Why Aircraft Engines Quit

let's see an engine, this is a mazda cx-5 which in the omni automotive tax is called a cuv crossover or if you prefer a compact suv among the japanese manufacturers, mazda is a niche company and its thing is sports performance, definitely out of the ordinary. common this car. It has a very good six-speed manual automatic transmission which I really like anyway the engine is a 2.5 liter inline four cylinder of about 150 cubic inches and 184 horsepower, four valves per cylinder with double overhead camshafts and injection Direct fuel, non-ported, direct fired coils for ignition and variable valve timing. It has cable drive for throttle body control as car

engines

run.

It's not cutting edge, but it's pretty sophisticated. Mazda makes about one and a half million cars a year, which is small by Toyota and Honda standards, but still. Many

engines

have that number in mind, so if you look at this level of mechanical sophistication, it's logical to wonder how reliable this is. I mean, there are a lot of moving parts and electronic components. How often does it just turn on and off? What if he never does this? It never worked for you, it sure works for me. Now let's compare the Mazda engine to a typical

aircraft

engine, so this costs about 40 grand on the hook.

This is a Continental i or 550n. This is possibly the best or one of the best piston engines overall. Today's aviation has a soft reduction in this application, which is a cirrus sr-22 and

quit

e efficient as an

aircraft

engine. It's a 310-horsepower, 550-cubic-inch, air-cooled, horizontally opposed six-cylinder engine, making it 70 percent more horsepower in three and half the displacement compared to the mazda engine, we only look at large cylinders with a diameter of 5.2 inches and pistons the size of beer kegs, two measly valves per cylinder, but what they lack is a number that they compensate for with a tendency to seize and burn, this is a pushrod engine in case you're not familiar, pushrods first appeared on buicks in 1904 and haven't changed much since then, the 550 has port fuel injection, not direct injection, and the continental It has a pretty good engine.

The system for that also has a throttle by cable, a large steel one that goes directly from the middle of the pilot to a mechanical throttle body, there is no mass air sensor here, but it has an air filter. Ignition is carried out by a pair of magnets. Magnetos were invented by Bosch in 1897 and aren't much different today, although it took Bendix a few years to figure out how you could really irritate owners by requiring an inspection every 500 hours, and by the way, this engine has a while between 2200 hour overhauls you might get as much service out of one it could even last 3000 hours the one behind me and the cirrus lasted 2400 hours or it could be tanked at 900 hours with soft cylinders or rotating bearings either way you would It will cost over 40,000 dollars for an overhaul or about 10 thousand more than the cost of a new Mazda CX-5 in the entire known universe there are about 225,000 engines like this flying between continental and lycoming less than 10,000 new engines per year made plus a bunch of revisions, keep those numbers in mind too, so there's nothing wrong with old technology, the less complicated it is, the simpler it is, the less it can break, that means an airplane engine should be much more reliable than that of a car. and you'd want it to be that way for reasons I'm sure I don't have to explain, but is an airplane engine actually more reliable than a car engine?

It's a little difficult to answer, although you know I'm going to try, but we can see why they abandoned in the first place define abandon for my purposes abandon means that the plane loses thrust and the fight ends either as an accident or a reportable incident this is it a resignation or much closer to home this or even much closer to home this is our mooney that ended up in a muddy swamp in south carolina engine failures for various reasons started all these accidents but how often does this happen? It's not an everyday occurrence, but engine failures kill and injure people every year and damage or destroy many airplanes.

The good news is that both fatal accidents and accidents in general in aviation have been on a downward trend in recent years. 10 years, but it seems we have reached a plateau. One thing that hasn't changed much is that mechanical problems are things that break. Airplanes account for about 15 percent of all accidents each year, but only about one percent of fatal accidents. It's a small number but not exactly zero. Nor is it statistically insignificant nor is the fact that of those 15 percent of accidents are caused by mechanical or maintenance problems. More than half (about 60 percent) are the result of power plant problems, if not total failures, then they are serious enough to cause an accident that bites the wrists a little more, but still not zero and not all engine faults make it to the database. the pup's engine shutdown a few weeks ago probably won't be included in the data nor is it necessary, we don't know how many engine failures are never reported, it's certainly part of the imaginary world he might have lived in when controlled dangerous substances were all the rage .

You could crunch the numbers and get a failure rate based on hours, but the flight hour data available is not very accurate for this report. I rely on the raw ntsb accident data, which themselves are somewhat incomplete reports, often lacking detail and sometimes reaching no conclusions. and you'll probably be wrong sometimes, but you play the hand you're dealt, so we'll have to do some estimating, so for this exercise I'm using 2018. That year there were 1,224 accidents in total, of which 117 are listed as plant failures of energy according to the null data which is a little more than two per week using the FAA's approximate trick on flight hours, really approximate, which is equivalent to 0.45 engine accidents for every hundred thousand flight hours, which is a little more than half the fatal rate, but one tenth of general aviation's overall rate. accident rate Another way to look at this is to survey owners and ask them how often their engines shut down.

NASA did this same type of survey in 2001 after they postulated this wonder. The current reliability of the complex GA aircraft systems is unknown. No joke, a lot of data on airline-level equipment, but almost nothing on general aviation piston aircraft, so domestic researchers built a statistical model based on an imaginary six-hour flight, used this complicated formula to shuffle the things and calculated some probabilities, extremely low probabilities, as it turned out in this The list of things that break in airplane engines is third out of six, not very good, maybe if you are flying over the Rocky Mountains at night with a stable norm, but the odds are pretty good, so why do they abandon what causes the engines to suddenly stop?

It can be an unsettling surprise, but a third of the time no one knows. 29 percent of engine shutdowns have no demonstrable cause. The accident report explains what happened but does not say why. There seem to be several reasons for this. The most important thing is that in many cases. In many accidents, the plane lands, the investigator shows up, and the engine starts and runs fine, thus frustrating the hopes and prayers of the hapless pilot who desperately wanted to be vindicated for a tank cylinder or a fractured fuel line. this plane the venerable cessna 150 whose continental o200 is famous for suffering from carburetor icing so the plane lands in a muddy field after the engine shuts down the still hot engine melts the ice and as if by magic the plane starts immediately ice ferry crashes again more than one Some of these have happened with an onboard instructor who forgot to use carburetor heat is a good way to turn a practice emergency landing into a real one.

Also, a lot of this research is pretty minimal and that leads to an interesting story about our lunar retreat. In this video I explained that the engine shut down while taking off from an airport in South Carolina and one of my colleagues landed in an assault swamp with no injuries, the NTSB did not send an investigator, which is not uncommon for accidents without injuries to occur. Instead, the FAA sent a As designated representative of air capability, he dragged the plane out of the swamp, folded the propeller with a lever, and started the engine without problems, a couple of months later, after we cashed the check. six-figure insurance, the plane appeared in pieces.

On eBay I actually called the guy who bought it and asked him about the engine. I'm pretty sure it was an amplifier. Hey, yes, the good engine runs very poorly. Too bad about your accident. Further investigation revealed that this was not an isolated incident. I found other accidents, mainly Moonies and Piper. arrows using the same lycoming i or 360 with a similar pattern that made me wonder why insurance companies didn't investigate them as part of a loss prevention program. Well, they don't, at least not very often, one executive explained it to me this way. If your claim was for a hundred thousand dollars and that's about the same as ours, we could spend a third of that amount trying and not succeeding and figuring out why the engine

quit

, so if the insurer is doing their quarterly numbers, everything gets lost in the wash, it's all about benjamins and that's just the harsh reality, then there's the ntsb, a lot of these investigations are too superficial and don't produce the usable data that they should, the safety board is sending their investigators to the FA motor school to improve this, but they have a long way to go, the cause is unknown in a third of motor accidents and in my opinion it is unacceptable.

Going back to the pie chart here, for my purposes I'll group fuel depletion, fuel starvation, and fuel contamination into one category, even though they technically aren't. t engine failures, but they cause engine failures and they are all preventable. Together, they represent the largest known portion of engine failures, and most of them are aviation pilots' favorite stupid trick, running the plane out of gas just a decade ago, which used This will happen twice a year. week, but now it's only less than once a week, so we're getting better. About half of fuel-related engine shutdowns are due to poor management, for example selecting a tank that only has air and being surprised when the engine shuts down.

I'd like to say I've never done that. I would also like to say that I am dating Jennifer Lopez. Both are equally true. One of the reasons for poor management is this. It's the fuel valve setup on a mid 1970's cessna twin. It's easy to make a mistake. and selecting the wrong tank, another reason is panicking after flooding the engine and not being able to switch to a tank that has fuel before the glide crosses the planetary surface. Fuel contamination accounts for about five percent of engine shutdowns, most of them undetected water. in the tanks, watch this video I made last year on this topic and always remember tank sump, the fuel is always missing, it's also part of that five percent, that's what put this 421 in the grass, it was covered with jet, same advice, sump the tanks and smell the fuel if in doubt, use the method of dumping the suspected fuel on a paper towel and look for a faint straw-colored stain as the fuel evaporates, indicating the presence of jet a.

If the fuel is primarily avgas, this method may not work well. However, carbice accounts for six percent of all engine shutdowns, some engines are more susceptible than others, mainly continental ones, lyco means less, so you know what to do. Engines have heat in the carburetor for a reason, so use it when the engine is idling or at low power. make sure the link is in good shape and actually working, sometimes that wire has come loose and not connected to anything, now this is where things get serious, get rid of all that other stuff and both pieces of the pie What remains are structural flaws. and maintenance cause failures, the two are sometimes related first, structural failures when a connecting rod flies off the side of the county and the windshield is stained with oil, that's nature's way of telling you that you've had a failure in the engine.

Structural failures include broken crankshafts, fractured magnets. Burnt valves and camshafts or broken rocker armscrumbling cracked cylinders failed oil pumps and broken crankcases among a list that is not much longer than that, fortunately these are not common but neither are they numerically rare 17 represents about half a dozen per year year, but the actual number is probably a bit higher because not everything is reported, remember our doubts, why do large pieces fail well? Let's consider the crankshafts of an airplane engine, it is a large, heavy and robust part that also carries a lot of load, isn't it? Long ago, crankshaft failure appears to be caused by manufacturing defects in 2002, Lycoming had a multimillion-dollar crankshaft recall after a series of in-flight failures related to metallurgical problems.

Continental went through a series of these two crankshafts also failing due to propeller strikes. that go unreported and caused cracks that then failed in the shaft, none of this is common, but they happen in the typical engine overhaul shop these days, up to a third of all work done like propeller clearance inspections and For some shops it is About half of some crankshafts fail due to lubrication failures, either low oil level or blocked oil galleries. Camp axles also fail more commonly due to corrosion or lack of lubrication, causing spalling. These appear in accident reports from time to time and that produced this oddity of the ntsb features a real runway turn that didn't work is a cessna 172 returning to the runway after a partial engine failure caused by a worn cam the plane appears to be in a stall but obviously did not turn it landed hard and although everyone survived there were injuries this is what the cam looked like this is what a really chipped cam looks like and this is a set of valve lifters that were within the few hours of failing;

Fortunately, these are often caught before an accident occurs. through an oil filter inspection occasionally oil analysis or simply loss of power of the valves also of the tank due to poor fit and guides or inferior quality materials, leaded navigation gas does not help and lack of lead is not a valve lubricant when the connecting rods fail, the engine falls apart spectacularly, shedding parts and oil. This often happens because connecting rod bearings fail due to lack of lubrication, they rotate on the crank journal and simply separate, sometimes they are difficult to categorize, for example I found some reports of connecting rod bearing failures that were caused by insufficient torque of the bolts when the engine was overhauled, so it is maintenance or structural. failure, well, it's a notable number of failures that occurred after the engine had work on the cylinders, which, as any owner will tell you, is a routine occurrence in aircraft ownership, which seems to happen in these failures okay, let's just look at an engine and for that I'm at my favorite engine shop, Zephyr Engines in Zephyr Hills, Florida, let's start with a basic look at an airplane engine.

This turns out to be a small Continental O200 four-cylinder engine, but it's pretty typical of how airplane engines are built. Two great pieces cast with these heavy ones. -service bolts that hold the same thing together, the crankshaft goes in here, the cam goes up here and the cylinders bolt on from each direction and this is what the connecting rods look like and as you can see they are quite strong and when an engine is built, it is made from the crankshaft outwards and because this entire structure is lightweight and highly stressed, all torque values ​​on these fasteners are critical, especially the rod bolts and the case bolts, when an engine is opened partially to work with the cylinder, those rod bolts sometimes don't get retightened properly, especially on continental engines, and they end up breaking or loosening and that's a really loud, ugly failure.

If the case bolts are not properly tightened, the case halves begin to work, causing jitter and that can cause the crankshaft bearings to rotate in their mouths, resulting in rapid wear and progressive failure. It is common for the front bearing to fail first. This can produce a serious partial failure, not the engine-explosive thrill of a thrown rod cap, but the engine will eventually fall apart. While we're here, we take a look at the cylinders. Cylinder failures are not as common, but when they do break, they sometimes crack around the spark plug holes or crack radially where the cylinder head mates with the cylinder or sometimes at the base of the cylinder.

They also tend to be partial failures as the piston and rod stay together rather than blowing a hole in the hood and treating innocent bystanders to a shower of oil, now what type of engine fails most frequently continental or lycoming nor is it rotax . I was surprised, well, they let me explain. How I got this, we don't have reliable flight hour data, so we can't calculate a rate per flight hour that way, but we do know how many registered engines are in service, so this data represents failures per 1000 registered engines. The caveat here This data is based on small numbers, just 13 accidents in total over the six-year period we studied, so a few accidents may sway the conclusion one way or another.

I couldn't determine why Rotax is higher than the other two, but it may be a small numbers effect. Some of those Rotax accidents were fuel related, but there were also mechanical failures, but why do continental engines seem to have a higher failure rate than lycoming ones? I have a theory that the continental engine population leans more towards large displacement, high performance engines like the i or 550. The series and the cirrus that I showed you at the beginning of this video, those engines are more stressed and often run with the cylinders half way, this probably makes them more susceptible to failure.

Barely a month goes by without us reporting a spectacular fatal takeoff accident. leading to the conclusion that engine failures on takeoff are more deadly than other phases of flight, but is that really true? Let's analyze the pie chart again according to this accident. Sweep failures occur approximately equally on takeoff and cruise as a percentage of the entire engine. Failure accidents are in the mid 30th percentile range, but takeoff represents a smaller portion of the total flight time, so yes, per hour, the risk during takeoff is definitely higher, but what about the risk? of fatal accident? This is where it gets interesting for both takeoffs. and cruise engine failures resulted in fatalities in a third of accidents, so in the event of an occurrence, takeoffs are no more deadly than crews.

The telling thing is that engine failure on takeoff can be fatal for sure, but two-thirds of pilots. survive the same goes for cruise engine failures, two thirds also survive, but engine failures and crews should be able to survive much longer because you have more time and more options to find a place to park the plane, you can bend it a little , but Should you be able to survive either way, you should regularly practice both emergency landings outside of cruise flight and responding to a power outage immediately after takeoff. I've posted a couple of videos on this topic and the links are down in the description to sum it all up, we know.

All this about engine failures, at least half of them and probably more are caused by pilots or mechanics doing something wrong or something stupid out of nowhere. No mechanical failure is below 20 of the total engine failures, so it is a small fraction of one percent. of fatal accidents caused by mechanical failure in general, you can reduce them even further by doing a few things and I really shouldn't have to say this, but put enough gas in the plane check the tanks for water, never skip it put enough oil in the aircraft and consider using an oil analysis that can detect a failure trend before it is a failure.

The same applies to engine monitors. Modern ones record engine data that can be analyzed to detect trends. This is how airlines do it. They know how the fuel lines work and how much fuel is in the tanks. they don't really stand what the poh says, they use carburetor heat when you're not supposed to skip engine maintenance including mag inspections, and once every two months I like to take the cowlings off the plane and spend 15 minutes inspecting things. for loose fasteners, shaped hoses, leaks, that sort of thing, I once found an oil filter that hadn't been safety wired and was starting to unscrew and a loose alternator wire, it's worth looking for; otherwise, don't worry too much about an engine failure in the A long list of things that will kill you and blow you up, it's actually pretty low on the list, thanks to L.J Warren, here at Zephyr Engine, for letting me poke around in the pile of scrap metal Bravo, I'm Paul Bertorelli, thanks for watching.

Oh, and by the way, my Mazda is gone. yesterday the battery ran out