercial pilots and the tower controller reasonably assumed that the military helicopter would comply. In fact, why would they have any reason to question that assumption?

A question has arisen about night-vision goggles. Were the crew of the military helicopter using these devices? Night Vision Imaging System (NVIS) are not new[2]. They are used in both in military and commercial flying. There are a series of technical requirements that address their safe use. For commercial flying helicopters, that use such visual systems, they must additionally be equipped with a Terrain Avoidance and Warning System (TAWS). 

One of the down sides of night-vision systems are that the greatly enhanced capability can lead to overconfidence and potential misjudgements by pilots. When used by pilots these systems amplify ambient light and thus help pilots maintain visual references. That’s good for night flying over difficult terrain at low altitude. It’s not so good when there are multiple bright light sources all around, as there are in a big city.

I’m sure that the accident investigators will be giving the above subject a great deal of consideration. Afterall, the evening of this tragic accident was one of fine weather and fair visibility. The investigators have a significant task ahead analysing data and verifying the performance of both humans and machines in the accident situation.

NOTE 1: Worth a watch https://youtu.be/hlMTpIAlpw0

NOTE 2: Key safety system off in Army helicopter that collided with American Airlines jet, senator says | Reuters

NOTE 3: Night Flying “there are factors that can make it more challenging, like the lack of visual references and encountering visual illusions”. Flying into the Dark. What You Need to Fly at Night | by FAA Safety Briefing Magazine | Cleared for Takeoff | Jan, 2025 | Medium

[1] Evidence of a last-minute manoeuvre may still come to light. Sadly, the outcome remains the same.

[2] https://skybrary.aero/articles/night-vision-imaging-system-nvis

Advancements in Flight Recorder Technology and Regulations

My last posting addressed accident flight recorders and airworthiness requirements. That’s not enough. It’s important to note that aircraft equipage standards are addressed in operational rules. So, the airworthiness requirements define what an acceptable installation looks like but as to whether an operator needs to have specific equipage or not, that’s down to the operational rules in each country.

Internationally, the standards and recommended practices of ICAO Annex 6 are applicable. These cover the operation of aircraft. Flight recorders are addressed in para 6.3.1. and Appendix 8. Let’s note that ICAO is not a regulator. There are international standards but operational rules in each country apply to each country’s aircraft.

One of the major advances in accident flight recorders technology is the capability to record more data than was formerly practical. This has led to standards for Cockpit Voice Recorders (CVRs) advancing from 2-hour recording duration to 25-hours.

Proposed rule changes have been hampered by the impact of the global pandemic. Some new operational rules apply only to newly built aircraft. That means some existing aircraft can retain their 2-hour CVRs.

Another technology advance is what’s known as Recorder Independent Power Supply (RIPS). RIPS can provided power to the CVR for at least 10 minutes after aircraft electrical power is lost. The RIPS is often offered as a relatively straightforward aircraft modification.

I do not know if the South Korea Boeing 737-800 was required to have accident recorders with the capabilities listed above. If they were not, then there’s a good basis for recommending that changes be made to existing aircraft.

Understanding Aircraft Accident Recorders

There’s quite a bit of chatter on social media about accident flight recorders.

One of the skills required by an aircraft accident investigator, and not often mentioned, is the ability to grapple with rules, regulations, and technical requirements. This is given that civil aviation is one of the most highly regulated industries in the world.

The story of the development of the accident flight recorder is a long one. No way can a few words here do justice to all the efforts that has been made over decades to ensure that this vital tool for accident and incident investigation does what it’s intended to do.

In fact, that’s the first technical requirement to mention for accident recorders. Namely, FAR and CS Subpart F, 25.1301: Each item of installed equipment must be of a kind and design appropriate to its intended function. That basic intended function being to preserve a record of aircraft operational data post-accident.

Aircraft accident recorders are unusual. They are mentioned in the airworthiness requirements, however they play no part in the day-to-day airworthiness of an aircraft. The reality is more nuanced than that, but an aircraft can fly safely without working flight recorders.

FAR and CS 25.1457 refers to Cockpit Voice Recorders (CVR)[1] and 25.1459 refers to Flight Data Recorders[2]. Both CVR and FDR receive electrical power from the aircraft electrical bus that provides the maximum reliability for operation of the recorder without jeopardising service to essential or emergency electrical loads. Both CVR and FDR should remain powered for as long as possible without jeopardising aircraft emergency operations.

Before drawing too many conclusions, it’s important to look at the above certification requirements in relation to their amendment state at the time of type certification of an aircraft.

If the aircraft of interest is the Boeing 737-800 then the FAA Type Certification date is 13 March 1998 and the EASA / JAA Type Certification date is 9 April 1998. Without wading through all the detailed condition, the certification basis for the above aircraft type was FAR Part 25 Amendment 25-77 and JAR 25 Change 13 [Note: EASA did not exist at the time].

FAR and CS 25.1457 and 25.1459 were in an earlier state than that which is written above. That said, the objective of powering the recorders in a reliable way was still applicable. There was no requirement for the CVR or FDR to be powered by a battery. What hasn’t changed is the requirement for a means to stop a recorder and prevent erasure, within 10 minutes after a crash impact. That’s assuming that aircraft electrical power was still provided.

So, when it’s reported that the South Korea Boeing 737 accident recorders[3] are missing the final 4 minutes of recoding, the cause is likely to be the loss of the aircraft electrical buses or termination by automatic means or the removal of power via circuit breakers. We will need to wait to hear what is found as the on-going accident investigation progresses.

[1] https://www.ecfr.gov/current/title-14/section-25.1457

[2] https://www.ecfr.gov/current/title-14/section-25.1459

[3] https://www.bbc.co.uk/news/articles/cjr8dwd1rdno

Fatal Boeing 737 Crash in South Korea

Jeju Air Flight 7C2216, arriving from the Thai capital of Bangkok, at South Korea’s Muan Airport (MWX), crashed at around 9am local time (00:00 GMT/UTC) on Sunday, 29 December 2024.

My condolences to the families and loved ones of those who died or were injured in this fatal aircraft accident.

Pictures of the Jeju Air Boeing 737-800 landing[1] show that no landing gear can be seen deployed. A video image shows the aircraft skidding down the runway at high speed. The aircraft is wings level. It is reported the aircraft overrunning the runway and colliding with a wall or ramp. The video image does suggest that the aircraft engine thrust reversers were deployed. This is wrong. Weight on wheels is needed for deployment.

MWX runway 19 has a Landing Distance Available (LDA) of 2800 m. The local visibility was reported as 9000m and the wind speed at 2kt.

Was the pilot in command trying to go around? The accident flight recordings should answer this question. That is from the aircraft Flight Data Recorder (FDR) and Cockpit Voice Recorder (CVR).

This remains a hope. Reports are that the FDR has been damaged. This should not be a surprise given the nature of the impact it suffered. However, both FDR and CVR are designed and tested to survive extreme cases.

The South Korean Ministry of Land, Infrastructure and Transport says that the accident flight and voice recorders have been recovered[2].

Jeju Air is a popular South Korean low-cost airline. The airline was established in 2005.

A full independent accident investigation will no doubt take place. That is in accordance with the standards and recommended practices of ICAO Annex 13.

Current media speculation surrounding possible causes of this Boeing 737 accident do not offer any satisfactory explanation for the sequence of events. For example, it would be astonishing if the root cause of the accident was a bird strike or multiple bird strike shortly before landing. The aircraft has several means to deploy its main undercarriage.

It is likely that safety culture, controller and pilot training, and airport facilities are bigger factors in this fatal accident than the fact that it involved the loss of a Boeing 737-800 aircraft.

NOTE: Boeing 737 “If the gear fails to extend properly or hydraulic system A is lost, the gear can be manually extended by pulling the manual gear extension handles, located in the flight deck.” Landing Gear

POST: The impact test in the applicable technical standards EUROCAE ED55 (FDR) and ED56A (CVR) are demanding. The recorder’s crash protected memory module is fired out of a canon into a shaped target to simulate an accident scenario. It must be readable afterwards.

[1] https://www.independent.co.uk/tv/news/south-korea-jeju-air-crash-b2671085.html

[2] https://www.bbc.co.uk/news/live/c4glr85l2ldt

SONAR in Ocean Wreckage Recovery

Finding aircraft wreckage in the deep ocean is possible. However, it requires a degree of good fortune. Most of all, it requires the searcher to look in the right places. Lots of other factors come into play, particularly if the ocean floor is uneven or mountainous.

The primary tool for imaging the ocean floor is SONAR. That’s using the propagation of sound in water. SONAR can be of two types. One is called “passive” and the other called “active”.

The first case is like using a microphone to listen to what’s going on around. Of course, the device used is named appropriately: a hydrophone. It’s a device tuned to work in water and not air. Afterall, sound travels much faster in a liquid than it does in air.

Passive SONAR depends on the object of interest making a noise. Just like we have directional microphones so we can have directional hydrophones.

Passive SONAR is only useful if the aircraft wreckage is making a noise. Since in the case of Flight MH370, the battery powered underwater location beacons attached to the accident flight recorders have long since stopped working this kind of SONAR isn’t going to be much use.

Active SONAR is analogous to RADAR. That is where a pulse of high frequency sound is sent out through a body of water. Then sensitive hydrophones pick up a reflection of that pulse. It is detected and all sorts of miraculous digital signal processing is done with the acoustic signal, and an image is then formed. From that displayed image the human eye or sophisticated algorithms can make sense of what they are looking at on the sea floor.

Active SONAR can give both range and bearing (direction). Timing the sound pluses from their transmission to reception can give a way of calculating range. Or distance from the object providing a reflection. Bats know how to do this as they navigate the dark.

In sea water, there are complications. Sound does not always travel in a straight line in sea water. The speed of sound in water depends on salinity, temperature and pressure. All three of these factors can be measured and compensated for in the SONAR signal processing that I mentioned above. Helpfully at ocean depths beyond a kilometre the calculations become easier.

The average depth of the Indian Ocean is over 3 kilometres. It’s mountainous underwater too. So, what are the chances of finding flight MH370 on the ocean floor after 10-years[1]? This prospect goes back to my earlier comment. It requires the searcher to look in the right places.

Just imagine encountering the Grand Canyon for the first time. It’s nighttime. An important object is lost in the canyon. You only have the vaguest theories as to where the object has come to rest. With a handheld touch you go out to search. What are the chances of finding the object?

There are several factors that are in your favour. One, you know what the object might look like or, at least, in part. Two, the easy search locations (flat/smooth) may be covered relatively quickly. Three, certain areas of the rocky canyon have already been searched. Still the odds are against finding the lost object without a high degree of good fortune. 

I wish the new planned searchers much good future[2].

NOTE 1: one of my student apprentice projects was to design and build a Sing-Around Velocimeter for use in relatively shallow sea water[3]. It worked but was cumbersome in comparison with the simple throw away devices used for temperature depth profiling.

NOTE 2: To get down to the ocean depths required it’s a side-scan sonar that may be used. This active sonar system consists of a towed transducer array that can be set to work at different depths. Imaging objects on the seafloor and underwater terrain is done as a towed array moves slowly forward through the water. The scanning part is the acoustic beam sweeps left and right. Each scan builds up part of an image.

In operation, as the frequency of the sound in water goes up so does the resolution of a potential image but, at the same time, the range of the sonar system goes down. Thus, a sonar system used for surveying may have low and high frequency settings. Unlike sound in air, here high frequency means above 500kHz.

NOTE 3: What will an aircraft accident recorder look like after a decade in the deep ocean? It might have survived well given the nature of the dark cold pressured environment. This picture is of an accident recorder recovered from relatively shallow sea water (Swiss Air Flight 111).

POST: Nice view of what SONAR can do, at least in shallow water Bristol Beaufort wreckage found

[1] https://www.cbsnews.com/news/mh370-plane-malaysia-new-search/

[2] https://www.bbc.co.uk/news/articles/cewxnwe5d11o

[3] https://apps.dtic.mil/sti/tr/pdf/AD0805095.pdf