Build your house in the winter, the wood's cte is much higher than the tape's. Cutting in winter will give you more wood at nominal Temps while cutting in summer will give you less.
@@Inconspicuoustwrex Huh, I never expected that to be the case, but I looked it up and it looks like wood does indeed have a roughtly three times higher coefficient of thermal expansion than spring steel (if I didn't missread anything by an order of magnitude).
So I ended up looking it up, and my base assumption was wrong. Wood has anisotropic material properties due to its grain structure. It has about twice the cte as steel, 30 micrometer per meter per degree C for wood vs 16 micrometer per meter per degree C for steel, in the radial direction of the tree, tree getting fatter, while having 5 times less cte as steel, 3 micrometer per meter per degree C for wood vs 16 micrometer per meter per degree C for steel, in the direction of the grain, tree getting taller. We cut timber to length along the grain direction, so cutting in summer would result it more wood than in winter. If we are cutting 1 meter of wood when it is 30 degrees C outside, our tape measure would read 1 meter but would actually be 1.00016 meters long. If we then cut the wood, it would be a piece of the same 1.00016 length. Now, if we wait till Temps cool to a more nominal 20-degree C, that 1.00016 m piece of wood would shrink to 1.0001299952 m, which is about 130 micrometers more wood than if you had cut it initially at 20 degree C. If we do this when it's 10 degree C out, the 1 meter tape measure would actually be 0.99984 meters long, and when measured at 20 degrees C the wood would end up being 0.9998100048 meters long. Cutting your wood at 30 degrees C results in 0.0003199904 m more wood that cutting your wood at 10 degrees C.
@@PetesGuide There's that angle to it. But of the two blocks that Bureau Drive runs, NIST is the entirety of one of them, and the Subway just barely outside is all the way down at 6 Bureau Drive.
There’s an entire video done by a a UA-cam channel about NIST, I’m trying to recall which one it was. Look for it, it’s quite fascinating, especially the mega heavy weight they have for calibration of mega capacity scales.
In case anybody was wondering, "NIST traceable" basically means the manufacturer (or a 3rd party calibration service) sent their equipment to NIST, and then they use the calibration report and NIST approved procedures to calibrate things. This can actually go several degrees of separation, but you can always trace back to NIST.
nice to see an upload from you again. its been a while. not sure if you know 'inheritance machining' but they mostly focus on accuracy and machining tools and other machines that make other things. would love to see a collaboration between you two.
Being someone who's spent time in a metrology lab, I know just how bad tape measures can be. Seeing that you can get NIST certified ones has be a bit to excited.
As an engineer who thinks that engineering as a whole is horrendously underestimated by the general public its easy to see why metrology is even more obscure because most engineers take calibration for granted. I work in industrial control systems and its staggering just how much we rely on the manufacturers getting things right. We simply take it on face value that the manufacturers made every pressure, flow, temperature,...... sensor accurately and that they measure what they claim to measure at the scale on the box. So at the practical application level we are separated from the metrology people by the manufacturing layer. Ever since I saw MT's vid on the Jo Blocks that was followed by his more recent one on the NIST I have gained a lot more appreciation for metrology and wish it was covered a lot better at the college level.
When siding my house with my dad, he ran the chop saw, and I was getting pretty frustrated that his cuts were coming up 1/4 short. Turns out we were both exact on the mark with different tape measures...
Way back, I did a summer job in a metrology lab. It was really interesting and gave me an instinct to question "how accurately do we really know what we think we know". Was also useful for helping me think about various ways experiments could be a bit off, which has served me well over the decades. It probably should be a required topic for any engineer and recommend for scientists to learn about.
If anyone is curious, NIST SOPs 10-12 has some great guidance on calibrating rules. And of course NIST has the best single axis vision bench in the world. What an amazing experience, I’m so happy you were able to record this. I’ve been dying to see an in depth tour of this. Thank you so much.
It bugs me no end that NIST uses a weight measure to specify the tension, when a force measure is what they meant to specify. But then, US customary units have always been very confused about this particular distinction.
@@MrCrackbearexactly. It's fortunate that the distinction generally doesn't affect anything practical since a pound mass exerts a pound force under standard gravitational conditions
my dad has one of those Stanley tapes from my grand father, he managed a local paint store after WW2 and it was a give away cause grandpa still had like a half a box of them in the 80's before he died
Sigh. I did. Thankfully it was from my 'own' set of tools so it only hurt me. I was in like middle/high school. "This tape measure is garbage! Can't believe it's worn itself so loose." Took a punch or something and hammered the rivets tight. Wasn't until years later...oh no.
Same here. I decided to super glue it to take all that "slop" out. Luckily before the end of the project I had let it slam when it retracted a few times and it broke my "repair"
I would love to see how NIST analyses modem theodolites for land surveying. We also have to take into account temperature and barometric pressure because they really do matter over thousands of feet or miles. Appreciate you.
Did you happen to ask Mr. Lee what model of tape measure was the first one you tested? He said it was his personal one that stays in his desk. I would be very interested in what model it was.
Tape measures do have accuracy classes, most of the tapes you see at the big box stores are considered class 2, you can order class 1 tapes like Fisco or Advent. I have found as far as standard home depot tapes that Milwaukee and Dewalt were less accurate than the Stanley's (as boring as they are they are a staple in the category).
Oddly enough, Stanley/Black+Decker own DeWalt so I’m sure there’s some sharing of manufacturing techniques, engineering, etc. Stanley certainly has more experience historically. Milwaukee and Ryobi are owned by the same parent company, and they don’t necessarily share the same reputation either.
I genuinely delight in how the UA-cam algorithm throws up such “left field” stuff from time to time. What’s even more surprising is what I discover about myself in the process. Thanks to this video, I now know I’m a measurement nerd! 😂
I'm a former calibration tech in the Marines and for Lockheed at Stennis Space Center, and I used to calibrate tape measures for the satellite blanket department using our Renishaw and HP laser interferometers. I'll never forget the first time one came through the lab. At first we thought it a bit odd that someone would need a type 1 lab to calibrate a tape measure, until we found out why. I'm also from Denver, and grew up listening to the "atomic clock" from NIST on the phone using a local number back in the 80's, which kick started my career. Made many visits to NIST, and met with a few of them, throughout my career as well. Small world.
Okay... I have a question that's been driving me nuts. Sticking with freedom units, ya know on tapes how the whole number has a thicker mark? Where is that value relative to the mark? I thought centre, but when I compared the tape vs. my 6" Starrett ruler, it shows the whole value is at the beginning of the mark. Is there an industry standard? Cause every time I bring a new ruler into the shop, I go through this insane making cross-checking.
That is interesting, no joke, In repair or in existing work that would be relavent but in new construction does it matter? Everything is relative, if everything in a newly constructed biulding that was measured at 12' 6" was really 12' 6.1 " I dont think there would be a problem. When I was a helper as a young guy I always did check my tape measure against my bosses' so I could give him what he asked for.
Not sure about that. Because NIST is very specialist, only certain types of scientists and engineers want to go down that road. There's lots of scientists that jump around to adjacent fields because of how much cross disciplinary research there is nowadays. I don't know how much of that can happen in NIST.
The latter fun fact was something I didn’t learn until the last few years, but it totally changed my view of measuring tapes. I never took them seriously below the 1/4 inch level or so because I always assumed the looseness and presence of the hook on the tape assured a minimum error.
10:12 - "If you didn't like it, really let me know by giving it the double thumbs down" How do you give a "double thumbs down" ??? Or, if meant in the logical/obvious way, by clicking thumbs down once to indicate a thumbs down rating, followed by clicking thumbs down a second time to remove the thumbs down rating, how would this be distinguished from not clicking at all (no rating)?
Will depend how long it is. Steel's coefficient of thermal expansion is well known. I assume since you used Fahrenheit we are using customary units. The thermal expansion of steel is somewhere between .000006 and .00000694 in/in/F. So for a 25 foot tape measurer you can expect it to be between 0.216 and 0.2498 inches longer at 120 degrees than at 0 degrees. The laser interferometer was showing a measurements to the 100,000th of an inch in the video. Even a tenth of a degree Fahrenheit would cause a difference of 0.0002 inches over 25 feet. Easily detectable with this instrumentation. Which is why they control and measure the temperature in their lab so precisely, and probably the imprecision of temperature control and measurement over such a long space is a larger contributor to their uncertainty calculations than anything to do with the instruments.
This is something considered in calibration labs. The reported temperature of dimensional calibration is important to know. You can theoretically determine the expansion/contraction, but if you really need to know you should always perform the experiment. Most would just calculate with a CTE for the expansion/contraction, and estimate the uncertainty. Detecting the difference would be trivial for the laser. Their accuracy is in the millionths of an inch or less, considering optimal operating conditions.
That's funny you posted this with the McMaster Carr site. I work in an aerospace machine shop, we make parts for SpaceX and Blue Origin. We just bought one of those Starrett certified tape measures for our shipping/receiving to do receiving inspection on raw materials. The certificate says it is +/- 1/32 over 12-feet.
I wonder about the uncertainty measured to microns in 9:09 when the comparison is made just by visual inspection. The visual inspection shown in 3:53 will not provide such precision. Maybe that is why it is not called deviation but uncertainty.
4:20 A computer could verify the accuracy of _Every_ _Single_ _Graduation_ _Mark_ on a ruler or tape in a few minutes. I'm very surprised that they are still doing this "old-school".
Just how traceable is that 5kg tensioning masspiece. Has it been verified against any of the standard kilograms, which are obsolete, but which are very small due to their density. I have some masspieces that are traceable to it with only a few steps. Verified against a balance, that was in turn verified by a replica masspiece, which in turn was referenced against one of the prototype kilogram units that were sent all over the world. 1g, to the milligram though you can measure the mass change that occurs due to corrosion from the atmosphere, and the tiny mass lost due to handling.
Great video, it was very interesting to see a legacy tape on the calibration machine, I would be curious to see how it compares to a consumer grade tape measure from a box store.
I’ve often wondered how NIST does these tests if the enity requesting it needs a temperature outside of 20°C. They can’t change the temperature in the lab because it takes weeks to adjust it
I got in trouble as a machinist once for making parts too short. I said no way, I made them correct within tolerance. Turned out at one metre and before my tape was great. At two metres and beyond my tape was great. Between one and two metres the error grew to 1mm then decreased again. My parts were 1mm short at at 1.5m We stretch the tape measure out along a 2 metre steel ruler and couldn’t believe what we were seeing. It was a Stanley brand. Needless to say they sent me out a free refill for the tape. Scrapped parts were worth 10x that. However I was off the hook 🤠
That hook to the 2 inch mark accuracy is probably more important to the accuracy of that tape than the next 48 inches. In the sheet metal sheer shop my company would only allow a few different brands of tapes and we had to check them on a standard ruler before sheering tons of stainless for production.
Assuming 1/16 is the finest division that's extrapolates to under 1 division over the length, I consider anything good to give or take 1 count full scale to be dead on.
Unfortunately, you do not get the world best tape meter which is the Class I branded BMI, from Germany. It is also the lightest, sturdiest, with a lot of perks, like self expanding, stays flat, etc.
a video on optical interferometry wouldn't be a bad idea. I'm assuming that an infra-red laser was used for this work, with a wavelength on the order 800 or 900 nm. My guess is that the reflected beam is compared (i.e. added) to the transmitted beam, and the max and min are counted as the reflector is slid down the track? I've heard of interferometry being done with white light, and I can't imagine how it works with the range of wavelengths found in white light.
I mean, a light can *look* white, but still have sharp emission bands that you could (in theory) use for interferometry. I'm honestly not sure why you would want to do this, however. It occurs to me that a completely different application of interferometry would be to measure the thickness of a transparent coating. The idea is that the surface and the bottom both reflect some light, and the reflected light interferes with itself. This is the same effect that causes that rainbow sheen on oily water. By observing the exact wavelength that is reflected back the strongest, you can determine the thickness of the coating, and of course you need white light to make this measurement.
I remember a project that needed graduated tapes for aligning conveyer tracks. Nothing super accurate, but it was surprising the error between three measuring tapes from the same package had 2mm discrepancy over one meter. Didn't matter here, but was quite startling.
Every Starrett tape I have has a Roman numeral 2 inside a flat oval apparently signifying it as Class 2 with a tolerance of +/- 3/32” over 100ft. But I have only seen one other tape with the marking. I think class I being 1/32” over 100ft. Milwaukee tapes with the finger hold gap are the best.
Having used a lot of different tape measures, i would say that the most i have ever seen them be off by is around 1mm per meter compared to each other obviously, Which is not at all bad, its 1 meter per kilometer, its 1/1000. Even the really cheap ones they just tend to last a lot less time in my experience. Its not hard to figure out how they do it either, they buy a precision tape measure to create their own ones and if you sell just a thousand tape measures the price of a precision one isnt that much.
I want to give you a metaphor about calibrating something inherently inaccurate to high state of accuracy to illustrate how silly this is, but this IS the metaphor.
probably my bias talking but, as a machinist, ive had good luck with my starrett imported tapes. just the cheapie ones off of amazon. def seem to be tighter tolerance than fat max or the like. not that anyone using a construction tape needs +/- .003" for anything, but still.
@@vankuipland this is true and I thank you. But I would argue that real tool and die makers are a subset of mechanical engineering. It's like half of the same discipline at least, just a more functional approach.
half a mm off at 4 feet seems perfectly acceptable when the mark you're making with it is probably a 3mm wide sharpie tip, or 1-3mm wide pencil depending on how long since you sharpened it, lol. Impressive for nearly 100 years old.
Ok I came here on Tom Scott's recommendation and I'm sure the rest of the video will be fun, but the third "we'll talk about that later in the video" after 1:40 playtime is a bit obnoxious.
Whats the point of raising the tape? It will never be used that way so shouldnt that error be accounted for in the scale? Also theres no way to ever use the little hook on the end and put a 5kg load on it it will pull off every time so that seems an incredibly useless standars
I use a stanley at work to hold +/- .010" tolerances over 60"+ on insanely expensive hydraulic systems going into some very well known excavating equipment
At this point everyone knows (or does now) that the thumbs down button is as useful as the door close button on an elevator. (They only work for firefighters with a key.) Basically, you don't see anything change, but it can get recorded by the admin. Sometimes.
close door button does work in my office. if I do nothing it takes about 4 seconds longer for the door to close compared to pushing the close door button. (push floor button + close door vs push only floor button) and yes I did measure that, hey I am somebody who watches this type of videos....
Imperial (or better said US Customary) is defined by SI units for length (and much more). The standard for metrology in most cases is 20C the world over.
Just curious. I understand it’s climate controlled but why are these incredibly precise tests not performed in a clean room? Where ingress and egress are controlled, real time temperatures are monitored, etc. And why is the calibration apparatus not constructed of a stone material such a granite? Something that has a much higher thermal resistivity. I designed high precision laser metrology devices and tools in my younger years and these things and many more were norms in that world. When having the ability to measure to the 50 millionths of an inch, I could be looking at a measurement of a ball bearing under a laser and could tell you when someone opened the door to the room because the temperature of the ball would change and it’s diameter would change.
He kept saying “calibrate” the tape. Except you can’t calibrate a tape after its manufactured. You can check the calibration, but outside of reprinting it, there’s nothing to recalibrate.
Nist surely failed with their treatment of 9 / 11 . . . BUT , on 'important things' , I'm sure they can do admirable work ! EDIT : Way back in the STONE AGE , ( 1970's ) . . . I worked to 4th order U.S.G.S. survey standards . . . * Take a circle , divide it into 6400 units . . . THEN work to a nine digit decimal 'mantissa' . . . THAT would be the billionth part of one sixty four hundredth of a circle ! * * Then there is the atmospheric temperature and humidity to consider . . . Yes kids , it's CHILD'S PLAY ! ! ! *** These days , it is done by electronics / lasers . . . I did it the 'old fashioned way' : with a pencil ! Children , can you say 'accurate' ? Other 'fun stuff' : faceting gemstones to an accuracy of a single wavelength of light . . . that's a WHOPPING 7 to 11 millionths of an inch ! *** This kind of accuracy can be done by 'eye' . . . 20 X magnification , and good lighting . . . along with a 'steady hand' ! ! ! ( check polishing OFTEN , or you may miss THE MARK by a millionth or two ! ) DAMN THOSE SCRATCHES ! ! !
Build your house in the summer. The thermal expansion of the tape will result in more house for the money.
Build your house in the winter, the wood's cte is much higher than the tape's. Cutting in winter will give you more wood at nominal Temps while cutting in summer will give you less.
@@Inconspicuoustwrex Ahhh which is it
@@Inconspicuoustwrex Huh, I never expected that to be the case, but I looked it up and it looks like wood does indeed have a roughtly three times higher coefficient of thermal expansion than spring steel (if I didn't missread anything by an order of magnitude).
So I ended up looking it up, and my base assumption was wrong. Wood has anisotropic material properties due to its grain structure. It has about twice the cte as steel, 30 micrometer per meter per degree C for wood vs 16 micrometer per meter per degree C for steel, in the radial direction of the tree, tree getting fatter, while having 5 times less cte as steel, 3 micrometer per meter per degree C for wood vs 16 micrometer per meter per degree C for steel, in the direction of the grain, tree getting taller. We cut timber to length along the grain direction, so cutting in summer would result it more wood than in winter.
If we are cutting 1 meter of wood when it is 30 degrees C outside, our tape measure would read 1 meter but would actually be 1.00016 meters long. If we then cut the wood, it would be a piece of the same 1.00016 length. Now, if we wait till Temps cool to a more nominal 20-degree C, that 1.00016 m piece of wood would shrink to 1.0001299952 m, which is about 130 micrometers more wood than if you had cut it initially at 20 degree C.
If we do this when it's 10 degree C out, the 1 meter tape measure would actually be 0.99984 meters long, and when measured at 20 degrees C the wood would end up being 0.9998100048 meters long.
Cutting your wood at 30 degrees C results in 0.0003199904 m more wood that cutting your wood at 10 degrees C.
@@Inconspicuoustwrex This is the sort of curiosity and pedantry which makes me occasionally have a bit of faith in humanity. Hats off to you
I like how NIST's address is a nice round number. Wonder what the tolerance on that is
+/- .003"
1ppb
Would a letter addressed to, say, _101_ Bureau Drive, successfully get delivered to NIST? 105? That seems like a rather fun experiment to try tbh.
There’s a 1 in their address, and that digit doesn’t appear to have any curves in it.
@@PetesGuide There's that angle to it. But of the two blocks that Bureau Drive runs, NIST is the entirety of one of them, and the Subway just barely outside is all the way down at 6 Bureau Drive.
Touring NIST would probably sound very boring to the average person but my god I bet its exciting as hell
@Fatvod I had the opportunity to tour NIST about 35 years or so ago -- it was _fascinating_ to see what they were up to and could do!
Metrology people are weird for sure.
I should know, I am that people.
@@prgnify but it's the _right_ kind of weird.
(This coming from an electronics engineer 🤓)
@@somedutchguy7582 btw, before it's too late;
Happy Metrology Day!
There’s an entire video done by a a UA-cam channel about NIST, I’m trying to recall which one it was. Look for it, it’s quite fascinating, especially the mega heavy weight they have for calibration of mega capacity scales.
In case anybody was wondering, "NIST traceable" basically means the manufacturer (or a 3rd party calibration service) sent their equipment to NIST, and then they use the calibration report and NIST approved procedures to calibrate things. This can actually go several degrees of separation, but you can always trace back to NIST.
We are the knights that go "NIST"!
but what is the airspeed velocity of an unladen swallow?!?
@@vanguardcycle Unknown, but we do know that the airspeed of a resting parrot is exactly zero.
Reading comments from my comfy chair.
Imagine their shrubbery out front
@@AttilaAsztalos of course he's resting. he's dead. i have a dead parrot
nice to see an upload from you again. its been a while. not sure if you know 'inheritance machining' but they mostly focus on accuracy and machining tools and other machines that make other things. would love to see a collaboration between you two.
This needs an upvote
Love that dude- he's the reason I'm going into machining lol
The most surprising thing about this video to me was how you used the tape as a compass to scribe a circle at the start. Neat trick!
Being someone who's spent time in a metrology lab, I know just how bad tape measures can be. Seeing that you can get NIST certified ones has be a bit to excited.
Your last 5 words "be a bit to excited" being corrected to "to be a bit exciting" would be very satisfying.
As an engineer who thinks that engineering as a whole is horrendously underestimated by the general public its easy to see why metrology is even more obscure because most engineers take calibration for granted.
I work in industrial control systems and its staggering just how much we rely on the manufacturers getting things right. We simply take it on face value that the manufacturers made every pressure, flow, temperature,...... sensor accurately and that they measure what they claim to measure at the scale on the box. So at the practical application level we are separated from the metrology people by the manufacturing layer.
Ever since I saw MT's vid on the Jo Blocks that was followed by his more recent one on the NIST I have gained a lot more appreciation for metrology and wish it was covered a lot better at the college level.
When siding my house with my dad, he ran the chop saw, and I was getting pretty frustrated that his cuts were coming up 1/4 short.
Turns out we were both exact on the mark with different tape measures...
Way back, I did a summer job in a metrology lab. It was really interesting and gave me an instinct to question "how accurately do we really know what we think we know". Was also useful for helping me think about various ways experiments could be a bit off, which has served me well over the decades.
It probably should be a required topic for any engineer and recommend for scientists to learn about.
@@travcollier Obligatory mention here of Simon Winchester's book _The Perfectionists._
If anyone is curious, NIST SOPs 10-12 has some great guidance on calibrating rules. And of course NIST has the best single axis vision bench in the world.
What an amazing experience, I’m so happy you were able to record this. I’ve been dying to see an in depth tour of this. Thank you so much.
At my work the tension we use on our tapes is roughly whatever we feel is adequate that day.
At my work the tension is whatever the boss’s mood is. 😁
It bugs me no end that NIST uses a weight measure to specify the tension, when a force measure is what they meant to specify. But then, US customary units have always been very confused about this particular distinction.
@@MrCrackbearexactly. It's fortunate that the distinction generally doesn't affect anything practical since a pound mass exerts a pound force under standard gravitational conditions
Does this mean we can look forwad to more than 2 Machine Thinking videos a year? How exciting of a prospect!
Hopefully.
my dad has one of those Stanley tapes from my grand father, he managed a local paint store after WW2 and it was a give away cause grandpa still had like a half a box of them in the 80's before he died
I remember as a kid trying to "fix" the wiggle on one of my Dad's measuring tapes.
Sigh. I did. Thankfully it was from my 'own' set of tools so it only hurt me. I was in like middle/high school. "This tape measure is garbage! Can't believe it's worn itself so loose." Took a punch or something and hammered the rivets tight. Wasn't until years later...oh no.
My dad would holler at me when I let his tape roll its self in and SLAM on on the wiggly hook. I'd say but you do it look it's already loose! 🤷
Same here. I decided to super glue it to take all that "slop" out. Luckily before the end of the project I had let it slam when it retracted a few times and it broke my "repair"
The amount of critical systems our society keeps perpetually running on Windows XP will never cease to get a chuckle out of me.
Nuclear missile silos often run on floppy discs.
Why not use a microcontroller for the computations and any operating system for the display?
I would love to see how NIST analyses modem theodolites for land surveying. We also have to take into account temperature and barometric pressure because they really do matter over thousands of feet or miles. Appreciate you.
Did you happen to ask Mr. Lee what model of tape measure was the first one you tested? He said it was his personal one that stays in his desk.
I would be very interested in what model it was.
Love your channel. Always looking forward to seeing more.
Brilliant. I’m always fascinated by all your vids.
What kind of inch is that old tape using? The spacing could be 48.02276 international inches, or 48.022664 survey inches!
Tape measures do have accuracy classes, most of the tapes you see at the big box stores are considered class 2, you can order class 1 tapes like Fisco or Advent. I have found as far as standard home depot tapes that Milwaukee and Dewalt were less accurate than the Stanley's (as boring as they are they are a staple in the category).
Makes sense, Stanley used to make precision machinist's gauges and scales, Milwaukee and Dewalt did not.
Oddly enough, Stanley/Black+Decker own DeWalt so I’m sure there’s some sharing of manufacturing techniques, engineering, etc. Stanley certainly has more experience historically. Milwaukee and Ryobi are owned by the same parent company, and they don’t necessarily share the same reputation either.
@@wcropp1 Then why do Black & Decker tools suck so much? (lol)
I genuinely delight in how the UA-cam algorithm throws up such “left field” stuff from time to time. What’s even more surprising is what I discover about myself in the process. Thanks to this video, I now know I’m a measurement nerd! 😂
That wiggle is calibrated to the thickness of the hook to allow for push and pull measurements to read equally.
Missed you bro. Made my day.
I'm a former calibration tech in the Marines and for Lockheed at Stennis Space Center, and I used to calibrate tape measures for the satellite blanket department using our Renishaw and HP laser interferometers.
I'll never forget the first time one came through the lab. At first we thought it a bit odd that someone would need a type 1 lab to calibrate a tape measure, until we found out why.
I'm also from Denver, and grew up listening to the "atomic clock" from NIST on the phone using a local number back in the 80's, which kick started my career.
Made many visits to NIST, and met with a few of them, throughout my career as well.
Small world.
I love this channel! Thank you!
Okay... I have a question that's been driving me nuts. Sticking with freedom units, ya know on tapes how the whole number has a thicker mark? Where is that value relative to the mark? I thought centre, but when I compared the tape vs. my 6" Starrett ruler, it shows the whole value is at the beginning of the mark. Is there an industry standard? Cause every time I bring a new ruler into the shop, I go through this insane making cross-checking.
That is interesting, no joke, In repair or in existing work that would be relavent but in new construction does it matter? Everything is relative, if everything in a newly constructed biulding that was measured at 12' 6" was really 12' 6.1 " I dont think there would be a problem. When I was a helper as a young guy I always did check my tape measure against my bosses' so I could give him what he asked for.
Working at NIST has got to be a dream job for science and engineering nerds
Not sure about that. Because NIST is very specialist, only certain types of scientists and engineers want to go down that road. There's lots of scientists that jump around to adjacent fields because of how much cross disciplinary research there is nowadays. I don't know how much of that can happen in NIST.
One of the channels we need more than a 👍 button, need a love ❤️ button.😊 Thank you for sharing, we love this channel.
The mysteries of the universe are hidden in the folding tape measure.
I like this kind of thing way more than someone should.
The latter fun fact was something I didn’t learn until the last few years, but it totally changed my view of measuring tapes.
I never took them seriously below the 1/4 inch level or so because I always assumed the looseness and presence of the hook on the tape assured a minimum error.
A machine thinking upload? Truly a red letter day
10:12 - "If you didn't like it, really let me know by giving it the double thumbs down"
How do you give a "double thumbs down" ??? Or, if meant in the logical/obvious way, by clicking thumbs down once to indicate a thumbs down rating, followed by clicking thumbs down a second time to remove the thumbs down rating, how would this be distinguished from not clicking at all (no rating)?
It would be interesting to measure a tape at 120-deg F and 0-deg F to see how much of a difference their laser can detect.
Will depend how long it is. Steel's coefficient of thermal expansion is well known. I assume since you used Fahrenheit we are using customary units. The thermal expansion of steel is somewhere between .000006 and .00000694 in/in/F. So for a 25 foot tape measurer you can expect it to be between 0.216 and 0.2498 inches longer at 120 degrees than at 0 degrees. The laser interferometer was showing a measurements to the 100,000th of an inch in the video. Even a tenth of a degree Fahrenheit would cause a difference of 0.0002 inches over 25 feet. Easily detectable with this instrumentation. Which is why they control and measure the temperature in their lab so precisely, and probably the imprecision of temperature control and measurement over such a long space is a larger contributor to their uncertainty calculations than anything to do with the instruments.
This is something considered in calibration labs. The reported temperature of dimensional calibration is important to know. You can theoretically determine the expansion/contraction, but if you really need to know you should always perform the experiment. Most would just calculate with a CTE for the expansion/contraction, and estimate the uncertainty. Detecting the difference would be trivial for the laser. Their accuracy is in the millionths of an inch or less, considering optimal operating conditions.
That's funny you posted this with the McMaster Carr site. I work in an aerospace machine shop, we make parts for SpaceX and Blue Origin. We just bought one of those Starrett certified tape measures for our shipping/receiving to do receiving inspection on raw materials. The certificate says it is +/- 1/32 over 12-feet.
I wonder about the uncertainty measured to microns in 9:09 when the comparison is made just by visual inspection. The visual inspection shown in 3:53 will not provide such precision. Maybe that is why it is not called deviation but uncertainty.
Love your videos
4:20 A computer could verify the accuracy of _Every_ _Single_ _Graduation_ _Mark_ on a ruler or tape in a few minutes. I'm very surprised that they are still doing this "old-school".
Just how traceable is that 5kg tensioning masspiece. Has it been verified against any of the standard kilograms, which are obsolete, but which are very small due to their density. I have some masspieces that are traceable to it with only a few steps. Verified against a balance, that was in turn verified by a replica masspiece, which in turn was referenced against one of the prototype kilogram units that were sent all over the world. 1g, to the milligram though you can measure the mass change that occurs due to corrosion from the atmosphere, and the tiny mass lost due to handling.
Here at my work the tolerance is +/- 5mm. So i measure with my fingers instead
Great video, it was very interesting to see a legacy tape on the calibration machine, I would be curious to see how it compares to a consumer grade tape measure from a box store.
I’ve often wondered how NIST does these tests if the enity requesting it needs a temperature outside of 20°C. They can’t change the temperature in the lab because it takes weeks to adjust it
I got in trouble as a machinist once for making parts too short.
I said no way, I made them correct within tolerance.
Turned out at one metre and before my tape was great.
At two metres and beyond my tape was great.
Between one and two metres the error grew to 1mm then decreased again.
My parts were 1mm short at at 1.5m
We stretch the tape measure out along a 2 metre steel ruler and couldn’t believe what we were seeing.
It was a Stanley brand. Needless to say they sent me out a free refill for the tape. Scrapped parts were worth 10x that.
However I was off the hook 🤠
Is the center of the tick mark what we should align to?
Did anyone get weird audio artifacts whenever people talked? Like a buzzing high pitch overtone of what was being said.
1:25 Didn't you yourself say that since inch is defined as 25.4mm, it's just a funny metric unit? =)
That hook to the 2 inch mark accuracy is probably more important to the accuracy of that tape than the next 48 inches. In the sheet metal sheer shop my company would only allow a few different brands of tapes and we had to check them on a standard ruler before sheering tons of stainless for production.
Assuming 1/16 is the finest division that's extrapolates to under 1 division over the length, I consider anything good to give or take 1 count full scale to be dead on.
Cool. I've got one of those little round tapes around here, somewhere. Never looked at the markings on the case before.
Unfortunately, you do not get the world best tape meter which is the Class I branded BMI, from Germany. It is also the lightest, sturdiest, with a lot of perks, like self expanding, stays flat, etc.
Just bought an inch decimal (10ths) tape measure, something I've been meaning to do for 20 years. thanks for the inspiration !
“Double-click thumb’s down” 😁😁 (see what you’re doing there) 😉
So... what's the tape measure that the NIST guy keeps in his drawer?
i work at ESZ Germany, can confirm this is amazing, also at 1:00 the statement should come with a "untill you drop it 😂😂
Tip for next time: check your audio levels when you start recording. The audio was clipping a bit in this video.
This video measured up, and was right on the mark. 😃👍
😂😂
I would like to know more about how the tape measure was printed to begin with.
a video on optical interferometry wouldn't be a bad idea. I'm assuming that an infra-red laser was used for this work, with a wavelength on the order 800 or 900 nm. My guess is that the reflected beam is compared (i.e. added) to the transmitted beam, and the max and min are counted as the reflector is slid down the track? I've heard of interferometry being done with white light, and I can't imagine how it works with the range of wavelengths found in white light.
I mean, a light can *look* white, but still have sharp emission bands that you could (in theory) use for interferometry. I'm honestly not sure why you would want to do this, however.
It occurs to me that a completely different application of interferometry would be to measure the thickness of a transparent coating. The idea is that the surface and the bottom both reflect some light, and the reflected light interferes with itself. This is the same effect that causes that rainbow sheen on oily water. By observing the exact wavelength that is reflected back the strongest, you can determine the thickness of the coating, and of course you need white light to make this measurement.
not even ten seconds into the video, and my life has been changed forever. my tape measure can be used to make a radius?
I remember a project that needed graduated tapes for aligning conveyer tracks. Nothing super accurate, but it was surprising the error between three measuring tapes from the same package had 2mm discrepancy over one meter. Didn't matter here, but was quite startling.
Every Starrett tape I have has a Roman numeral 2 inside a flat oval apparently signifying it as Class 2 with a tolerance of +/- 3/32” over 100ft. But I have only seen one other tape with the marking. I think class I being 1/32” over 100ft. Milwaukee tapes with the finger hold gap are the best.
Having used a lot of different tape measures, i would say that the most i have ever seen them be off by is around 1mm per meter compared to each other obviously, Which is not at all bad, its 1 meter per kilometer, its 1/1000. Even the really cheap ones they just tend to last a lot less time in my experience.
Its not hard to figure out how they do it either, they buy a precision tape measure to create their own ones and if you sell just a thousand tape measures the price of a precision one isnt that much.
I want to give you a metaphor about calibrating something inherently inaccurate to high state of accuracy to illustrate how silly this is, but this IS the metaphor.
So you measure it to the middle of the line not the start or end!?
I want a tape measure that's both extremely accurate and looks like it's not accurate at all.
XP used so it does not shut down to add Edge and Edge spywear. Haha. Have you lost control of your windows yet?
probably my bias talking but, as a machinist, ive had good luck with my starrett imported tapes. just the cheapie ones off of amazon. def seem to be tighter tolerance than fat max or the like. not that anyone using a construction tape needs +/- .003" for anything, but still.
You will find plenty in Swinleys forest just south of Bracknell, Berkshire
Just GRATE. Thankyou!
Metrology is such a big part of science and engineering, but doesn’t get much attention, especially from the theoreticians lol. Cool stuff 👍
Monday (May 20th) is worlds metrology day
How did they do it before the laser?
did anyone say, digital tape measures 👀
talking about precision and showing footage of somebody measuring depth to water in a monitoring well is a good bit lol
it's actually windows XP because that was before they put all the spyware and backdoors in
Mechanical engineers build the world.
Yet are paid and treated like crap
hello? Civil engineers? we had bridges and plumbing long before standardized/precise measurement systems
@@suibora Then grasp the subtext of "mechanical engineers build the modern world"
And us toolmakers ( tool and die makers ) are responsible for all those tools etc. that make this world go round
@@vankuipland this is true and I thank you. But I would argue that real tool and die makers are a subset of mechanical engineering. It's like half of the same discipline at least, just a more functional approach.
half a mm off at 4 feet seems perfectly acceptable when the mark you're making with it is probably a 3mm wide sharpie tip, or 1-3mm wide pencil depending on how long since you sharpened it, lol. Impressive for nearly 100 years old.
i would guess the tape measure is accurate to 30 thou over 4 ft, if it has not been used excessively
Ok I came here on Tom Scott's recommendation and I'm sure the rest of the video will be fun, but the third "we'll talk about that later in the video" after 1:40 playtime is a bit obnoxious.
Whats the point of raising the tape? It will never be used that way so shouldnt that error be accounted for in the scale?
Also theres no way to ever use the little hook on the end and put a 5kg load on it it will pull off every time so that seems an incredibly useless standars
That's from the Heart of East Texas - Tyler. That's not just "somewhere in Texas"!
i am waiting to watch the new tally ho video so i can watch this guy talk about tape measures.
almost a year and the two videos are clickbait-y low effort titles and ideas
Great Video :o-)
im guessing the old tape measure is more spot on then the new ones
I use a stanley at work to hold +/- .010" tolerances over 60"+ on insanely expensive hydraulic systems going into some very well known excavating equipment
Fascinating!
Good for you. Well done. Get yourself a treat.
This wasn't as in depth as most of your videos but thumbs up anyway.
Not as amazung as your normal videos, but interesting none the less. I gave a thumbs up.
At this point everyone knows (or does now) that the thumbs down button is as useful as the door close button on an elevator. (They only work for firefighters with a key.) Basically, you don't see anything change, but it can get recorded by the admin. Sometimes.
close door button does work in my office. if I do nothing it takes about 4 seconds longer for the door to close compared to pushing the close door button. (push floor button + close door vs push only floor button)
and yes I did measure that, hey I am somebody who watches this type of videos....
@@Blackadder75 yup, some do work.
Seems like there's room for error when positioning the carriage with the microscope.
It's fascinating that someone in USA deep underground is using meters and Celsius scale for measuring tapes in imperial units.
Imperial (or better said US Customary) is defined by SI units for length (and much more). The standard for metrology in most cases is 20C the world over.
I think the old tape will be one of the better ones
Just curious. I understand it’s climate controlled but why are these incredibly precise tests not performed in a clean room? Where ingress and egress are controlled, real time temperatures are monitored, etc.
And why is the calibration apparatus not constructed of a stone material such a granite? Something that has a much higher thermal resistivity.
I designed high precision laser metrology devices and tools in my younger years and these things and many more were norms in that world.
When having the ability to measure to the 50 millionths of an inch, I could be looking at a measurement of a ball bearing under a laser and could tell you when someone opened the door to the room because the temperature of the ball would change and it’s diameter would change.
Its always weird to see stuff from East Texas in a video in a place nowhere near east Texas lol
He kept saying “calibrate” the tape. Except you can’t calibrate a tape after its manufactured. You can check the calibration, but outside of reprinting it, there’s nothing to recalibrate.
Nuclear fabrication has some wild requirements for their tools, and trying to substitute items during shortages are like pulling teeth
Nist surely failed with their treatment of 9 / 11 . . . BUT , on 'important things' , I'm sure they can do admirable work !
EDIT : Way back in the STONE AGE , ( 1970's ) . . . I worked to 4th order U.S.G.S. survey standards . . .
* Take a circle , divide it into 6400 units . . . THEN work to a nine digit decimal 'mantissa' . . . THAT would be the billionth part of one sixty four hundredth of a circle ! * * Then there is the atmospheric temperature and humidity to consider . . . Yes kids , it's CHILD'S PLAY ! ! ! *** These days , it is done by electronics / lasers . . . I did it the 'old fashioned way' : with a pencil ! Children , can you say 'accurate' ?
Other 'fun stuff' : faceting gemstones to an accuracy of a single wavelength of light . . . that's a WHOPPING 7 to 11 millionths of an inch ! *** This kind of accuracy can be done by 'eye' . . . 20 X magnification , and good lighting . . . along with a 'steady hand' ! ! ! ( check polishing OFTEN , or you may miss THE MARK by a millionth or two ! ) DAMN THOSE SCRATCHES ! ! !
liked this so much, I gave it a double thumbs up 🫣
The contractors are all using laser gizmos now.