Boeing 7*7: the saga continues…

[PakledHostage]

[Pakled]: It is broken, can you make it go?

[Vogon]: Please complete form 127.8(b) "request for form 976.3(a) form" in triplicate...

[Pakled]: ...It is broken, can you make it go?

[Geordie La Forge]: bangs head on desk

[StrandedonEarth]

1 hour ago, PakledHostage said:

[Vogon]: Please complete form 127.8(b) "request for form 976.3(a) form" in triplicate...

... or be subject to a Vogon poetry reading...

[tater]

1 hour ago, PakledHostage said:

(I have been involved first-hand in FAA rulemaking processeses. Trust me, it is very bureaucratic... The meetings are T.E.D.I.O.U.S. tedious. ...Imagine a room full of Vogons and Pakleds and you'll be on the right track.)

"Start Printed Page 9763"

(one of many in the doc you linked, lol)
 

They'll start a new process with whatever is state of the art in 2024, and by the time it passes it will seem old, and a few years later people will say, "Why  does the FDR only store 4 TB of data, and not many petabytes?"

[mikegarrison]

1 hour ago, PakledHostage said:

Please help me out here @mikegarrison!? They don't seem to be understanding what I am saying...

 

Yes, two different boxes that record two different things.

The CVRs are something of a compromise because pilots don't want everything they say being recorded but also understand that being able to access the CVR ultimately can help keep them safer.

Many years ago, I actually used the cockpit ambient mic from a CVR to estimate the amount of hail an airplane flew through before both engines flamed out. It was of course an extremely rough estimate, but I was able to show that the plane very likely did fly through more hail than the engines were certified to be able to ingest.

[PakledHostage]

Seems United has found problems with some of the plugs on their aircraft:

https://twitter.com/cnnbrk/status/1744473546070344134

[tater]

11 minutes ago, PakledHostage said:

Seems United has found problems with some of the plugs on their aircraft:

https://twitter.com/cnnbrk/status/1744473546070344134

Wonder if it was a specific employee? Those are held with a cotter pin though, right?

[StrandedonEarth]

59 minutes ago, tater said:

Wonder if it was a specific employee? Those are held with a cotter pin though, right?

Or safety wire. Lazy git probably center-punched the threads and called it good. But isn’t someone supposed to independently verify it’s done correctly? Oh right, probably got laid off as redundant…

[Shpaget]

FAA has already proposed rule change to increase CVR from 2 to 25 hours. That is likely to happen soon, however, the regulation would apply only to newly manufactured planes.

NTSB lady was saying that the new rule should apply to all planes.

https://www.faa.gov/newsroom/faa-proposes-requiring-25-hour-cockpit-voice-recorders

EASA and ICAO have had this rule for years, so devices are in actual use. There is no development issue. It is literally an off the shelf part at this point.

Edited January 9 by Shpaget

[darthgently]

Just leaving this here

 

 

[Shpaget]

13 hours ago, PakledHostage said:

Seems United has found problems with some of the plugs on their aircraft:

https://twitter.com/cnnbrk/status/1744473546070344134

Alaska Airlines reports that they too have found "some loose hardware".

https://news.alaskaair.com/alaska-airlines/operations/as-1282/

Quote

As our maintenance technicians began preparing our 737-9 MAX fleet for inspections, they accessed the area in question. Initial reports from our technicians indicate some loose hardware was visible on some aircraft.  

NTSB says the plug moved upwards and disengaged all twelve stops of the holding mechanism. The four bolts that were supposed to hold the door down have not been recovered yet and NTSB has "not yet determined if they existed there". That will be determined in their lab in Washington DC.

https://www.youtube.com/watch?v=DI8tv9h0PPg

timestamp 2:50

[tater]

Wow.

So it is tightened until tight (some min torque?) such that the holes align with the gaps in the castellated nut, and pinned. Loose seems difficult unless never tightened properly, or unpinned.

For older aircraft, the nut can wobble, wearing down the cotter pin, maybe (easier than wearing down the bolt, certainly)... pin fails, nut free to fall off—but the bolt would still have to then drift out for the assist spring to be engaged. For a plane only in service since October, seems more likely no nut at all? Slowly drifts out over a few flights (X2)?

[RCgothic]

I have spent a fair amount of time on fastener tightening in my career. Basically, if the fastener isn't tight enough to bind the joint without slip, then it will fret and chafe and fatigue and fail under repeated load cycles.

Castellated nuts and cotter pins are a *terrible* way of securing a bolted joint. The nut has to have enough clearance to get the pin in which means in practice it can back off a little. Depending on torque and wear it's not unusual to shear the pin entirely. At best this will temporarily retain a nut that's come a bit loose before the bolt fails.

The best way to make sure a nut doesn't come loose is to ensure it's correctly fastened to begin with (and the best way to do that is with a stud puller) and definitely not with a castellated nut.

Yikes.

[darthgently]

40 minutes ago, RCgothic said:

I have spent a fair amount of time on fastener tightening in my career. Basically, if the fastener isn't tight enough to bind the joint without slip, then it will fret and chafe and fatigue and fail under repeated load cycles.

Castellated nuts and cotter pins are a *terrible* way of securing a bolted joint. The nut has to have enough clearance to get the pin in which means in practice it can back off a little. Depending on torque and wear it's not unusual to shear the pin entirely. At best this will temporarily retain a nut that's come a bit loose before the bolt fails.

The best way to make sure a nut doesn't come loose is to ensure it's correctly fastened to begin with (and the best way to do that is with a stud puller) and definitely not with a castellated nut.

Yikes.

I don't know about aerospace, but Loctite Red™ is a wonder in ground vehicle applications though a possible challenge during disassembly later.  A propane torch is your frenemy here

[tater]

That image is of a 737 door plug hinge, BTW (if it wasn't clear).

I still don't get how this could fail so quickly unless they were not installed—except if not installed the assist would have pushed the door off the stops immediately, right? I'm unsure what the 2 bolts at the top do—maybe those were in place, bottom ones never installed, and in the short service life (plus sitting around beforehand) the springs are exerting an upward force. In flight, that's a sort of hammering on the 2 upper bolts, and those just shear?

[darthgently]

 

 

[AngrybobH]

1 hour ago, tater said:

I still don't get how this could fail so quickly unless they were not installed

Just a guess but over-torqued bolts could do this. Torqued past yield the bolt will lengthen under load and then stay that length making the bolt effectively loose.

[darthgently]

47 minutes ago, AngrybobH said:

Just a guess but over-torqued bolts could do this. Torqued past yield the bolt will lengthen under load and then stay that length making the bolt effectively loose.

It would be interesting to find out that all the bad installs came down to one botched torque wrench at a common shop or one rookie mechanic at that shop with arms like an ape and who goes through torque wrenches like popcorn 

[magnemoe]

3 hours ago, RCgothic said:

I have spent a fair amount of time on fastener tightening in my career. Basically, if the fastener isn't tight enough to bind the joint without slip, then it will fret and chafe and fatigue and fail under repeated load cycles.

Castellated nuts and cotter pins are a *terrible* way of securing a bolted joint. The nut has to have enough clearance to get the pin in which means in practice it can back off a little. Depending on torque and wear it's not unusual to shear the pin entirely. At best this will temporarily retain a nut that's come a bit loose before the bolt fails.

The best way to make sure a nut doesn't come loose is to ensure it's correctly fastened to begin with (and the best way to do that is with a stud puller) and definitely not with a castellated nut.

Yikes.

Remember seeing an F-16 jet engine a very long time ago. It had wires connecting the nuts or bolts on the engine, I assume it was holes they went trough and believe they was looped around the wire again before on to the next so if one break the line the rest would still be rigid. Probably makes inspection easier 

[Shpaget]

32 minutes ago, darthgently said:

It would be interesting to find out that all the bad installs came down to one botched torque wrench at a common shop or one rookie mechanic at that shop with arms like an ape and who goes through torque wrenches like popcorn 

I have no trouble imagining that. It appears to be a lot harder to get some guys to understand that a click of the torque wrench means "stop pulling" than I would have thought a few years ago. The usual excuse when asked "Did you not hear and feel the click?" is "Yeah, but I though I'd tighten it just a little bit more."

I have three collet nuts (ER16, that's M22 nut) sitting on my desk that are split completely on one side by overtightening them. What is worse is, I found them by chance. They were being actively used in the milling machines. No one knows who did it, no one noticed they broke.

[mikegarrison]

57 minutes ago, magnemoe said:

Remember seeing an F-16 jet engine a very long time ago. It had wires connecting the nuts or bolts on the engine, I assume it was holes they went trough and believe they was looped around the wire again before on to the next so if one break the line the rest would still be rigid. Probably makes inspection easier 

That's called lockwire.

[CBase]

Seriously how antique is Boeing manufacturing ? From automotive I do know that all "A" quality class screws are fastened with an eletric screwdriver which documents final torque and angle in a quality system. No way you could skip any screw or not reach appropiate torque. I do understand that 737 MAX production shutdowns made it hard to get good production line worker, but that is a poor excuse for lack of quality systems in place. And in 2023/24 another person with a pen to set a checkmark on paper is not industry standard quality control.

[PakledHostage]

5 hours ago, RCgothic said:

Castellated nuts and cotter pins are a *terrible* way of securing a bolted joint. The nut has to have enough clearance to get the pin in which means in practice it can back off a little. Depending on torque and wear it's not unusual to shear the pin entirely. At best this will temporarily retain a nut that's come a bit loose before the bolt fails.

The best way to make sure a nut doesn't come loose is to ensure it's correctly fastened to begin with (and the best way to do that is with a stud puller) and definitely not with a castellated nut.

Yikes.

Castellated nuts are used all over the place in aircraft. This isn't a case of some sort of incompetence.  They're typically not used in strucural joints, but they are everywhere on an aircraft. Smaller strucural joints are typically done with Hi-Loks and big structral joints will use suitably torqued conventional bolted joints. The joint in question isn't a structural joint because the bolts just hold the plug in place relative to the stops. The stops carry the structural (pressurization) loads, not the bolts.

Edited January 9 by PakledHostage

[tater]

43 minutes ago, PakledHostage said:

The stops carry the structural (pressurization) loads, not the bolts.

Maybe a dumb question that someone knows the answer to, but what's the nominal door latching mechanism look like? Clearly not locked nuts... Why not have the door-plug otherwise more door-like, and have the latch only on the inside, and nonetheless covered by the interior trim panel. It can then be removed (opened) easily for maintenance, and the only care required is to lock the door. All the functional doors have latches, after all, and they certainly take intent to unlatch.

Edited January 9 by tater

[mikegarrison]

35 minutes ago, PakledHostage said:

Castellated nuts are used all over the place in aircraft. This isn't a case of some sort of incompetence.  They're typically not used in strucural joints, but they are everywhere on an aircraft. Smaller strucural joints are typically done with Hi-Loks and big structral joints will use suitably torqued conventional bolted joints. The joint in question isn't a structural joint because the bolts just hold the plug in place relative to the stops. The stops carry the structural (pressurization) loads, not the bolts.

When I first joined Boeing, they had a program where engineers spent about four weeks at a voc-tech school, where we had to build various airplane parts following blueprints. This was supposed to teach us to make better blueprints.

I spent my entire career in Airplane Performance (later called Flight Sciences) doing Noise and Emissions engineering. Never made a single blueprint.

Anyway, it was a fun class. I still have the part I built (part of a 727 PSU) in my garage.

I guess the point, if there was one, was that Hi-Locs were one of my favorite fasteners. Fun to use.

Anyway I agree that, to the extent I understand the design, the bolts in question are not highly loaded and probably it made sense (or seemed to make sense) to use a castle nut just to retain the bolt, not to ensure some level of torque on the nut. Castle nuts should only be used in low-torque applications.

[tater]

Given they found the door, but no bolts, I wonder how they move forward? Look for indications of how a bolt failed in the parts the bolts pass through?