368 NTSB Accident Investigation Process with Jeff Guzzetti
01:03:03 Share

Have you ever wondered about the process the NTSB goes through to investigate an accident? Or about the differences between investigating a major accident, such as the Reagan National Midair, versus a smaller GA accident? If so, stick around because today we're talking with NBC analyst and former NTSB investigator Jeff Gazzetti.

Hello again, and welcome to Aviation News Talk, where we talk general aviation. My name is Max Trescott. I've been flying for 50 years. I'm the author of several books and the 2008 National Flight Instructor of the Year. And my mission is to help you become the safest possible pilot. Last week in episode 367, I gave an update on the midair collision at Reagan National Airport and also talked about the crash of a Learjet in Philadelphia. So if you didn't hear that episode, you may want to check it out at aviationnewstalk.com slash 367.

And if you're new to the show, welcome. Glad you found us. Now, if you would take a moment right now, go out on whatever app you're using and touch either the subscribe key or if you're using the Apple podcast app or Spotify, the follow key. So that next week's episode is downloaded for free.

And see if you know any of these people who've recently joined the club to support Aviation News Talk. They include one new mega supporter. These are the folks that donate $50 a month. Our thanks to Frank Kimmel. Other donors include David Tokopf, Asher Brand, Chuck Colley, Mike Blade, and Douglas Olson. And if you'd like to hear your name next week, take a moment right now, stop what you're doing, go on out to aviationnewstalk.com slash support, where you can sign up to support the show.

You can also support the show whenever you buy a new Lightspeed headset. When you do, Lightspeed will send a check to support Aviation News Talk, but only if you first go out to the web to a special URL that we've set up for you, which is aviationnewstalk.com slash lightspeed.

Now let me tell you about Jeff Gazzetti. Jeff spent 18 years at the NTSB, first as an investigator and later as the deputy director in charge of the 10 regional offices that manage GA investigations. Later, he was the director of FAA's Accident Investigation Division. Now here's our conversation with Jeff Gazzetti.

Jeff, welcome to the show. Great to see you again. It's good to be here, Max. Well, it's been a terrible week or two in aviation. I think a lot of people are interested more than ever in the process that the NTSB goes through to investigate an accident.

Let's start with Annex 13. Just what is that and how does that guide an investigation? Annex 13 is a document from ICAO, the International Civil Aviation Organization. It's been around since the 1940s or 50s. It's basically the playbook by which all countries across the world should play in the sandbox of aircraft accident investigation. It allows

It allows a country of manufacturer or state of design to participate in another country that one of their airplanes happened to crash in so that everyone shares information. And so there's a set of standards by which adequate aviation accident investigation can proceed. Yeah. And I guess that makes it easier for different countries to interact because they all kind of know what's expected from each other. Well,

Let's talk about some of the definitions. Talk a little bit about accidents, incidents, things like that. How are those defined? So ICAO has a definition for accident as well as incident and serious incident. And most of the countries adopt this, like the NTSB here in the United States pretty much adopted word for word the definition of an aircraft accident. And essentially, it's an event that occurs that's associated with the operation of the airplane.

which means if a catering truck happens to slam into a parked airliner, that doesn't count as an accident because it had nothing to do with the operation of the aircraft. So you have to have people on board and there has to be intent for flight. So if you meet all of those criteria and you have substantial damage and there's a definition of exactly what substantial damage is,

And it excludes things like wingtip damage or landing gear damage or an engine failure. All those are not counted as substantial damage. Substantial damage is mostly big. The metal, the aluminum is being bent and or serious injury. So even if you don't damage the aircraft, but there's a serious injury on board, then it's still an accident. A serious injury, most serious injuries, as per the definition, is due to a broken bone.

But again, they have exclusions. It doesn't include your nose. It doesn't include your fingers or your toes. But any other bone in your body breaks, if you break an ankle or break an arm due to turbulence or something, that's an accident. It's a serious injury, which means it's an accident that has to be tracked and investigated by an investigative authority.

An incident, the definition is anything other than an accident. So that's pretty easy. But then there's something called serious incidents, which is an incident in which the probability of an accident was substantial. That's all it says.

And there are specific reportable events that pilots or operators must report if it comes into the serious incident list in which an accident investigation authority may pick and choose and say, we want to investigate that. So I guess the definition becomes important when the NTSB is picking and choosing. Am I correct that they're going to investigate virtually all accidents, but just some set of incidents? And how do they choose?

That's correct. The NTSB has a congressional mandate. They must investigate and produce a report and determine cause on every single civil aircraft accident. Now,

Not military, not intelligence agencies, but everything else. Airliners, Cessna 172s. So if it meets the definition of an accident, the NTSB must investigate. Serious incidents, because they're so busy investigating accidents, they don't do too many incidents. They do pick and choose. They might do 30, 40 airline incidents investigations each year out of the hundreds of incident reports that they get.

And so the NTSB has to uses its own judgment to pick and choose the response to each and every accident. So they know that have to respond in some fashion.

But if you're flying your Cessna 172 and you have a hard landing and you bend the firewall, that's substantial damage. It's an accident you have to report. But the NTSB is not going to come out and examine your airplane. There'll be some investigator assigned to a field office that'll pick up the phone, collect some information, write up a one-page report, get the pilot statement, and boom, they're done with that.

On the other extreme is an investigation into a major airline accident involving multiple fatalities, like what occurred this past week in Washington, D.C.,

Well, again, the NTSB has to respond. They could respond with one investigator picking up the phone, but obviously that's not appropriate. So they'll launch a go team. They'll send 20 investigators to the accident site. They'll break up the investigation into groups and deep dive into every aspect of that. And the NTSB does this because they

If a major airliner crashes, Max, I mean, the public trust has been violated. People paid money to be flown safely in a scheduled air carrier in a system that's supposed to have lots of redundancy to protect paying passengers, to protect this transportation system we have. And so they're going to really, they're going to deep dive into why did this happen? It shouldn't have happened.

That's not to say that your little 172 hard landing with firewall damage isn't important, but you know, it's a private pilot. They're not flying for hire. Uh, no one was killed or injured. It's a data point that needs to be looked at and investigated, but at much lower level than a, an air carrier, an airline accident involving fatalities. Let's,

Let's touch briefly on a couple of other definitions. Talk about state and why that's important. And then let's talk about investigator in charge and accredited representatives. Sure. So a state in ICAO Annex 13, it says states have a right to participate in certain investigations. So for example, the 737 MAX accident that happened in Indonesia, as well as Ethiopia,

The United States was the state or the country of design of the 737. They were also the country of manufacture, the state of manufacture. So the state of occurrence for that first 737 accident in Indonesia, the state of occurrence usually leads the investigation. So the Indonesian aircraft accident investigation entity says,

led the investigation into that 737 MAX in Indonesia. But the United States had a right to participate. And so...

United States was the state of design and the state of manufacture of the Boeing, a U.S. company, 737 MAX. So ICAO Annex 13 says that the United States can plug into the Indonesian investigation. And the way they do that is they elect an accredited representative from the state. In this case, it has to come from the Accident Investigation Authority. So the NTSB sent

an accredited representative as representing the United States and attached to her. Her name was Lorenda Ward. She's still at the NTSB. Attached, supporting her was someone from Boeing, people from the FAA, people from Pratt & Whitney, all of the U.S. entities involved.

Visited Indonesia under as a team, a U.S. team led by the accredited representative. Now, that accredited representative plugged into the investigator in charge, which is the state of occurrence, the state where the accident occurs. That agency leads the investigation. And in any country, there needs to be one person to lead the investigation personally.

And that person is called the investigator in charge. So there was an investigator in charge in Indonesia. In fact, I've worked with him before. He's very capable. And the U.S. team became a party to his investigation. State of registry is important. You probably wonder, why do they care about who registers the airplane?

That comes into play if you lose an airplane in international waters, right? So there's no state of occurrence. There's no sovereign territory. Who leads that investigation? Well, ICAO Annex 13 says the state that registered the airplane leads the investigation.

So if United Airlines 767 crashes 100 miles off the coast of New York, the NTSB is in charge. Or even if that same airplane crashed 100 miles off the coast of France, France doesn't lead the investigation. The United States does, just by virtue of the state of registry over international waters. Talk about the relationship between the NTSB and for FAA, both for major accidents and for smaller GA accidents.

Certainly. So they're two separate agencies. The FAA is massive, right? 45,000 employees. The vast majority are air traffic controllers. But the NTSB utilizes their FISDO inspectors, air safety inspectors from the FAA. There's about 3,800 of them spread across 100 field offices across the country.

The NTSB only has 50, five zero investigators to investigate these 1200 GA accidents each year. So obviously they can't travel to each one. The NTSB does not travel to non-fatal accidents. They will travel to every single fatal general aviation accident, one person, one field investigator. But for all the rest of them, the NTSB relies on the FAA to send one or two FISDO inspectors

to be their eyes and ears on the accident site. The NTSB is still in charge. They're on the phone, they're writing up the report, but they're getting their photographs and their on-scene information from one of those local 3,600 FAA inspectors out there. So it's really a symbiotic relationship. However, both agencies are very different, right? So you have the NTSB with a total

400 employees, I believe, total. And only about 120 of those do aviation. The rest do marine accidents, highway accidents, railroad accidents. So 120, 125 aviation investigators.

The FAA is a totally separate agency. It comes under the Department of Transportation, and it has, like I said, 45,000. It's in charge of regulating aviation. It can require airlines and pilots to do things, obviously. The NTSB cannot. It's a small, independent agency, and the only thing it can do

is make recommendations to the FAA to require things. Now, the NTSB is in charge of the investigation. Their report is the official one, not the FAA's.

But the FAA does their internal investigation to try to clean up their own house if it's involved with the investigation. But at times, the NTSB and the FAA can come to loggerheads, NTSB making recommendations for cockpit video recorders. And the FAA says, well, thank you very much, NTSB. But guess what? We have to consider rulemaking. We have to consider how much this is going to cost the airline industry. We have to consider all these things.

And so they may disagree with the NTSB and not adopt recommendations. But they got a pretty good track record, about 85%.

of all recommendations made to the FAA by the NTSB have been adopted. And that's pretty good in my views. It sounds like percentage-wise on small accidents, the FAA has a bigger role, spends a greater percentage of the total time, less so in major accidents. Talk about the respective roles for a major accident, such as the Reagan National Midair Collision. Sure.

So for a major accident like that, the NTSB is going to be much more hands-on and they're going to utilize their specialists and engines and in structures and in human performance and systems. And they're going to break up the investigation into bite-sized chunks because you're going to deep dive. The public expects them to, uh,

to have a pretty full story about what happened so that they can develop a probable cause and make recommendations to prevent the next accident. So the NTSB is going to launch what they call their GO team. And they're out of Washington, D.C. So obviously in this most recent accident, they didn't have to travel far. But the GO team is on call, ready to launch within a two-hour time period. They're put on an on-call list. Investigators are on call, and that list changes every week.

And if you're on call and you get a phone call or you get a text, you got to be on your way to the accident site within two hours. And you have to have your go bag packed. You could be traveling from Washington, D.C. to Los Angeles, or you could be going to Indonesia. So, you know, they got to be ready. They're trained for this. They signed up for this. So the GO team launches several investigators. And then behind them,

are supporting players, FAA. You might have FAA test pilots. You'll have more than just a FISDO inspector there. You're going to have specialists from all over the FAA's immense aviation research facility. You may have air medical people. You'll have FAA flight test pilots. You'll have FAA certification engineers.

You'll have FAA safety people from Washington, D.C., air traffic control specialists, all assisting the NTSV. And then not only that, you're going to have manufacturer experts working

assist in the investigation. In this case, with the collision over the Potomac River, I think there's 10 different entities that are assisting. You're going to have PSA Airlines. The airline itself will have a flight safety department. You're going to have Bombardier, who designs and builds the CRJ-700. Now, guess what? They're from Canada. They're going to be plugged in through that

ICAO Annex 13. There'll be a Canadian accredited representative from the Canadian Transportation Safety Board. And then in tow will be Bombardier and Transport Canada, which is their FAA. They'll be playing in this. You may have engine manufacturers involved. You'll have the FAA, of course. You could have the airport authority. So everyone plays a subordinate role at

as a party to the NTSB's investigation. The NTSB conducts this orchestra of different parties and different disciplines, but in the end, the NTSB is in charge of developing the final report and determining the official probable cause. So you might have 100 people

uh, investigating a major airline accident. Yeah. I think this is going to be unprecedented in terms of the scale of resources that are going to be thrown at it. I read that the accident, which happened around 6 PM that resulted in the FAA delivering a package of materials by 3 AM to the NTSB. That sounds pretty unusual to get things to them that quickly. What kinds of things would you expect to see in that package and in subsequent days from the FAA? Yeah.

Sure. So I used to work, I used to lead the office at the FAA that would provide those packages to the NTSB. It's not that unusual. The FAA has this information. There's no reason why they can't immediately share this factual information to the NTSB. It's usually air traffic control recordings.

So they can even play that over a conference call. They probably had the controller talking to this helicopter, you know, two hours after the accident. The NTSB probably listened to that recording. They'll have radar data. They'll have all kinds of air traffic control information, who was working on the shift.

what airplanes were coming in before and after. Because again, the FAA registers these airplanes. So they have probably had the FAA registry information. So there's just any, anything and everything the FAA has in its database, they will attempt to gather and quickly turn over to the NTSB. Now, during the days leading up to that,

The NTSB may ask for other things. They may ask for internal FAA memorandum describing the helicopter routes, one over the last two or three years. They may ask for legal rulings inside the FAA about helicopter flights. They may ask for how often PSA Airlines underwent operational surveillance by the FAA over the last year or so. They're certainly going to want to interview controllers

at the Washington Reagan National Airport Tower, and they'll ask the FAA to, the NTSB will put in a request to the FAA saying, we want to interview this controller. We'd like to do it tomorrow, sooner than later. And so the FAA communicates with the controller and the controller's union, and they set up the interview so that the NTSB can interview the controller. So

So those are the kinds of things that the NTSB will request of the FAA. So let's talk a little bit more about day one. You've mentioned the GO team, which is headed out there. I'm kind of curious when they get there. What are the first things they do and other things they do later in the day? And I'm guessing things are a little different with securing the scene on a river. So the NTSB is not, it's the operator that's required to secure the scene until the NTSB arrives. And then when the NTSB arrives, their first order of business is,

is to meet the incident commander, the local law enforcement or search and rescue. The incident command system in this country is a system by which an emergency responder has a leader, right? Could be the fire chief, and that fire chief may be relieved by another fire chief as the shifts are. So there's one incident commander that's in charge of

this disaster scene that the NTSB investigator in charge will meet and say, who's the incident commander? Hello, my name's Jeff Gazzetti. I'm the investigator in charge of the NTSB's GO team. Whenever you're done saving lives, whenever you say the scene is safe, we would like to take over. We, the NTSB, would like to take over. And in fact, please help us secure the scene. So that's their first order of business is securing the scene. They're also going to

Reach out to all of the potential parties to the investigation. So they'll call PSA Airlines and say, please lock down all of your maintenance records for this airplane. They'll speak to the U.S. Army to say, who can we talk? Who is your point of contact at the U.S. Army that was operating this helicopter? We would like to involve them. So they'll get to meet and introduce themselves to all of the experts that they need.

The first thing they do is it's called an organizational meeting. It's a big meeting. It's formatted. The NTSB leads it. All the parties come into a big ballroom in a hotel. So the NTSB will rent out a ballroom or someplace where all the parties can come in and the NTSB wants them all to introduce themselves.

They'll kick out any media. They'll kick out any attorneys. They just want the experts from all the different parties, FAA, airline, airframe manufacturer, engine manufacturer, foreign authorities.

And then they'll say, okay, this is what we're doing. We're going to break up our investigation. We're going to have an engines group. We're going to have a survival factors group, a human performance group, a air traffic control group, a flight data recorder group, a cockpit voice recorder group. And they'll put an NTSB person in charge of each of those groups. That's the go team members. And then all the parties will fill in. There'll be one representative from each party on each of those specialist group groups.

So the first day they're organizing. And the second day they'll probably begin to examine the wreckage. And this is whether it's on the ground or in the water, quite frankly, if it's in deep water, like I was the systems group chairman for the Alaska Airlines flight 261 investigation, which by the way, I attended the memorial service. So it was exactly 25 years ago last week. It happened yesterday.

the Memorial was two days after this tragic accident in Washington, DC. So it was really, uh,

But I remember that airplane crashed in the Pacific Ocean, right? And 200 feet of water. I was the systems group chairman. I had no wreckage to look at the first couple of days. So I just went as part of the go team and started meeting the people that would be helping me examine the systems of the wreckage, ALPA engineers, Boeing engineers. And then we just waited on shore until the U.S. Navy began to bring up parts of the wreckage.

If that same airplane would have crashed on the ground, we would have had quicker access to the wreckage. But in the end, we organized just the same on that first day. And the second day, we just continued to get more organized and we requested radar data and looked at things like that.

So organization is one of the first things you do. One of the second things you do is attempt to extract the flight recorders, the cockpit voice recorder and the flight data recorder, because it has information on there. And the sooner you can get to that information, the quicker you can make a potential decision to ground the fleet or to make a safety recommendation.

So those are some of the first things that the GO team does. For folks who don't recall the Alaska accident, just briefly talk about the jack screw and the cause of that particular accident.

Yeah, so that was an MD-80 operated by Alaska Airlines. And in January 31st, 2000, it took off out of Puerto Vallarta, Mexico. It was heading to San Francisco. And after takeoff, the pitch, the horizontal stabilizer jammed and the crew needed more and more force to keep the airplane going. So they adjusted their airspeed. They decided not to turn back to go to Puerto Vallarta. The airplane was in complete control.

They waited until they got near Los Angeles to make a landing in Los Angeles.

Uh, but then the, the horizontal stabilizer moved a bit, caused a, an initial dive. They pulled out of the dive and it declared an emergency. And then about eight minutes later, that horizontal stabilizer completely came out. There was a jack screw, uh, that, that came out. The threads of the nut were completely sheared out because it hadn't been lubricated by the airline in several months. And so, uh,

It just fell in an uncontrolled fashion from 18,000 feet into the surface of Pacific Ocean, killing all 88 people on board, sparking a major NTSB investigation. I worked that investigation full time for three years. And in the end, the probable cause was Alaska Airlines failure to adequately lubricate the jack screw.

and also check for the wear over the years leading up to it. And as a contributing factor, the FAA allowed Alaska Airlines to extend their lubrication interval and extend their wear check interval to the point where it just became meaningless at that point. And the jack screw wore out. It was allowed to wear out. And also Boeing was cited

for having a single point failure, the jack screw. If the jack screw fails, you lose the airplane and that's not allowed nowadays in design. Yeah, single points of failure are definitely bad. Let's talk a little bit about parties again for a moment. What are some of the responsibilities of parties and how can parties get kicked out from the investigation? So yeah, so first of all, the NTSB has to invite you as a party. They have to invite Boeing, they have to invite the engine manufacturer or the flight attendance union.

they're not automatically in the only entity that's automatically a party to every single investigation is the FAA. Congress decrees that the FAA is, has their fingers in everything. So, uh, they, Congress says, FAA, you're going to be joined at the hip with the NTSB. The other parties are at the behest of the NTSB. So when a party, and it's good to be a party because that party can understand on a day-to-day basis, what may have gone wrong with their product.

But a party has to be able to provide technical expertise to the NTSB, and they should be in a position to, if needed, to enact an immediate safety change without having to wait for NTSB or FAA. So the parties bring a technical expertise, and they get granted access, these parties, to the day-to-day findings of the investigation, things the public does not get access to.

And they can bring this back to the factory and look into it and assist the NTSB. But they have to play by rules. Each person that participates in an NTSB investigation has to sign a form. It's called the party form. There are 17 pages of restrictions in that form. They're not allowed to talk to the news media.

Only the NTSB can. And that's the number one reason, by the way, parties get kicked off is because they have press conferences. They put out information without asking the NTSB and the NTSB says you're confusing the public. There needs to be one voice for the investigation and that's the NTSB. But anyway, other restrictions include you can't withhold anything from the NTSB, right? If you see a problem in your organization that may have led to the accident, you must immediately reveal that to the NTSB.

You have to comply with NTSB direction on the investigation, and you have to keep information. I want to say it's kind of like a cone of silence, basically. You're learning information inside the yellow line, and you're not allowed to just tell your wife or your husband or your friends about what you're learning. You're just taking direction from the NTSB. The NTSB can kick you off as an individual,

They could say, hey, Boeing, John Smith from Boeing, you're not playing well. You're not playing nice in the sandbox. Boeing, you find someone else to take this person's place. Or the NTSB can kick off an entire company. Like, for example, 10 years ago, maybe it was 15, there was a midair collision over the Hudson River involving a helicopter and a small airplane.

The controller that should have been warning the airplanes from hitting each other was on the phone with his girlfriend talking about a barbecue or something.

And it looked bad. But NATCA, the controllers union, they had a press conference like the second day after the accident. And the NTSB gave them a warning, said, don't do this again. And then the next day, NATCA just did the same thing again, put out a press release saying it wasn't our controller. So the NTSB put out a press release saying, you're gone. You have violated our rules.

Hey, aviation world, this is what happens if a party to one of our investigations speaks out of line and talks to the news media without going through us first. So that's one example. There's other examples. And the NTSB likes to publicly broadcast this sanction.

so that others understand that they're serious about the integrity of having one voice for the investigation. Talk about some of the unseen hazards that investigators may have to deal with and what they do to assure their own safety when they're at an accident site. Accident investigators are trained to keep themselves safe. For example, there could be biohazards on the accident site, right? Human remains that could have

Infectious diseases there. So all accident investigators have to be trained with protecting themselves against biohazards. We call them poopy suits. They're basically Tyvek suits and rubber gloves and goggles. And if an investigator is going to go into any area that has human remains, they must wear protective gear.

That's just one aspect, biohazards. Other aspects include sharp edges, right? You got aluminum that's torn. So investigators should wear leather gloves. They should wear overalls that have material covering their arms and legs. Their skin should not be exposed because you're going to have sharp objects everywhere with torn aluminum. And so you got biohazards, you've got sharps.

Probably one of the biggest hazards is remote accident sites or your own personal health, hiking up a mountain and not having the health to be able to handle that, not having enough water with you and becoming dehydrated, not protecting yourself from the hot sun as you're investigating wreckage on a hot tarmac in Florida, for example. So those are the three main classifications of hazards, biohazards, sharp materials, and

and just the environmental hazards of the sun and lack of water and those kinds of things. Investigators receive training in these areas, and they follow the procedures pretty well. Yeah, this was an area that I was not really well aware of until I started to take some of the courses at USC that you and others teach down there. One other class of hazards, which totally came out of left field and is actually kind of new, had to do with carbon fiber. Talk about the danger of that at sites these days.

Yeah, carbon fiber. I guess that should be classified as a fourth category because it's, first of all, it can be sharp, but it can also be toxic. So carbon fiber, if there's a fire and they burn, it can release small little fragments of carbon fiber. And if you breathe those in, it can do a lot of damage to your lungs. And so you should wear respirators that filter out those carbon areas.

You can even use floor wax to keep those fibrous particles down. Sometimes investigators will use floor wax and coat...

the area of the burned composites so that you don't get those fibers coming up into the air. But you might have a carbon fiber, even if it's not burned, just kind of sticking out of the ground. These little fibers are very strong and they can pierce through a boot and get into your bone and your flesh. So they're very, it's very hazardous. So there's a special set of safety precautions to handle working around destroyed and burned composite materials.

Yeah, it's interesting how with new technology, we come up with new hazards as well. You started to talk a little bit about the black boxes, CVRs and flight data recorders. Talk about the process from the beginning to the end of how you locate those boxes and what you do with them throughout the entire investigation. Sure. If an airplane crashes on land...

investigators usually try to search for the black boxes are usually in the tail, the tail cone of most aircraft. If it crashes in the water, there's actually a locator beacon that's water activated, attached to the black box that gives out a signal so that you can use acoustic equipment underneath the water to find the black boxes. Because you'll hear this, the battery allows the acoustic signal to beep

for like 30 days. Regardless, you're going to require help to try to find the black box. Now, as you know, black boxes aren't black. They're painted orange, reflective orange, so they're easy to spot. But sometimes they are black due to the fire and the soot that covers them. But regardless, there'll be an effort to attempt to find those boxes. Those boxes are extremely robust, by the way, and they have to meet certain standards. They can survive nightly

massive temperature fireballs, things like that. They can survive massive G-forces in an impact.

One of the things they can't survive is sitting in a slow cooking for more than a day or two, smoldering, because that gives time for the heat to seep through the insulation and eventually melt the parts inside. But other than that, black boxes, the data usually survives if the airline maintained it properly because it needs maintenance once a year, I believe. So

investigators will attempt to find the black boxes, and there's usually two of them, a cockpit voice recorder, which records the sounds in the cockpit, as well as the pilot's intercom and air traffic control transcripts. And then there's the flight data recorder,

which is wired to all the sensors in the airplane to determine the exact position of the ailerons and the pitch control and the elevator and the rudder and the heading and the altitude. Several different times a second, the sampling rate is. So there's just immense amount of engineering data from a flight data recorder. Sometimes these recorders are what they call combined, where the CVR and the FDR are in one unit.

Sometimes the two separate boxes kind of in the same area of the tail cone. Investigators will determine exactly what the aircraft was equipped with.

Large airplanes, especially airliners, always have a CBR and FDR, cockpit voice recorder and a flight data recorder. Smaller airplanes like the business jet, the Learjet that crashed in Philadelphia last week, jets like that of that size are not required to have flight data recorders. And there was no flight data recorder on that Learjet, but there was a cockpit voice recorder. It's an easier, simpler, more inexpensive way to wire up aircraft.

an airplane. A flight data recorder requires a lot of wiring and sensors, and that's why it's really only reserved for larger passenger-carrying aircraft of a certain size. Anyway, once the NTSB finds these recorders, they carefully extract them from the wreckage,

And they are transported under escort as quick as possible to the NTSB's laboratories in Washington, D.C. They have three labs. Actually, they have four. They have a materials lab to look at broken metal. But then they have a cockpit voice recorder lab, which is a soundproof room with a big screen where you can do transcriptions. They have a flight data recorder lab.

which is a room with, again, a big screen where they can download the data and chart it to see it on a chart. And then there's the video recorder lab because they might get surveillance video from an airport or they might have someone sitting in the cockpit that has an iPhone taking a video of the approach. And they will do what they need to do to download the iPhone or whatever to get the video images.

These are all engineers that work at the NTSB. They're trained to carefully extract the ones and zeros, the electronic data from these black boxes. And they got to be careful because if they don't use the proper procedures, they can end up erasing the recording. So these are very skilled people that know what they're doing. Let's talk about the two cases last week. The Potomac River, the CRJ and the helicopter both had black boxes.

FDR and CVR, but they were waterlogged. They landed in the water. So what the NTSB does is they take them out, they flush them with distilled water, and then they kind of bake them in an autoclave to let it properly dry.

Because when they attach the wires to the black box to download the data, it's got to be dry and it's got to be a clean, clear connection. So they methodically do that. The cockpit voice recorder in Philadelphia, they found that thing at the bottom of an eight-foot crater where the airplane came into from.

So they extracted that. Now, one of the potential problems with a super high impact is it might damage the electronic connections on the black box. So it's difficult for the NTSB to plug in. So they may need just more time, might take a few days, might take a few weeks working with the black box manufacturer to gin up a better connection to better download that data.

Once you download the ones and zeros, they put it into algorithms and computer programs, and they begin to generate plots versus time versus altitude, anything that could be helpful to them. That's the flight data recorder. Cockpit voice recorder, they will pick, the NTSB will form a cockpit voice recorder group with one NTSB cockpit voice recorder specialist, someone from the FAA, someone from the airframe manufacturer, someone from the airline. They'll sequester themselves in a soundproof room

They'll put earphones on and they will listen to every utterance in there and they will bang out a transcript. That transcript eventually will become public several months down the road. But the actual recording, and this is important, the actual recording will never be heard by anyone else other than the people in that cockpit voice recorder room.

That was the agreement between industry and FAA and NTSB about protecting the sanctity of the voices and the sounds in a cockpit voice recorder. ALPA, pilots unions and airlines didn't want

Those recordings to be played on the evening news. And so it's protected by the NTSB, very, a lot of protection for the actual recording. But eventually the transcript will be released when the majority of the factual information is ready to go. So the people that are in the room listening to that, is that strictly NTSB people or some of the parties to the investigation? It's one NTSB person and then one representative who,

And they've got to be qualified. It's got to be a pilot that understands the communications and the cockpit noises from each of the parties. So like in the case of this most recent accident, I'm sure there'll be a pilot from PSA Airlines in the room. There'll be an FAA person in the room that's a pilot.

There will be probably someone from the U.S. Army because they're a party to the investigation. So there'll be one representative from each of the parties to get together and build consensus on hammering out a transcript of every utterance

of every second of the final approach of that airplane. Regarding the Learjet crash in Philadelphia, I had heard that although there wasn't a flight data recorder on board, that there was some hope they might be able to get data from a ground proximity device. What possible information might they get from that? Boy, that's a good question. In fact, a lot of Cessna 172s and Pipers have these types of boxes. You

Your Garmin 1000, for example, these avionics units are not crash protected. Okay. So they're not crash protected like a flight recorder, but oftentimes you can still get data off of them. They have what they call non-volatile memory on them, NVM. And I guess an easy analogy to that is when you go into your car and you turn on the ignition and you want to play your favorite station or turn to another station, your radio has stored in its memory, the different, your favorite radio stations.

Well, that's similar to your Garmin 1000 or in this case, in the Learjet, EGPWS, Enhanced Ground Proximity Warning System. And for maintenance purposes and for tracking purposes in a normal operation, if an EGPWS sounds, it will record what altitude you are at, what airspeed you are at, how far above the ground were you. And so it captures a moment in time when the EGPWS was firing.

And so if the ground proximity warning system provided a warning as that Learjet was screaming out of the sky,

It could store a certain amount of parameters, airspeed, altitude, those types of things that investigators can use in lieu of a flight data recorder. We call it a poor man's flight data recorder, all of this information. And you can get a lot of information. Like with the Garmin 1000, if it survives, you can actually get the track of the 172 during its accident flight because it's recording GPS position and everything.

From that, you can derive airspeed and it's recording altitude. So these are like potentially flight data recorders. But again, they're not crash hardened. But many times you just need that chip, that one memory chip inside of these avionics. And if it survives, you go to the avionics manufacturer. They have a lab set up.

They remove the chip from the damaged unit and put it onto a good unit, and they power it up, and they can download the memory from that chip. So it sounds like they might not get a whole lot of data from that particular device. So, for example, if there was a loss of thrust in the Learjet, we probably have nothing that's going to kind of tell us how the power decreased and on what engine and things like that. That's correct. They probably won't do that. Although you bring up a good point about engines, probably not on this Learjet.

I'm not sure, but engines that are FADEC, full authority digital engine controller, they have the same non-volatile memory. So the FADEC pack is attached to the engine. And several times I've been involved in investigations where, in fact, an example is Swiss Air Flight 111, which is an MD-11 that crashed in Canada. There was a fire in the cockpit. The fire burned...

the connections between the cockpit and the cockpit voice recorder. So we did not get the last six minutes of the flight for FDR or CVR. However,

We found the circuit chips on the FADECs for all three engines. And in it did provide engine performance data because it's a FADEC. And so we were able to determine that all three engines were operating properly and it provided the last six minutes of altitude and airspeed. So FADEC is one. And I don't think the Lear had a FADEC on there. If they did, maybe it was so destroyed that

There was just no way for the NTSB even to identify it. They're always looking for things like that. Yeah, and I think that was a 1981 aircraft. So I'm guessing in general, you're going to have a lot less information with 40- and 50-year-old aircraft than 10- and 20-year-old aircraft. Right. Yep, I agree. Talk a little bit about some of the other devices that passengers might be carrying, iPhones and watches and video cameras. What kind of information do you end up finding from those?

I tell you, it's been a boon for investigators. It's got a double-edged sword, though, because in one case, you've got a lot more data, video data, inside of the airplane, or even external data with people filming the accident. But it takes time to...

analyze that data, to download it and to analyze it. And a lot of time. So it's just a waterfall of data that these investigators have to parse through. But I think the benefits from it far exceed the detriment of taking more work. So yeah, a picture's worth a thousand words. So if you've got good video inside the cockpit from someone that's filming in-flight fire or smoke in the

That's very helpful to the NTSB. An accident that I oversaw involved a fatal balloon accident. It was the Heart of Texas balloon. 15 people died when this balloon took off in instrument conditions, basically, and then impacted everybody.

High tension towers, electrical towers, and the canopy fell to the ground and 15 people died. It was very tragic. All those people had iPhones, right? And they were filming the whole thing, the whole balloon trip into the clouds and impacting the high tension towers and even their fault or their deaths. There must have been 15 devices in the NTSB when they showed up.

They immediately tried to find these devices and keep them charged, right? And then they downloaded everything they could. And it basically helped solve the accident because, again, it chronicled this balloon flight into the clouds and towards the high-tension towers. So they're basically mini flight data and cockpit video recorders. And they also track...

Even if you don't have video, they kind of track their location, right? They leave little breadcrumbs of where you were flying when you were in the balloon, for example. So they're extremely handy. Yeah, and I've heard of two GA crashes now where the accident notification occurred because someone's iPhone's car crash notification let authorities know that, hey, something happened at high impact here. And that's really kind of fascinating that now we've got instant notification of accidents.

Yes. Yeah, it is. I wear a watch that I just bought, one of these Apple Series 10 watches. And if I happen to take a fall and spill onto the pavement...

It'll alert 911 automatically. It's just pretty amazing. Let's get back to the Reagan National Midair Collision. Could you give us a rough timeline over the next year or two or whatever of all the major milestones that we would kind of expect for this investigation? Sure. So in accordance with ICAO Annex 13 and NTSB's own procedures, the NTSB is going to come out with a preliminary report within the next two

Two weeks, probably. Usually they put out a preliminary report within 30 days of the accident. It's going to just be factual. It's not going to give a cause, but it will verify the basic facts of the investigation, the what, where, when. And so the public should be looking for that to be posted. Then you probably won't hear a lot more from the NTSB for a while. They're going to be busy conducting their investigation.

They will probably have for this one, what they call a public, an investigative hearing, a public hearing where maybe four or five, six months from now in Washington, D.C., they're going to bring in witnesses from the army and from FAA and they'll ask them questions under oath and the public can attend. And it's a way for them to get more information out for the investigation, as well as raise awareness of what the NTSB is doing in the public eye.

then probably about a year, year and a half from now, and it's going to take that long, the NTSB is going to release all the factual information that they have on this accident. Each of those group chairmen that are on the GO team I talked about, they're going to be tasked with writing a

cockpit voice recorder report, an engine report, a structures report. All that's going to be made public. And then a couple of months later, the NTSB will have a final board meeting. So this is going to happen again, probably about a year and a half from now. And in that board meeting, the members of the board will be receiving presentations from NTSB staff, and they will publicly proclaim the probable cause of the accident and the recommendations that

And then a report, a final report of the whole thing will come out shortly thereafter. So it's a long process, a year and a half, two years for a major accident like this. But it's going to answer all the questions that need to be answered about what happened here. So it's unusual to have an accident of this size and scale happen in our nation's capital at an airport that lawmakers fly out of. How do you think that might affect the investigation in any way, shape or form?

I just think that investigators might be a little more on pins and needles. They're going to make sure they go thoroughly by the book with this investigation. It's the higher-ups, the NTSB chair and the board member inmate that's the spokesperson on this.

Those are the ones are going to have to be very careful with regards to what they tell Congress. But it will be advantageous for the investigation because no stoner is going to be left unturned because there is such congressional interest. And it did happen in Congress's backyard. But by the same token, I don't think the NTSB is going to be shy about commenting on

why National Airport has become so busy over the last few decades. It's because these congressmen have added more and more and more nonstop flights out of National Airport so that they can get to their constituent locations more quickly, right? So,

The NTSB is apolitical and, you know, they're just going to try to shy away from the politics and stick straight to the facts. They've done that before. I worked the JFK Jr. investigation. That was highly visible. President Bill Clinton was actively involved in what was happening in that investigation at the time. There's been other investigations. Senator Wellstone was a U.S. senator that died in a twin engine turboprop in Minnesota.

About 20 years ago. So the NTSB is used to navigating the turbulent political waters to get to the facts and put out a great report. I was thinking about one other hazard to investigators that we haven't talked about, and that's mental health. It's got to really be tough for people to go out year after year to accident scenes and see the kinds of things that they're seeing at accident scenes.

Talk a little bit about that. How does the agency work with that? How do individual investigators deal with that psychological trauma? So, Max, it's a good question. I think the more accidents NTSB investigators work, the better they're able to deal with the issue. Now, every now and then you might come across an accident with little kids. That's the worst. When you know the children have died.

But regardless, after every accident and during the accident, the NTSB has an employee assistance program, an EAP program, where they monitor and just chat with and get a debrief from the NTSB investigators that are working these accidents.

So investigators usually find out pretty in pretty short order after the first couple of accidents, they work, whether or not they're cut out for the position. And it's, you know, it's no nothing bad about people that say after the first couple of accidents, I don't want to do this anymore. I don't want to look at dead bodies. This is too much for me. That's fine.

Other investigators kind of get over that hump pretty quickly. I did, and most of the people I, you know, I kind of had my own way of relating to human remains. In my mind's eye, I just said they were mannequins that you see in a shopping center in the wall, or I wouldn't look directly at them. I'd just keep them in my peripheral vision.

And I would talk to other investigators about what I saw. The peer-to-peer investigator discussions after an investigation is really one of the major ways that investigators deal with this kind of potential post-traumatic stress. But for the most part, it's the people's love of accident investigation that overcomes their fear of trauma and overcomes

Over the past couple of decades, NTSB and FAA have gotten better and better at taking care of their own investigators in the wake of every accident. You just raised a question in my mind. For people who have been listening and are interested in the investigative process, what

What kind of opportunities are there for them to join the NTSB? How do they find out about those opportunities and what kind of qualifications is the agency looking for from people? There are opportunities at the NTSB. Every year they hire a handful of summer interns. They might not be paid, but it's really good experience. And they work side by side with investigators. They might help type up reports or observe accident sites.

The NTSB is usually always hiring in some fashion because, you know, there's people that retire and it depends on what position they're hiring for. So to be a general aviation accident investigator, a field investigator that does mostly small aircraft, you have to be a pilot and preferably an instrument rated pilot.

And then you usually have to have some other skill. Maybe it's a degree in aviation safety. Maybe it's experience with a company investigating accidents. That's how I kind of got into it. I worked for the Cessna Aircraft Company as an accident investigator before I was hired by the NTSB.

So getting some sort of accident or incident investigation experience, even from the periphery of the investigation is important. Having an aviation degree of some sort and being a pilot is very helpful. Now for the go team,

If you were a former air traffic controller or a former airline pilot or a former psychologist or a former meteorologist, well, then you might be hired for your specialty, for your expertise. So there's different ways to get in there. But the most important thing that I would say is a lot of colleges and universities now have aviation safety programs where you can at least get a degree and learn about the theory and

And then the second thing is to become a pilot, at least a private pilot with an instrument rating. And then the third best thing is to start getting experience as doing investigations. The MTSP is very small and it can be very competitive.

And sometimes you'd have to work in Washington, D.C. or places where you don't want to live. But they put out on their website their job openings all the time. So I would encourage frequently checking the NTSB's website for these job opportunities. I'm going to ask you in a moment about USC, but I'm wondering, is there anything else that we haven't talked about that you'd like to mention?

No, I guess the only thing I would warn people is don't jump to conclusions about what you think caused the accident. There's red herrings in every investigation where initial piece of evidence comes up and you think there it is. There it is. You know, but I got to tell you, it's usually there's usually a lot more involved.

And these investigations, any accident really has layers of root cause. So don't be fooled by just making an assumption about what caused an accident. There's usually a lot more data that you're not aware of that the NTSB needs to get before they determine what the actual verified, well-thought-out probable cause is.

Yeah, and I think you or perhaps one of the other instructors at USC had said that there is often an issue where investigators look at an accident and they tend to instantly jump to a similar accident that they've seen in the past. So even investigators can kind of fall into that trap. Absolutely. We have to fight. Probably that was the toughest thing, and it's still an evolution for me, is to have the discipline not to...

impart previous accident investigations onto the current one or not to have a bias in some fashion. So it takes some mental discipline to try to remain objective and open to other potential issues. You and I met about a year ago at USC where you were an instructor and I was in your class there. Talk a little bit about the program at USC. University of Southern California offers short courses in aviation safety.

And it's really a long story tradition. Most of the astronauts and Air Force officers were trained at USC back in the 50s and the 60s. The school has been around for decades and they offer certificates if you attend enough of these short courses. So I teach the first two days of their two week basic accident investigator course, but they offer probably 20 other courses. I also teach accident photography,

I teach safety management systems. I teach emergency response planning. So USC offers these courses. They're open to anybody and they're held in the Los Angeles area. They have a crash lab near Alhambra. So that's where you would stay. But I've been teaching for them ever since I retired from the FAA. Most of their instructors have a lot of experience working for FAA and NTSB.

And I've been teaching there since 2019, since I retired six years ago. And I love it. The school is constantly upgrading their curriculum to meet the needs of the students. It's got an excellent reputation. It's got support from the University of Southern California. It's underneath their engineering department.

And I think they're one of the best short course schools out there. Yeah. And I had gone down to attend what I thought would be just the two week long aircraft accident investigation course. But after taking that course, I decided, oh, I need to do more of this. And I went on to complete the certificate program. So I also took the helicopter accident investigation course. I took the human factors course.

And then I also took the SMS safety management systems course, which I highly recommend to anybody who's involved in safety within their organization. So yeah, great, great program they have at USC. So I wanted to ask you just a little bit about you and your work. I know that obviously you appear on NBC from time to time. Tell us about your day-to-day work that you do.

Sure. So since I retired six years ago, after 32 years of government service, I'm basically a one-man band. I'm a consultant. I'm an aviation safety consultant. But quite frankly, what I mostly do is teach. I teach at USC, some of the same courses that you took. I also teach at the Transportation Safety Institute, TSI, which is basically the FAA Academy in Oklahoma City. That's

that FISDO inspectors receive some training on how to be an accident investigator. Because again, as we've discussed earlier, 3,600 FISDO inspectors across the country that are the eyes and ears of the NTSB for these small GA accidents, they receive a two-week training

training down in Oklahoma City to get the lay of the land of that. So I teach and I also consult with a few corporations, flight business jet operations on how to improve their safety, or I might answer their question if they have a fender bender. But basically, it's just I just do some part time aviation safety consulting.

My gig, if you can call it that with NBC News, is just an exclusive agreement. I'm not salaried by them, but if there's an accident, I'm their aviation analyst, one of their aviation analysts. The other one is John Cox, who also teaches at USC. He's a former airline pilot. And, you know, we might appear a dozen times.

over the course of a year, on average once a month. It's not a full-time job and you can't make a living doing that. But I mostly do teaching and I do safety surveys of small operators. And what is your website where people can find out more about you? It's

GazzettiAviation.com. All one word, my last name, Aviation.com. Jeff, thanks so much for joining us here today. I've really enjoyed spending the time with you here again. Likewise, Max. I always enjoy trying to educate the public about the great work that the NTSB and the FAA does to keep everyone safe.

And my thanks to Jeff Gazzetti for joining us here today. You can find out more about Jeff and his work at Gazzetti, that's G-U-Z-Z-E-T-T-I, aviation.com.

And just a reminder that I love hearing from you and I read many of your emails on the show. If you'd like to send me a message, just go out to aviationnewstalk.com, click on contact at the top of the page. That's absolutely the best way to send me a message. And of course, I also want to thank everyone who supports the show in one of the following ways. We love it when you join the club and sign up at aviationnewstalk.com slash support to

To support the show financially, you can also do that at aviationnewstalk.com slash PayPal. We also love it when you leave a five-star review on whatever app that you're listening to us on now. And of course, if you're in the market for a headset, please consider buying a Lightspeed headset and using one of the links in our show notes, because if you use those links, they will donate to help support the show. So until next time, fly safely, have fun, and keep the blue side up. And remember that you can always go around. Go around.

♪ Coming down ♪ ♪ Your silence baby ♪ ♪ Sliding upside down ♪