After a door panel came loose on a Boeing 737 Max aircraft in January, the company was subject to a federal safety inspection. It hasn’t gone well.
Boeing is having a rough time of it right now, with parts falling off its planes left, right and center. Just last week, a wheel came loose and smashed through a car, and earlier this year the door from a 737 Max aircraft broke off mid-flight. That mid-air disaster sparked an audit from the Federal Aviation Administration, which has gone far from well.
The issue is that if they are doing this, it means that the workers doesn't have the proper tools for the job.
The keycard should be replaces with a go/nogo custom card, and the soap should either be specified by brand in the manual or swapped to a certified lubecricant, that has been tested to work fully with the gasket and not cause deteriation or on any way affect the quallity of the seal.
I can't know for certain what is specifically going on there but I do work in contract manufacturing for high end scientific equipment and critical medical electronics so I do know a fair bit about the processes used. For me the dishsoap and keycards on their own don't raise any alarms. It sounds like the main issue is poorly written incomplete manufacturing instructions, which is a big enough issue on it's own and is an absolute monster to try and fix once your production workers have gotten used to working like that.
the soap should either be specified by brand in the manual or swapped to a certified lubecricant, that has been tested to work fully with the gasket and not cause deteriation or on any way affect the quallity of the seal.
The seals used are most likely silicone (it's what we use on environmental chamber doors). If so there are very few chemicals that will harm them let alone dishsoap. We actually use 409 (a bathroom cleaner) spray to lubricate our seals where I work.
The dishsoap is almost certainly something they order and stock with their own internal shop supply number. The instructions most likely reference that number but that number would be meaningless to anyone else so the news article just said dawn dishsoap. It's not going to be any random dishsoap because that's not how industrial supply works. It would be more expensive for them to go pick up random dishsoap than to just keep ordering the same part number (that specific dawn dishsoap) in bulk from their industrial supplier.
The keycard should be replaces with a go/nogo custom card
Why in the world would you make custom tooling when there is a readily available off the shelf solution? You can just buy packs of keycards for dirt cheap and they are going to be a known thickness because they need to be to keep working in the same keycard slots. That thickness should be documented somewhere but it isn't going to be in the manufacturing instructions because the production people don't need it; they just need to know that the go/nogo gauge (the keycard) should fit. The more extraneous information you include on manufacturing instructions the greater the chance you have of someone missing or misreading something. If someone needs that extraneous info or something on the production floor isn't right that's when you bring in the engineer or process support staff who will have access to that info and the authority to make decisions based on it. If your production staff are making critical decisions on their own then something is very wrong with your manufacturing instructions (which sounds like the real problem here).
Submariner here. After several incidents in which submarines imploded, burned, or otherwise caused death and/or endangered thermonuclear weapons systems, our current procedures specify every single item used down to specific serial numbers, with specific authorized substitutes. If the authorized substitute cannot be found, the procedure is simply not done, and if necessary for ensuring the actual safety and conduct of the submarine's primary mission, the entire multi-million-dollar mission is cut short and the ship surfaces to either receive the requisite supplies or goes back to port. Specific serial numbers for lubricants, specific stress-tested seawater-proof pressure-resistant alloys for bolts, specific serial numbers and part numbers for fuses, specific torque wrenches, even specific serial numbers for indicator lights. Every single maintenance step of certain procedures are read out loud at least three times and re-confirmed and acknowledged by both the worker and supervisor before being conducted, including the opening and closing of maintenance panel doors.
If the dishsoap is standardized in the documentation I don't see any issue with it.
The hotel keycard, less so, since it is used to meassure how tight a fit is it will inevitably get worn, so the card needs to be durable with a predictable wear pattern, I have had hotel keycards made put of all kinds of plastic, paper even wood, they all have drasticly different thickness, wear patterns and durability.
If the documentation is too generic it looses it's meaning.
All of the big hotel chains use the same plastic key cards that are credit card sized, they are durable and can be reused many times but also cheap enough to not fret over them if a customer forgets to return it before leaving. As a former aircraft maintainer myself, I don’t personally think it would be an issue if Boeing or its contractor ordered a bunch of standard hotel card blanks for seal testing, but if they were meant to use that as their test device it should be documented , there should be a part number for that card and authorized suppliers, and there should be a specific procedure to follow when using them. The article mentions the lack of documentation, so this was probably an unauthorized improvisation on the fly. I doubt these were being used to measure a specific tolerance, this case was probably something stupid like “the cabin pressurization check failed after we replaced the door, let’s poke a card along the seal to find where the gap is and squeeze extra sealant in that spot.” My specialty was avionics though, so I will admit I don’t really know much about the pressurization checks and seals, I was always at the plane for some other work whenever I encountered them.
I'd suspect neoprene not silicone, for door-seals of aircraft.
the Dawn I've no problem with.
The checking-fit with hotel-keycards I have one HELL of a problem with.
It's an aircraft: tolerances should be specified, and should be made to fit those tolerances.
It's umpteen tens-of-degrees below freezing outside, when you're at cruising-altitude, so you've got a pressure-vessel ( the fuselage of the aircraft ), AND you've got a termperature-differential, AND you've got metal-fatigue ( or composite-aging/accumulating-cracks-in-its-reinforcement-fibers ), and tolerances are supposed to be engineered, not "oh, it seems to fit" bullshit.
Anyone who cares about such things, please read some in-depth stuff on aviation crashes.
There are youtube channels devoted to going through things, and I found out about a jetliner losing its tail because of 3 bolts that were the wrong steel, on one of those channels, but the written stuff packs more knowledge per hour of study..
Jan Roskam, aircraft-designer, has one book on it, old, but important, subtitle is "The Devil Is In The Details".
The Lessons From The Sky series has info on near-accidents, and you'll note they are more human-centered than the sometimes technical-as-hell items in Roskam's book..
When one discovers that a jetliner can kill everyone aboard, when it's being used for short island hops ( Hawaii ), and that means it's getting many more pressurization/depressurization cycles than the engineers intended, or that salt-spray in the air can corrode an airframe enough to cause catastrophic failure, or that a single failed cotter-pin can remove the controls from a homebuilt while in-flight ( another source )..
"The Devil Is In The Details" is the most-true subtitle I've ever seen in any book.
Prevent problems.
Catch All Lapses.
Discover problems you didn't know to be proactively preventing.
Prevent any discovered problem from ever EVER getting roots/legs to harm anyone else.
seems saner to me, than the jackassery that Boeing has been doing, since McDonnell Douglass did a reverse-takeover from the inside, after their merger.
Bottom-line "leads" the company, my ass: it's sunk Boeing.
At my old job we used PAM cooking spray and credit cards. It was written into the documentation, btw.
(This was a global company, btw, not a ma and pa shop).
I don't support Boeing, but I feel like this aritcle is sensationalizing a possibly mundane aspect of the manufacturing. Like none of us know the specifications for assembly. The tolorance for those specific gaps could be generous and a key card turned out to be an acceptable gauge. Automotive industry uses a variety of Go/No-Go gauges that are 3D printed which are far outside the accuracy of Feeler guages, because in some cases you dont need the precision. If the gap needs to be 2mm +/- .5mm and a key card is 2mm +/- 0.4mm or whatever then its fine.
Overall, what I am saying is that the issues with Boeing is systemic across all aspects, but one shouldn't be so quick to jump on details that make a eye grabbing article title.
I get where you're coming from but as someone who's worked in the industry where everything is super over regulated, something like this is a huge deal. Not only would they be required to use a feeler gauge. That gauge would need to be tested and calibrated every (where I worked) 2 weeks. The DIY'er in me thinks it's ridiculous to "calibrate" a piece of metal, but in the industry and others like it, that's what they're required to do.
When government auditors come in and see something like this people get fired. They also search deeper because if you're already doing something that blatant there's going to be more to be found.
Once again this is a question of tolerance. It could be a place you only care about minimum clearance. If the spec is for a gap of at least 15 mil then a pack of 30 mil key cards makes for a bunch of cheap easily replacable go/nogo gauges with enough leway that even a worn one won't put you under spec.
I'm most familiar with IPC standards for electronics but even in the most critical class 3 applications there are plenty of spots where the standards are effectively gauged by eyeball let alone with even a makeshift tool because those specific specs aren't that critical for the application.
I'm also Class 3 certified and where I come from inspection will absolutely measure clearances if it's not obviously within spec. And even then there are many that still do just to cover their own ass. God help them if QA steps in and gets a different measurement.
I have no doubt different facilities run differently. Just look at Boeing. But don't assume they're all run like a circus just because you worked somewhere that doesn't do things by the book.
I'm not saying we played fast and loose with the spec. I'm saying that there are plenty of places where the actual spec doesn't use numbers and it is a judgement call for example minimum lead protrusion.
There are also plenty of places where a number is given but it is not possible to measure such as barrel fill on through hole components. In those situations an inspectors best bet is to eyeball it and if it's even questionable to rework and correct the process so it isn't. You don't eyeball it when it's close, but you also don't need to measure, for example, lead protrusion on every lead when they all apear to be definitely under 1.5mm.
I agree there will be way more serious violations. The point I was really trying to make was that the FAA may have listed 100+ other more serious things that are not as catchy or mean little to people outside the industry. However, the media chose to report on this because its relatable to the average person and make it seems more significant than it really is.
"Failure to establish a validation system to ensure all components are present and installed" (4 door bolts) just doesnt have the same ring, dispite being significantly more serious, then Boeing using soap and keycards.
Soap is bad if you need to keep something lubricated but it works pretty well is you just need to lube something up a bit for installation because you can just hose it off afterwords. As far as the keycards go, that depends on the tolerance. If that tolerance is generous enough then an actual feeler gauge is overkill. A pack of keycards is dirt cheap and are all going to be about the same thickness. When one gets busted or worn then you can just grab a new one from the pack. Verses having a production guy break a feeler gauge and start guestamating because they don't want to tell the boss they broke the third one this week.