United 108 IAD-MUC Engine failure at 1,000 feet on takeoff

[repairman54]

I was going to ask about this… I recall when growing up, once driving on the NJTP near EWR, in a light rain. All of a sudden the wipers got real streaky. Weren’t working right. My friend’s father, who was driving, claimed it was because the aircraft was dumping fuel.

Dousing cars on the turnpike in the rain seemed dubious, but the windshield was horribly streaky. So maybe it was true!?!

Too close to EWR for dumping. I lived off Exit 12 for a lot of years. Back then trucks leaked fuel like crazy since it was so cheap and mileage wasn't an issue like today. Turnpike would be always be slippery when it started to rain.

 

[I<3oil]

I am certain that Boeing and GE already know. I do know it wasn’t a huge deal (the engine stayed on the wing) because the airplane flew again the next day, and is still flying today.

I see

 

[I<3oil]

Malfunctioning FMU (fuel metering unit) in what I'm seeing on an A&P forum I frequent.

So basically, fuel was cutoff? Not actually anything mechanical with the engine then

 

[IndyFan]

I am certain that Boeing and GE already know. I do know it wasn’t a huge deal (the engine stayed on the wing) because the airplane flew again the next day, and is still flying today.

So basically, fuel was cutoff? Not actually anything mechanical with the engine then

It could be a variety of things. I'm no expert, but these engines are complicated. Reliable, but very complicated. Might have been a part that rarely fails but was easy and quick to diagnose, access, and replace? Who knows?

 

[y_p_w]

It could be a variety of things. I'm no expert, but these engines are complicated. Reliable, but very complicated. Might have been a part that rarely fails but was easy and quick to diagnose, access, and replace? Who knows?

Isn't this sort of thing supposed to be caught during routine maintenance/inspections? Of course that doesn't preclude the possibility of random failure that wasn't a result of normal wear that would indicate the possibility of failure before the next inspection.

 

[ammolab]

Isn't this sort of thing supposed to be caught during routine maintenance/inspections? Of course that doesn't preclude the possibility of random failure that wasn't a result of normal wear that would indicate the possibility of failure before the next inspection.

Inspections look for things that are defective….you can’t see some things that are working now, but will fail in the future.

 

[Astro14]

Isn't this sort of thing supposed to be caught during routine maintenance/inspections? Of course that doesn't preclude the possibility of random failure that wasn't a result of normal wear that would indicate the possibility of failure before the next inspection.

The engines, the entire airplane, gets inspected and checked before going ETOPS/International. Every flight. Every time. This airplane was no exception.

I once had a 767-300 going to London one night, we were 5 minutes from pushback when the cabin temperature got warm, and I mean really warm. We had started boarding 45 minutes prior, and everything was working fine. Long story short - we couldn't fix it, and had to get another airplane. I stood by the door as folks deplaned.

One upset passenger, a tall, British person from 1st class, said, in an imperious, condescending voice, "this should have been discovered 3 hours ago!" I looked at him and said, "The airplane was thoroughly checked hours ago, but this fault didn't appear until 5 minutes prior to departure."

"Unacceptable!" he roared.

The next guy said, with a smirk, "You mean that you can't predict the future?"

Exactly.

 

[Astro14]

So basically, fuel was cutoff? Not actually anything mechanical with the engine then

Not exactly - the fuel was being improperly metered into the engine. The engine temperature was too high, the thrust was too low, the whole thing wasn't running right, so, it had to be shut down. jet engines aren't exactly like cars - there isn't a good analog.

 

[y_p_w]

The engines, the entire airplane, gets inspected and checked before going ETOPS/International. Every flight. Every time. This airplane was no exception.

There are some things that are hard to detect without disassembly or at least ultrasonic inspection. But yeah I think we all appreciate the safety checks that go into modern aviation.

 

[IndyFan]

The engines, the entire airplane, gets inspected and checked before going ETOPS/International. Every flight. Every time. This airplane was no exception.

I once had a 767-300 going to London one night, we were 5 minutes from pushback when the cabin temperature got warm, and I mean really warm. We had started boarding 45 minutes prior, and everything was working fine. Long story short - we couldn't fix it, and had to get another airplane. I stood by the door as folks deplaned.

One upset passenger, a tall, British person from 1st class, said, in an imperious, condescending voice, "this should have been discovered 3 hours ago!" I looked at him and said, "The airplane was thoroughly checked hours ago, but this fault didn't appear until 5 minutes prior to departure."

"Unacceptable!" he roared.

The next guy said, with a smirk, "You mean that you can't predict the future?"

Exactly.

Too bad you can't tell that bloke to go stuff himself.

 

[I<3oil]

Not exactly - the fuel was being improperly metered into the engine. The engine temperature was too high, the thrust was too low, the whole thing wasn't running right, so, it had to be shut down. jet engines aren't exactly like cars - there isn't a good analog.

Gotcha, that makes more sense.

 

[Astro14]

Gotcha, that makes more sense.

The bypass ratio on the GeNX engine is like 9:1 - which is crazy high. Much higher than previous generations of airliner engines.

It enables the 787 to get really incredible fuel economy. The 787 carries about as many people as the 767, but flies about 6-8000 feet higher, about 40 MPH faster, and uses half the fuel. Yep, twice the economy, at higher speed and altitude.

Amazing engine.

But, the aerodynamic/pressure balance between N1 and N2 (the two spools, inside the engine) has to be maintained exactly - and that is harder to do as the bypass ratio goes up. So, while I don’t know exactly what went wrong with this engine - improper metering of the fuel could cause an upset in that balance, nothing wrong with the engine mechanically, but it was running too hot, and not making any thrust.

So, it failed, in that sense. It wasn’t making thrust, and the exact reason becomes important only after the airplane is on the ground.

With electronic checklists and prescribed procedures built into the jet, I suspect that the engine told the airplane it wasn’t able to be restarted in flight, or the “engine relight” checklist would have been offered after the engine was shutdown. An engine relight may have been part of the other checklists they ran, but again, the engine wasn’t capable of being relit with that failed component (kind of like an engine start isn’t possible if the fuel injection isn’t supplying fuel).

So, they shut it down and left it off. The longer an engine runs with abnormal parameters, particularly EGT, the more likely it is to be damaged. Heating isn’t even when the airflow though the engine is disrupted, so you can get “hot spots”, where the local temperature is higher than the recorded temperature, and those hot spots can damage turbine blades and other components.

The fact that the engine flew the next day tells me that the crew acted quickly and correctly. Their quick action prevented thermal damage to critical components.

 

[MolaKule]

The bypass ratio on the GeNX engine is like 9:1 - which is crazy high. Much higher than previous generations of airliner engines.

Amazing engine.

But, the aerodynamic/pressure balance between N1 and N2 (the two spools, inside the engine) has to be maintained exactly - and that is harder to do as the bypass ratio goes up. So, while I don’t know exactly what went wrong with this engine - improper metering of the fuel could cause an upset in that balance, nothing wrong with the engine mechanically, but it was running too hot, and not making any thrust.

So, it failed, in that sense. It wasn’t making thrust, and the exact reason becomes important only after the airplane is on the ground.

With electronic checklists and prescribed procedures built into the jet, I suspect that the engine told the airplane it wasn’t able to be restarted in flight, or the “engine relight” checklist would have been offered after the engine was shutdown. An engine relight may have been part of the other checklists they ran, but again, the engine wasn’t capable of being relit with that failed component (kind of like an engine start isn’t possible if the fuel injection isn’t supplying fuel).

Excellent engine operation description.

When working at Boeing with the 777 propulsion group, we went to the GE engine test facility to see how well our newly designed GE engine pylon, the second attempt for the GE engine, was holding up. More about that later.

During one test the engine was shut down abruptly. Speaking to the test engine facility techs, we found out one of the EGT sensors failed, giving a false high temp output. The EGT sensor was replaced, and the engine test resumed without incident. I am not saying this was the cause of this engine shutdown, but a sensor replacement is a fairly easy fix.

The original pylon we designed for the GE 90 engine was stress analyzed and showed some structural deficiencies. We redesigned it and reanalyzed it because the latest engine design that was on the test stand was heavier and had produced more thrust, greater than the original specs GE had given us.

The other two engine pylons, one for the Rolls-Royce Trent 800 and one for the Pratt and Whitney PW4000, passed without a hitch.

 

[Variant_S]

Great write up Astro! I remember as a kid living in a house which was in the final approach of runway 22L at JFK. IIRC that was the most used/or one of most used runways in the USA for quite a while. They would often drop fuel which we'd see floating in the swimming pool.

I lived about 12 miles due east of ORD in the 1960s. We'd sit in the backyard in the late afternoon and watch the planes on approach stacked up over the Lake Michigan. We sometimes felt a fine mist that smelled of some petroleum product.

 

[4WD]

Nice to read a story about competent experienced pilots.

100% - I often doubt that the results of lawsuits prove much when a good law firm - and a dumb jury (or threat of) are involved …

 

[The_Eric]

The airplane flew across the ocean the next day - plane spotters will tell you that - so, while the failure is still proprietary, I’m not at liberty to disclose, it wasn’t a big deal - no damage occurred to it, thanks to the quick actions of the crew in getting the power back and the engine shut down.

Why are some failures considered proprietary? To help stave off industrial espionage?

 

[The_Eric]

A true dead stick landing is an extraordinarily rare occurrence, happening, perhaps once every couple of decades. We don’t train to glide the airplane in. It’s just so unlikely, that there are much more likely things that are more deserving of our attention in training. Things like windshear, a single engine failure, flap/slat problems, short field landings, etc. that we train to extensively.

So, no, they weren’t holding where they could “glide” in because the airplane isn’t built to glide.

Most modern twins, and certainly the 787, are certified to fly greater than 180 minute ETOPS. That means that the remaining operating engine is deemed to be sufficiently reliable to carry the airplane for at least three more hours. We fly far away from land, so it makes sense that the airplane is certified to operate for hours on one engine.

Here’s a recent flight path. Large parts of that flight were hours away from any runway. So, if we lost an engine at one of those points, there is no altitude at which we could fly that would guarantee a glide to anything. We are counting on the remaining engine providing thrust for as long as we need it to.

The number of “glider” landings is extraordinarily few and far between. Sully on the Hudson, for example, or the 767 in Gimli. Air Transat (a case of terrible fuel management followed by brilliant flying) is another. In 99.999% of cases, the other engine does just fine at getting the airplane safely to a runway. So, training focuses on those things that are likely to happen, not the things that are incredibly unlikely.

For much of this flight below (United 16 EWR LHR on Friday, July 18) the airplane was greater than two hours from a field. LPLA (Lajes, in the Azores) was one of our diverts. It was a very long way from our route of flight.

View attachment 293675

Are dead stick landings something that you're allowed to pursue and practice on your own? Do the sims possess that capability?