U of T news
  • Follow U of T News

U of T aerospace expert on investigation into fatal Ethiopian Airlines crash

Debris of the crashed airplane of Ethiopian Airlines, near Bishoftu, a town some 60 kilometres southeast of Addis Ababa, Ethiopia (photo by Michael Tewelde/AFP/Getty Images)

Investigators are just beginning to gather evidence to understand the causes of the Ethiopian Airlines crash that killed all 157 people on board, including 18 Canadians. It was the second Boeing 737 Max 8 – the latest version of Boeing’s best-selling aircraft series – involved in a deadly accident in just over four months. A Lion Air 737 Max 8 crashed in October killing 189 people. 

On Wednesday, Canada joined Germany, France, Britain, China and dozen of other countries that have either grounded or banned Boeing 737 Max 8s as a precaution. 

On Monday, before Canada's decision to ground the 737 Max 8s, U of T News spoke to Craig Steeves , a professor at the University of Toronto’s Institute for Aerospace Studies, about what we know about the crash and how investigators piece together the causes of an accident.

How unusual is this, two of the same type of aircraft – Boeing 737 Max 8s – crashing in the space of four months?

It's certainly surprising. There's some suspicion about why the Lion Air aircraft crashed. It doesn’t seem that anybody has any idea yet why the Ethiopian Airlines one crashed. 
In the Lion Air crash, what may have happened – and this isn’t conclusive yet – is that the angle of attack sensor stopped working and led to a cascade of problems. 

What does that mean? 

If a plane is flying perfectly level with respect to the incoming air, that’s zero angle of attack. To get lift, planes always have to fly at some angle of attack, tilted up slightly with respect to airflow. Increasing the angle of attack increases lift.

There's an angle that's not very large – 15 to 20 degrees – where the wings no longer generate more lift with higher angle of attack. That’s called stalling and it's bad. Several airplanes have crashed because they stalled. Colgan Air Flight 3407 in Buffalo and Air France 477 over the Atlantic Ocean, for example. In both cases, the angle of attack was too high and the aircraft stalled, but the pilots were still trying to increase the angle of attack more. Reflexively, if you want to go up, you point the aircraft up. But if the aircraft has stalled, pointing it up is not going to help. What you have to do is point the aircraft down, gain speed and get control of the aircraft again. Boeing has integrated a new system into the 737 Max that does this automatically.

So what are experts' suspicions about the Lion Air crash? 

These have been reported in the New York Times and the Guardian, and probably elsewhere. The suspicion is that the angle of attack sensor was faulty. It was incorrectly detecting an angle of attack that was too high and hence the automatic system reacted by trying to push the aircraft nose down when the aircraft was actually flying normally. The automatic system kept pointing the aircraft down and the pilot eventually lost control. At least, that is the current theory.

This automatic stall prevention system is integrated into an automatic system for trim control. For the pilot to turn this new system off they just have to turn off trim control. It's easy to do if the pilot knows that this is the cause of the problem.

Considering the suspicions about Lion Air, every 737 pilot in the world now knows that if the flight computer appears to be trying to crash the plane, turn off the trim control. Boeing sent out a notice reminding pilots of this. So it would be very hard to imagine the same thing happening again. Still, it's possible; we simply don't know yet.

What do we know about the Ethiopian Airlines crash? 

We hardly know anything. There are eyewitness reports that are contradictory. One eyewitness says the plane was on fire when it crashed. One eyewitness says it was not.

What will crash investigators be looking for at this stage?

The first thing they'll do is get as much perishable evidence as possible: statements from eyewitnesses, evidence from air traffic control, any radar plots that might exist.

The second thing they'll get is physical evidence on the ground. If it's a mechanical failure, you can often find the part of the aircraft that failed, eventually. If there was fire on board the aircraft, which can be catastrophic, investigators might be able to find evidence of that.

The aircraft apparently was on fire on the ground. It was just taking off so there was a lot of fuel on board, so this may make collecting physical evidence more difficult. But that's the kind of thing that an investigator would look for. 

The next thing are the flight recorders. The data recorder and the voice recorder.

The black boxes?

They're orange actually, but they get called black boxes. They're becoming more and more sophisticated and they record more things now. So it's possible to get a lot of information from them. 

There aren't a lot of maintenance records for this aircraft, I suppose, because it's so new, but certainly an investigator would look at that too. They would also probably interview people who knew the pilot.

The breadth of possible causes of crashes is enormous. In any crash that is related exclusively or primarily to the aircraft itself and what happened on board, a series of things have to interact in order to make that crash happen. It would very rarely be just one or even two things.

How long before we have answers? 

Sometimes never. For the Malaysian 777 that crashed in the Indian Ocean, flight 370, we still don't know why that happened.

Sometimes it can take a couple of years. If a plane is lost at sea it can take a very long time to establish what happened. In such a case even recovering the data recorders can take a long time. If there's dramatic destruction when the plane crashes on the ground – a high-speed crash and then fire – it can take a long time to piece together all the physical wreckage and figure out what happened. So, generally, many months or years before there is enough evidence to reach a firm conclusion.

Some countries and airlines have decided to ground Boeing 737 MAX 8s. Canada and the U.S. so far have not. Why is that?

That's really hard to know. One thing to remember is that the aerospace and the aircraft industries are probably the most political industries in the world. There could be a lot of considerations that are not obviously safety-related that we would not know about.

Given that two brand new aircraft were involved in broadly similar crashes in four months, it is possible to make a prudential argument for safety, saying that they have to be grounded until we know what's happened. The current evidence, though, seems to be very limited, and one could just as easily argue that we don't have any reason to ground these aircraft.

Tell me about this aircraft series.

The 737 has been around since the late 1960s. If you look up and see an aircraft flying overhead, it's probably either a 737 or an A320. They absolutely dominate commercial aviation because nearly all the two- to five-hour flights are flown with these planes, so that covers most flights that do not cross an ocean.

The more glamorous planes like the 787 or A380 are long-haul only. 

It's hard to know exactly what are the motivations of an aircraft manufacturer releasing a new plane, but the 737 Max could be seen as a response to the A320neo, and both of them may have been intended as a response to the Bombardier CSeries. The CSeries is a very, very efficient aircraft.

Neither Boeing nor Airbus wanted Bombardier poaching their customers.  Airbus improved the A320. “Neo” stands for new engine option. They put on a new engine, a geared turbofan, which is considerably more efficient than traditional engines. Boeing had to respond to this to remain competitive in this hugely important sector of the aviation market. They changed the engine on the 737 and improved the aerodynamics. Both Airbus and Boeing were aiming for better fuel efficiency and lower operating costs.

From an engineering perspective, what are the challenges when you re-engine an existing aircraft? 

Most new engines have very high bypass ratios. That means a lot less air goes through the core of the engine, where the fuel is burnt, relative to the amount of air that's going through the engine in total. This enables a quieter and more fuel efficient engine. But the engine has to be bigger.

If you look at a picture of a 747 from 1969 the engines look tiny. If you look at a new 747 now, the engines are huge in comparison. That's not because they're more powerful, but because they have a high bypass ratio.

For a bigger engine, the most obvious external change necessary is longer landing gear to fit the engines under the wings, but the larger engines also lead to changes in aerodynamics and handling, so these must be accounted for as well.

What do these events mean for Boeing? 

Regardless of what turns out to be the cause, it is bad news. Most importantly, a lot of people were killed, which is a tragedy.

Boeing's reputation will inevitably be hurt because these are brand new planes that are crashing. Plus, if regulators start grounding these airplanes, it's going to be a big problem for airlines. And if it's a problem for airlines it's a problem for Boeing.

In the medium-term it might affect their sales, although this will still be an enormously popular aircraft. But there are a lot of issues here: safety-related, aviation-related, regulatory and political. Untying all those things is really tricky.