If it's a gap between two parts that they're measuring, that's as simple as a feeler gauge: https://en.wikipedia.org/wiki/Feeler_gauge. The set pictured there looks like it goes down to 0.05mm.
With that in mind the phrasing sound to me like “the bushing is supposed to be moving freely, if it’s stuck to the surface you have to replace it” in aviation lingo without using ambiguous words like “freely”
Yep. Also, fun fact, these are used in dentistry and orthodontics to check spaces between your teeth. We use the same information in our 3D treatment planning environment.
Perhaps an odd question from someone who only just found out that feeler gauges exist - do they periodically calibrate or confirm the measurements of the gauges? If someone is sticking the same gauge into tight spaces all day wouldn't it wear down, even slightly, and alter the reliability of the measurement?
Can’t answer for aviation but I worked in the general testing and certification industry (Underwriters Laboratories) with tons of rules and externally audited standards about calibration of measurement and test equipment.
Yes, things like rulers and gauges are tested. The schedule depends on the use case and frequency of use. For relatively light lab duty stuff, rulers and gauges were checked 1-2 times a year. Things that are used more frequency or more aggressively are checked more often.
In aerospace everything like that is traceable and calibrated on a schedule. Traceable means that there is paper work that shows what it was calibrated against. Which itself is traceable to a NIST standard. The difference between traceable and ordinary measurement equipment is a couple of extra $.
Put is this way one time I found a set of hex torque drivers with cal stickers on them at a surplus place. So when I say everything I do mean everything. So yeah a feeler gauge should be traceable.
Yup. Traceable all the way back to the batch of material the part is made of. For every part in the plane. One of my customers makes parts for Rolls-Royce engines for the aviation industry and have to store certs and manufacturing records for even the smallest part. That includes certs on each step of manufacturing and material. Its pretty intense! During crash investigations they can be called upon to provide the full audit trail. Now imagine the amount of paperwork that is produced during an investigation when every nut and spring in a plane has the same amount of associated paperwork produced.
>The difference between traceable and ordinary measurement equipment is a couple of extra $.
This idea again. I don't know why people are saying this. It's not true. Aviation you're allowed to use any tool you want unless the workshop manual says to use a specific tool.
OEM calibration certs are useless to an aviation company because it's someone else promising accuracy that could see you thrown in jail if it's wrong. Before tooling is used the first time it will be calibrated. Then calibrated again at set intervals.
>hex torque drivers with cal stickers
It's most likely the part had stickers so it was in their system for visual inspection for good condition.
I have a pair of lockwire pliers that were pulled from the shop floor due to them being indicated worn during a visual inspection. I still have them 7 years later and they're still the sharpest pliers I own.
As for the traceability from birth, this applies to parts that go on the aircraft not tooling. Tooling traceability you need to know what tool was used if it's required to be calibrated. ie torque wrenches.
> If the feeler gauge doesn’t slip in with minimal force, then that tells you the gaps is smaller than the gauge.
The numbers of sets of comically bent out of shape feeler gauges I've encountered in my life demonstrates that there's a bunch of folks out there who don't understand how they're supposed to be used. As in "maybe if I just push harder this'll fit" :)
Yes. And places that deal with these types of tolerances will typically have a person who's job is to simply verify and replace gauging. If a gauge doesn't have a documented test verifying its accuracy, then it has to be assumed to be inaccurate and any parts or tests based on results provided by the instrument need to be redone or the part scrapped.
For things to measure, where a person finally have to sign for an OK, tools allways have to be calibrated and tested (not just in aerospace industrie). Thats also the reason, why there is not just the persons name but also the toolnumber on the test protocol.
When I was young, it was guaranteed that a lifter would let loose just before you picked up your date on Saturday night if you had not “run the valves” in the past month.
Who even uses calipers for critical precision? Calipers are for dimensions machinists append with -ish, as in four thousandths-ish. Important dimensions deserve a micrometer, especially the feeler guage that just said you need to tell your boss to scrap a 737. Decent machine shops even have different classes of micrometer; there's the one you carry in your pocket, and then there's the fancy set locked up in your boss's office that only comes out for special occasions, like when your boss has to tell their boss to scrap a 737.
I see this in my own shop. Digital displays instill people with greater confidence in the measurement than is warranted. This is something of a UI fail, as the tool manufacturer (Mitutoyo) has very clearly defined measurement precision for all their metrology equipment (+- 0.001" and 0.0015" for the calipers).
That is a very interesting observation, thank you. I've never thought about the digital vs analog in terms of actual "oh, it's digital, it's better." measurement variation.
In my experience, I think you're probably right, it does seem like people may take them as "more accurate."
This is why I still own most of the vehicles I've purchased over the last 30 years. It is much cheaper to replace worn parts, ideally before they fail, than it is to replace the entire vehicle. I have the skillsets needed to maintain my own equipment and the tools so I just monitor the operating condition of each and replace or repair things as needed.
These planes have regular maintenance checks and diagnostics to help identify issues before they become catastrophic failures. This whole thing is likely a normal maintenance issue, maybe unexpected due to the unusual nature of the scenario they face with so many idle jets, but not something that would have escaped their servicing routines. The system flagged a problem and now they can address it. If for some reason their diagnostics were not set up to detect this sort of issue then you have a problem.
Airplanes worry about fatigue. They have to be replaced after so many years or the wings will fall off in the air. If the airplane is only a few weeks away from replacement when it fails an inspection you scrap it now, if it has years to go you replace parts.
Not that I want to be pedantic, but that's a rather sweeping statement. The decision to scrap would depend on things like the number of flight hours and takeoff/landing cycles on the air frame and the number of other worn parts, hitting meantime-before-failure on large numbers of parts/assemblies, such that it may be more economical to scrap.
I love that out of all the more complex ideas about what may have been used, it turns out likely to be a solid piece of metal with of a given thickness, couldn't get more simple than that.
If you're into 3D Printing, feeler gauges are also a great way to level your bed... move the extruder 0.04 above the bed at four points, stick your 0.04 feeler under it and adjust the springs until it touches the feeler. Voila, one perfectly leveled 3D printer bed.
that's a weird in-between two more popular ways of doing it.
The non-precise way you've already heard about, sheet of paper, check for dragging.
The better (more precise) way, is using a magnetic base micrometer attached to your extruder head. You can then watch the bed run-out in real time; if your 3d printer supports many configurable sections across the bed ( I know Prusa style cartesian units mostly all support quadrants) you can record the run out everywhere across the plane without any more physical work than watching the gauge and recording the results.
P.S. be careful using both the micrometer and the feeler gauges on a 3d printer bed. Most work plates now-a-days are using PEI coatings that'll scrape off easily with metal-on-metal contact.
I like the sheet of paper method, but I have gotten a lot more use out of a $0.15 feeler gauge and I get consistently better prints... but I don't have the tools to scientifically quantify it so it could be placebo.
They're deceptively simple. If you want to get into the history of modern manufacturing, a lot can be written about the development of gauges of accurately known dimensions.
Go find a PDF of "Fundamentals of Mechanical Accuracy" by Wayne R. Moore, if you haven't already seen it. That book, along with George Daniels text "Watchmaking" are some of the top inspiring works on the mechanical arts in my collection.
Somewhat sadly a lot of the drive for that tolerance was weapons, until we could accurately make very high tolerance (for the day) parts you couldn't use parts from one rifle on another without a smith, it certainly wasn't field serviceable - this imposed a lot of maintenance overhead until they solved it properly.
It seems like we really like to push the limits when it comes to shooting each other.
That the tolerances are so precise is exactly why you can do that on most modern rifles. It depends on the design - specifically, how they achieve headspace. On an AR-15, or almost any firearm with a similar bolt locking arrangement, bolts - indeed, entire BCGs - are swappable in practice, and people do that routinely without bothering with gauges regardless of what the manual says. On AK and similar designs, yeah, that's a really bad idea.
> A complex system that works is invariably found to have evolved from a simple system that worked. A complex system designed from scratch never works and cannot be patched up to make it work. You have to start over, beginning with a working simple system. - John Gall
Somehow your comment reminds me of a pissing match I saw between a young aerospace engineer and an old retired one. The old guy was right in his argument that it doesn't matter what the constraints on materials and dimensions are, it that they come from somewhere and are documented.
His example was designing brackets for a Indonesian construction company that was using locally sourced tropical softwoods for constructing buildings. The brackets were designed to be made with 'steel' and the tolerances were compatible with a hungover guy running a drill press.
This is a thing that non-aviators often don't realize about aviation in general: the hard part is designing the artifacts and processes so that by the time you get around to operations, everything is routine and nothing is hard. Flying, building and maintaining airplanes seems hard, but that's mainly because of the volume of stuff you need to know, not because any single item is particularly difficult. All aircraft are built, maintained, and flown by mere mortals. (The designers, on the other hand... ;-)
It's not just aviation. Our entire industrial civilization works like that, and this is the only reason why it can work. We are surrounded by high-precision machinery even in daily life, and are completely oblivious to just how much technomagic it is - and how much historical effort went into building the industry up to the point where it can produce such things so easily and cheaply.