"Don't take a seat in the burst plane"

[turbguy + 544]

Don't you get a little disturbed when you hear of a jet's fan blade goes: "she loves me, she loves me not"?

I always refuse seating in the "burst plane".  I wanna be well ahead of the inlets.  I never stood to the sides (nor permitted anyone to, either) of a turbine during overspeed testing.  

Hollow blades that initiate cracking from the inside?  Weird.  You would think they would initiate from the exterior surface. Particularly near or in the root dovetail. I smell untrasonic or eddy current inspections ahead.  Even then, inspections don't catch 100% of the indications.  Perhaps a second, independent inspection, will be required.

The metallurgists going to have some fun with their SEM's.

Probably low cycle fatigue from repeated spool-ups.  Investment castings?  Anybody know? It's difficult (not impossible) to forge a hollow structure.

Some blade manufacturer is gonna get a lot of orders.

Do these ever suffer from icing?

Edited February 23 by turbguy

[Ward Smith + 6,185]

The blades all look intact. The cowling is missing. I'm going with it wasn't seated properly after the last inspection and or wasn't tightened down. I know someone who used to supervise the union mechanics for an airline. He told me stories that would make you want to walk rather than fly

[Jan van Eck + 7,556]

10 minutes ago, Ward Smith said:

The blades all look intact. The cowling is missing. I'm going with it wasn't seated properly after the last inspection and or wasn't tightened down. I know someone who used to supervise the union mechanics for an airline. He told me stories that would make you want to walk rather than fly

It is a real tribute to the designers that their babies actually manage to hold together, despite the screw-ups in the field.

I once arrived in Aruba and as the pilot dropped the flaps, there was this big wrench lying in the flap bottom pan, right next to the hydraulic lowering piston.  Let that thing jam the piston travel and on climbout when the pilot raises the flaps, only one side will come up.  Ouch.  I reported it to the pilot (!).  Gotta wonder about the sloppiness of the mechanic that left a big tool just sitting there  (then again, maybe the rest of his tool-chest bounced out on the way down).  Oh, well.  Always the human factor lurking out there. 

[turbguy + 544]

39 minutes ago, Ward Smith said:

The blades all look intact. The cowling is missing. I'm going with it wasn't seated properly after the last inspection and or wasn't tightened down. I know someone who used to supervise the union mechanics for an airline. He told me stories that would make you want to walk rather than fly

Naw, there's a blade-and-a-half missing.  I also just saw a close up of the fracture near (not in) the root.  There are reinforcing struts inside the hollow blade, which indicate they are either investment castings, welded fabrications (yuck), or perhaps even 3D printed (yuck-yuck).  I don't see how you can forge that structure.

If cast, you gotta get the investment material outta the blade, so there must be a hole in them somewhere.

Could a detached cowling do that?  Perhaps. I would expect the cowlling to show evidence, and many more blades showing impact damage.

One of the practices with such rotating machinery is to exclude damage from foreign objects (FME).  This looks like a case of domestic object damage.  Dropping stuff into a turbine-on-the-half-shell happens.  There was always a "drop list" on overhauls, with the understanding the only penalty was reporting it if you could not immediately retrieve it.  

"Don't put anything on that horizontal joint! Take everything outa your pockets and tie off that wrench"! 

Edited February 23 by turbguy

[Eyes Wide Open + 2,497]

Jet engines are quite remarkable, quiet and relatively small for the sheer power they produce. 

Until one gets into that slipstream of power it is hard to comprehend, the below video is mind boggling to me..

 

[turbguy + 544]

4 minutes ago, Eyes Wide Open said:

Jet engines are quite remarkable, quiet and relatively small for the sheer power they produce. 

Until one gets into that slipstream of power it is hard to comprehend, the below video is mind boggling to me..

 

I don't know about the "quite" part.  Yes, getting sucked into the inlet leads you to the Cuisinart.

[Dan Warnick + 6,085]

9 hours ago, Ward Smith said:

The blades all look intact. The cowling is missing. I'm going with it wasn't seated properly after the last inspection and or wasn't tightened down. I know someone who used to supervise the union mechanics for an airline. He told me stories that would make you want to walk rather than fly

Human error is, unfortunately, evident in our industry from time to time.

The photos of the inlet cowling showed a basically intact piece with one split, most likely at a riveted seam.  A blade out event would have had that engine shaking like Johnnie B. Good!  And it wasn't.  Debris from the inlet separation appears to have been ingested causing stalls, probably, in the compressor and some of the casing in the hot section area was penetrated allowing for fire to be visible around that section of the engine, at least that's what I saw in the video.  As for a fan blade crack or even failure, that investigation will take time, it won't have been announced by authorities or manufacturers so quickly.  We'll just have to wait for the investigations to complete.  Until then, they will inspect the heck out of those engines with visuals before every flight and detailed ones every 25 hours or so.

We just have to wait for the investigations to progress and be completed.

[Dan Warnick + 6,085]

9 hours ago, turbguy said:

Naw, there's a blade-and-a-half missing.  I also just saw a close up of the fracture near (not in) the root.  There are reinforcing struts inside the hollow blade, which indicate they are either investment castings, welded fabrications (yuck), or perhaps even 3D printed (yuck-yuck).  I don't see how you can forge that structure.

If cast, you gotta get the investment material outta the blade, so there must be a hole in them somewhere.

Could a detached cowling do that?  Perhaps. I would expect the cowlling to show evidence, and many more blades showing impact damage.

One of the practices with such rotating machinery is to exclude damage from foreign objects (FME).  This looks like a case of domestic object damage.  Dropping stuff into a turbine-on-the-half-shell happens.  There was always a "drop list" on overhauls, with the understanding the only penalty was reporting it if you could not immediately retrieve it.  

"Don't put anything on that horizontal joint! Take everything outa your pockets and tie off that wrench"! 

You are right, if there are broken fan blades the cause could be many things, including cowling ingestion.  You can't rule out a failed blade, true, but the composites are tough and don't have many flaws that I've seen, and as you point out structurally speaking.

The investigations will tell.

[Dan Warnick + 6,085]

Here's what the OEM calls the blades:

Pratt & Whitney's PW4000 112-inch-fan

"The 112-inch PW4000 also is our largest commercial engine, its diameter nearly as wide as the fuselage of a Boeing 737. Using hollow titanium, shroudless fan blades, the PW4000 provides high efficiency and low noise along with superb resistance to foreign object damage"

And here's the NTSB's most recent statement, according to Aviation Week:

NTSB: Fan Blades Fractured In UAL 777 Engine Incident

The U.S. National Transportation Safety Board confirmed late Feb. 21 that two fractured fan blades are among the damage found in its probe of a catastrophic engine failure which forced a United Airlines Boeing 777-200 to return to Denver International Airport on Feb. 20 shortly after takeoff.

Meanwhile, the FAA was finalizing an emergency inspection order targeting the blades, and Boeing is calling for operators not to operate affected aircraft until the new protocols are in place.

FAA Administrator Steve Dickson late Feb. 21 said the agency will order “immediate or stepped-up” inspections of Pratt & Whitney PW4000 fan blades.

“Based on the initial information, we concluded that the inspection interval should be stepped up for the hollow fan blades that are unique to this model of engine,” Dickson said.

FAA was meeting with Boeing and Pratt & Whitney representatives to finalize the inspection order, he added.

Boeing said operators of PW4000-powered 777s should keep them out of service until the U.S. agency acts.

"While the [National Transportation Safety Board, or NTSB] investigation is ongoing, we recommended suspending operations of the 69 in-service and 59 in-storage 777s powered by Pratt & Whitney 4000-112 engines until the FAA identifies the appropriate inspection protocol,” the manufacturer said late Feb. 21.

Boeing’s statement came after the only U.S. operator, United Airlines, pulled its 24 affected aircraft from service, and Japan’s regulator ordered its airlines to ground them.

"We will continue to work closely with regulators to determine any additional steps and expect,” United said.

Japan's Ministry of Land, Infrastructure, Transport and Tourism ordered all PW4000-powered 777s flown by its country's airlines grounded until further notice.

NTSB in an investigative update late Feb. 21 said two blades were damaged. One was fractured “near the root,” and an “adjacent fan blade was fractured about mid-span.”

Investigators found part of one blade imbedded in the engine’s containment ring, and the “remainder of the fan blades exhibited damage to the tips and leading edges," NTSB said. Damage to the airplane was “minor."

The incident, which took place as the 1994-built aircraft was in the initial climb phase bound for Honolulu, led to debris from the No. 2 Pratt & Whitney PW4077 being scattered over Broomfield, Colorado, a city to the northwest of Denver. Video of the event taken from inside the cabin and images of debris on the ground pointed to not only a fan blade failure, but also the loss of the inlet cowl lip and nacelle casing.

The 777-200, which was operating United’s Flight 328 with 231 passengers and 10 crew, landed safely. There were no injuries reported on the aircraft or ground.

The Colorado event appears to be strikingly similar to recent PW4077 fan blade-related failures in which inlets or cowling parts detached. These include an incident on Feb 13, 2018, involving a sister aircraft to the most recent 777-200 mishap, in which the No. 2 engine failed 30-min. before landing while en route from San Francisco to Honolulu. The NTSB cited shortcomings in Pratt's inspection processes for setting the stage for that accident.

More recently, on Dec 4, 2020, the No. 1 engine on a Japan Airlines 777-200 lost a fan blade shortly after takeoff on a flight from Okinawa, Japan to Tokyo.

In the wake of the 2018 incident the FAA issued a directive in March 2019 requiring initial and recurring inspections of the fan blades on the PW4000. It is not yet known if the blades involved in the United 328 event had been through this inspection.

This story has been updated with information from the FAA, NTSB, Boeing, and United Airlines.

[Dan Warnick + 6,085]

Here's a good article from FlightGlobal on the FAA and more from the NTSB.  Looks like @turbguy has something on the Fan Blade breakage.

FAA orders emergency PW4000 inspections as airlines ground 777s

(Excerpt)

TWO FRACTURED BLADES

“Two fan blades were fractured,” says the NTSB. “One fan blade was fractured at the root.”

“An adjacent fan blade was fractured about mid-span,” the NTSB adds. “A portion of one blade was imbedded in the containment ring. The remainder of the fan blades exhibited damage to the tips and leading edges.”

The agency says “most of the damage was confined to the number two engine. The airplane sustained minor damage”.

A Denver-area NTSB investigator began the inquiry shortly after the incident, working with local law officials to recover engine “components that separated from the engine, many of which landed in residential areas”.

The engine’s inlet and cowling separated from the powerplant, the NTSB confirms.

Source: Broomfield Police Department / Twitter

Parts of an engine from a United Airlines Boeing 777-200 which fell near a house outside of Denver, Colorado on 20 February 2021

The agency has sent a lead investigator and a powerplant expert to Denver, and the jet’s flight-data and cockpit-voice recorders were sent to Washington, DC for analysis.

Two other PW4000 failures in recent years involved a JAL 777 in December 2020 and a United 777 over the Pacific Ocean in 2018. Investigators traced the 2018 incident to a blade failure.

Cirium fleets data shows that six airlines globally operate 60 777s with PW4000s. Those include the United aircraft, 20 jets operated by Japanese carriers (11 by ANA and nine by JAL) and 18 operated out of South Korea (seven by Asiana, seven by Korean Air and four by Jin Air).

Edited February 23 by Dan Warnick

[turbguy + 544]

26 minutes ago, Dan Warnick said:

The Colorado event appears to be strikingly similar to recent PW4077 fan blade-related failures in which inlets or cowling parts detached. These include an incident on Feb 13, 2018, involving a sister aircraft to the most recent 777-200 mishap, in which the No. 2 engine failed 30-min. before landing while en route from San Francisco to Honolulu. The NTSB cited shortcomings in Pratt's inspection processes for setting the stage for that accident.

"Thermal Acoustic Imaging"?  That's a new one for me.  Never heard of the NDE process.  Anyone else?  From what I see below, inspectors could not distinguish between a true indication and an artifact of the process.  P&W IS SCREWED!!   So the blaldes are a titanium alloy, really hard to machine.  Probably investment-cast, with the hollow portions either cast or EDM'ed out the casting.

 

 

"The February 2018 incident’s roots trace back to 2005, when Pratt developed a thermal acoustic imaging (TAI) inspection process for the interiors of hollow-core PW4000 fan blades. When the process was introduced, Pratt followed “standard nondestructive testing industry practice” and categorized it as “new and emerging,” meaning it could be used while formal training requirements were being developed. But Pratt did not develop a formal, extensive training program until after the United engine failure. 

{COMMENT BY ME;  P&W IS JUST ASKING FOR TROUBLE!}

From the outset of the inspection program to the time the United Airlines fractured fan blade was inspected and up to the time of the incident, Pratt “did not have a defined training and certification regimen for the TAI inspectors” at the company’s East Hartford, Connecticut, overhaul facility, the NTSB says.

Instead, the technicians were given about 40 hr. of on-the-job training. By contrast, eddy current and ultrasonic inspection training programs include 40 hr. of classroom learning and at least 1,200 hr. of practical experience, the NTSB says.

The NTSB found that the lack of a formal TAI training program contributed to technicians misdiagnosing an issue multiple times on the United blade that failed, mistakenly believing the finding was a product of the TAI process, not a problem with the blade.

{WHOW, BOY!}

“At the initial TAI accomplished on the fractured fan blade in 2010, there was a small indication at the location of the origin of the crack,” the NTSB says. “Review of the records from the 2015 TAI shows that there was a larger indication in the same area. At the time of each TAI, the inspectors attributed the indication to a defect in the paint that was used during the TAI process and allowed the blade to continue the overhaul process and be returned to service.”

The 40.5-in., No. 11 fan blade fractured across the airfoil about 1.44 in. above its fairing at the leading edge and slightly below the surface of the fairing at the trailing edge, the NTSB says. “Laboratory examination of fan blade No. 11 revealed a low-cycle fatigue fracture that originated on the interior cavity wall directly below the surface.”

 

Edited February 23 by turbguy

[Dan Warnick + 6,085]

^^  Ouch is right.  Pratt is going to get slammed.  And the wheels on the bus go round and round...

Good call on the LCF.

No, I have not heard of "TAI" either.  I guess they didn't want to share the acronym with Thermal Anti-Ice?  I'll have to ask Google.   

[turbguy + 544]

Any cracking developing in a structure in service is actually a valid stress-relieving mechanism.  Sometimes, they self-arrest.

Here, it's also a stress-concentrating mechanism.

If and when you reach the "critical crack size" of that structure, for the loads imposed, all bets are off. 

"Stand back and stand down".

Edited February 23 by turbguy

[Dan Warnick + 6,085]

Airworthiness Directives; Pratt & Whitney Division (PW) Turbofan Engines

(Excerpts)

AGENCY:

Federal Aviation Administration (FAA), DOT.

ACTION:

Notice of proposed rulemaking (NPRM).

SUMMARY:

We propose to adopt a new airworthiness directive (AD) for certain Pratt & Whitney Division (PW) PW4074, PW4074D, PW4077, PW4077D, PW4084D, PW4090, and PW4090-3 turbofan engines. This proposed AD was prompted by an in-flight failure of a 1st stage low-pressure compressor (LPC) blade. This proposed AD would require initial and repetitive thermal acoustic imaging (TAI) inspections for cracks in certain 1st stage LPC blades and removal of those blades that fail inspection. We are proposing this AD to address the unsafe condition on these products.

DATES:

We must receive comments on this proposed AD by November 26, 2018.

-

Discussion

We learned of an uncontained 1st stage LPC blade failure and inlet separation on a PW4000-112 series turbofan engine that occurred during a revenue flight. The fracture in the blade initiated from a low cycle fatigue crack in the airfoil. This blade failure was contained by the engine case, but there was subsequent uncontained forward release of the inlet cowl, causing damage to the aircraft and prompting an emergency descent. This condition, if not addressed, could result in an uncontained failure of a 1st stage LPC blade, damage to the engine, and damage to the airplane.

Related Service Information Under 1 CFR Part 51

We reviewed PW Alert Service Bulletin (ASB) PW4G-112-A72-268, Revision No. 7, dated September 6, 2018. This PW ASB describes procedures for performing 1st stage LPC blade TAI inspections. This service information is reasonably available because the interested parties have access to it through their normal course of business or by the means identified in the ADDRESSES section.

FAA's Determination

We are proposing this AD because we evaluated all the relevant information and determined the unsafe condition described previously is likely to exist or develop in other products of the same type design.

Proposed AD Requirements

This proposed AD would require initial and repetitive TAI inspections of 1st stage LPC blades and removal of blades that fail inspection.

[turbguy + 544]

I don't know where P&W starts counting blade rows.  I thought the fan blades were not a "formal" part of the actual compressor section, but I guess they are.

 

Edited February 23 by turbguy

[Dan Warnick + 6,085]

1 minute ago, turbguy said:

I don't know where P&W starts counting blade rows.  I thought the fan blades were not a "formal" part of the actual compressor section, but I guess they are...

Not part of the HPC, or what you and I call the HPC Major Module, but they are the start of the LPC followed immediately by the booster.

[turbguy + 544]

I wonder what the inspection frequency is (probably based on engine hours an/or spool-ups), what the minimum detectable indication size is (you always assume there are defects in the structure, in the worst place at the worst orientation, sized just below the level of detection, for remaining life analysis), and what the minimum acceptable indication size and location is to permit return-to-service.

While I am aware of AD's, I never had use for them in a powerhouse.

Edited February 23 by turbguy

[Dan Warnick + 6,085]

1 minute ago, turbguy said:

I wonder what the inspection frequency is (probably based on engine hours an/or spool-ups), what the minimum detectable indication size is (you always assume there are defects in the structure, in the worst place at the worst orientation, sized just below the level of detection, for remaining life analysis), and what the minimum acceptable indication size and location is to permit return-to-service.

Dunno, but you can bet there were some PSE types that have been concerned about this for a long time.  Composite fan blades are things of beauty, until...

And they spend tons of money and man-hours trying to see all risks, but sometimes you can't see certain stresses until something like this happens.  I know, I know. 

Remember the Iowa DC-10?  That happened a few years after I graduated from A & P school and was working at, er, a certain engine MRO.  But in that case they traced it back to lifting/lowering the engines, and holding them in place with a fork truck while removing/inserting the mount bolts, which stressed the engine mounts in ways no engineer could have foreseen.

I don't know about these blades, but the ones on another type cost @ $100k/blade.

[turbguy + 544]

6 minutes ago, Dan Warnick said:

Remember the Iowa DC-10?  That happened a few years after I graduated from A & P school and was workint, er, a certain engine MRO.  But in that case they traced it back to lifting/lowering the engines, and holding them in place with a fork truck while removing/inserting the mount bolts, which stressed the engine mounts in ways no engineer could have foreseen.

The engine in the tail, the engine that typically got "choked" during take-off pitch?  I was lead to believe that failure was a hot section disc bust in service (LCF, again), no?

Edited February 23 by turbguy

[Dan Warnick + 6,085]

This is how Pratt calls them Stage 1 LPC:

[turbguy + 544]

14 minutes ago, Dan Warnick said:

This is how Pratt calls them Stage 1 LPC:

Some different about the fan blades.  I understand they are titanium, 40.5" long, but this cut-away shows mid-span "snubbers" (probably for vibration damping and/or twist control).  The photos I've seen of the actual engine that failed don't show those. They appear to be free-standing blades.

Edited February 23 by turbguy

[Dan Warnick + 6,085]

1 minute ago, turbguy said:

The engine in the tail, the engine that typically got "choked" during take-off pitch?  I was lead to believe that failure was a hot section disc bust in service (LCF, again), no?

Yes, there was that theory as well.  OEM arguing points with the airline arguing with the NTSB, yada yada yada.  But it did end up with an AD for the HPT disks in that case too.  Not to mention no more fork trucks to stick the engines up there.  By the way, that was WAY up there.  

The final version I heard was that the mounts failed, at least partially, causing a shift in the engine which caused the fan blades to contact the case resulting in an uncontained failure.  The HPT disk got loose too, and either the fan blades or the HPT disk severed all 3 independent hydraulic lines, followed by total failure of the mounts and the disintegration of the engine.  Later tests showed the HPT disk had a crack or cracks as well.  The HPT disk was found in a corn field some time later by a farmer cutting the crop, as I recall.

One hell of a "landing".

Another instance of a DC-10 crash was the Flight 232 disaster at Sioux City, Iowa, USA, on July 19, 1989. After the #2 engine (tail engine) suffered an uncontained fan disk failure in flight which ruptured critical hydraulic lines, the crew, led by Captain Al Haynes and assisted by a senior pilot flying as a passenger (Dennis E. "Denny" Fitch), performed an emergency landing by varying remaining engine power to control the plane. Although the aircraft was destroyed with the loss of many lives, the crew flew the aircraft onto the runway in a partially controlled manner and 185 of the 296 people on board survived.

The Sioux City crash concerned investigators because the total loss of hydraulic pressure aboard the DC-10 was considered nearly impossible. The design had lines from all three independent and redundant hydraulic systems in close proximity, directly beneath the #2 (tail) engine. Debris from the #2 fan disk separation failure penetrated all three lines resulting in total loss of control to the elevators, ailerons and rudder.

 

[Dan Warnick + 6,085]

7 minutes ago, turbguy said:

Some different about the fan blades.  I understand they are titanium, 40.5" long, but this cut-away shows mid-span "snubbers" (probably for vibration damping and/or twist control).  The photos I've seen of the actual engine that failed don't show those. They appear to be free-standing blades.

You're right.  Those blades in the cutaway are not the real McCoys.  I just posted it to show how they categorize the LPC.

[Dan Warnick + 6,085]

13 minutes ago, turbguy said:

Some different about the fan blades.  I understand they are titanium, 40.5" long, but this cut-away shows mid-span "snubbers" (probably for vibration damping and/or twist control).  The photos I've seen of the actual engine that failed don't show those. They appear to be free-standing blades.

Are you in aviation or oil & gas?  I've always wanted to experience the oil & gas side of the turbine business.

[turbguy + 544]

1 hour ago, Dan Warnick said:

Are you in aviation or oil & gas?  I've always wanted to experience the oil & gas side of the turbine business.

I'm retired from Large Power Generation (coal, oil, nuc, hydro, nat gas, a little wind). I retired when the term PPE was becoming popular.  Had a short stint in manufacturing at large steam, heavy duty CT's, and aircraft engines with a "generous"  "electric" company.

"Hard hats protect soft heads".

I really don't know much about oil & gas business at all.  Be gentle with me.

We have WAY too many TLA's   (Three Letter Acronyms)!

A typical workspace for me, and it's almost always warm in the winter.  Working CT's in winter was BRUTAL!

Edited February 23 by turbguy