What Time is it Anyway? Clocks, Timescales, and How the World Decides What Time It Is

first of all thank you for inviting me these are a lot of fun to give like to give these talks I'm going to talk to you about clocks what is a clock what is time how do we keep it how do we decide what time it is um you're going to come to know there's going to be a lot of agreements a lot of sort of human interaction involved and we're going to go all the way through atomic clocks now I work for the Naval Observatory it's a US government agency so the next slide is really boring it's my disclaimer so I'm here to talk to you about clocks we're going to have a lot of fun but nothing that I say should be construed as the opinions or policy of the US government or the Navy or anyone um so first a road map I'm going to question your assumptions about time and clocks and how you might think about them what we're going to build them then we'll start going through history I'll talk about early history of clocks I'll spend a little while talking about longitude then we'll move into time zones increasingly modern clocks and then how the world starts to agree about how we're going to keep time and then we'll segue into the U.S Naval Observatory and atomic clock Atomic fountains which is sort of what I do and then how that applies to the military in keeping time and finally we'll talk about time transfer so first question here is what time is it well the simple question it the simple answer is it's what your clock says um but in some sense it's a very unsatisfying answer it it depends on whether your clock is running whether I can even tell you what time it is and and how did you set it what did you set it against um and they're going to be more questions when you start digging in a little further like what um what did you set it against and why did you trust it what are the units are they seconds can you define that I know some of you can but we'll we'll get to that we'll Define seconds um how well did your clock tick and keep these time units that I'm not going to let you define yet since you said it and so then I'm going to claim we need to think about what a clock actually is okay so what's a clock then a simple answer it's an oscillator and a counter okay oscillators are just repeating features of nature so it's something that comes back to an initial state so the Earth spinning and orbiting is a repeating feature of nature it's sort of the classic repeating feature of nature hey it's light again something oscillating something repeated in itself um your pulse is another oscillator so it's a pendulum so is an atom but that's getting ahead of ourselves that'll be later in the talk so you've picked your oscillator now now what do you do you have to count repetitions so we're going to have a counter we're going to count these repetitions we're going to store them there's going to be hands on a dial those can be marked off marks on a wall we'll call that a calendar eventually if you're talking about electronic device it's going to be the state of some electronic register inside of your device one big problem is going to be in the construction to Mechanical clocks and that is that you don't mess with your oscillator while you're counting it the process of counting a mechanical oscillator is often very disruptive to the oscillator itself in general we're going to call this an escapement so now we have our counter we have our that's tied to an oscillator so what's the time well it's whatever number is in your counter and yes that's a very unsatisfying answer the units depend on what you counted so maybe it's days maybe you kind of when the sun came up again maybe it Seasons it's when the wheat started to grow again or when the ice melted maybe it's seconds but wrapped up in all of this is we must agree on conventions we must agree on what we're counting from and what we're counting with and how we're counting so we agree to count on some point from some point in the past and we're in agree upon the units and we're going to agree on how to realize them so we're going to start to build artifact standards so having an agreement that says oh we're going to count this following way is at the end we have to build actual physical clocks and count with them so a convention yep a convention it's it's exactly that squishy it's exactly that inexact so an additional complication to fold into the whole issue here because I talked about we can philosophically Define what clocks are how we're going to count what we're going to use for our system of times now we have to build our physical clocks and physical clocks always diverge from each other in the time they're going to tell you they always always drift apart so the standard saw in the clock construction industry like me is that you know it's either very depressing or it's job security right so um so to wrap up again we have to agree on a reference and we have to coordinate then because our clocks and drift apart and this is going to be called time transfer there's a wonderful quote by Seneca here um it's actually a much more wonderful quote as it's written here if you go back and find the original it's actually about a one-page run-on sentence where he makes this comment in the middle but it's a wonderful um so what do we want in a clock we're going to have to choose we're gonna have to pick these conventions we're gonna have to pick our construction methods we want our very regular rate so we want to have it tick at a rate it's the same today as tomorrow we also want to have very small influences from the environment we'd like to not be sensitive to temperature or humidity or barometric pressure or whatever so when I suggested that your pulse might be an oscillator that you might want to use for a clock that's a really poor choice because that is almost the archetypical bad oscillator influenced by its environment there are much better oscillators to pick so you want a universal rate you'd like everyone to come up with the same definition of let's call it a second we'll we'll jump ahead and we'll call it a second we'd also like to be accessible so cheap would be good so everyone can have their own clock and not tied to geography we're going to see that's going to be a really difficult problem so now I'm going to start going through history once we've framed the problem the idea of what what we're going to try and solve so we'll start with early clocks the earliest clocks are called calendars um now a couple points of the at a couple points in this presentation they're going to be a little kind of well they're orange on my screen they're slightly less orange here um references that I will commend to you uh to read they're wonderful backgrounds and you're all in the library so go find the books read them they're wonderful um David Ewing Duncan wrote a book about calendars so wonderful interesting deep history that I won't cover so you a calendar is just simply marking off days on something so you mark off days the the repeating interval once again is very easiest when the sun came up next but this whole pattern has a larger cycle and you might want to detect that as well that might be very useful to you now a very crude way to detect that cycle would be to do something kind of macroscopic physical in your environment look and see when the ice melts in your pond when the river starts running now that's a very useful marker through a lot of history when the ice melts in your pond we'll probably give you a pretty good idea of when you can go down the river and catch fish when the fish will be running when you can plant your crops and they won't die to do better however though you need to start looking at the Suns and the Stars that's those are really much more precise ways of deciding when we've gone around and the orbit has repeated itself and there's the aha moment here Chris is coming from an observatory and this is going to be a theme Here the observatories are going to be important now once again labeling is going to be a pure convention what you call March is completely arbitrary what you call the beginning of your year is completely arbitrary and people long ago made choices that were completely arbitrary and may not be our choices Stonehenge is a calendar um going forward for with early clocks we'll have water fire in the earth now not the Earth rotating sand so water clocks you just have water drip out of container and you see how much water you have left or how much water you you've accumulated off here we have a water clock this is actually cascading water going down in buckets and there's a little Clapper that's part of the escapement that's going to bring when you get to a certain time that's a Chinese clock that's a current clock it's on display in an observatory in China you can have a an hourglass you're going to flip it over sand is going to go through and then if you have an attendant and this is what people used to do you have an attendant that's going to flip your hourglass back over and put a mark on it on the wall that can be a clock as well candles were used as clocks you would mark off hour markings and you would burn them during the night now in all instances these things are calibrated they have to be related back to what you're going to call for Truth for your time keeping system and at this point in history really it's been established that it's the Sun so these are all devices that get calibrated back to the Sun for truth the nice thing about these oscillators they don't require the sun to be out so they're used when it's dark mostly okay I've mentioned the sun's the best clock you have and at this point in history and we're still fairly early history sundials are going to be the most precise way to read out the time from the Sun sometimes are beautiful wonderful objects that are much more subtle than you might under might guess so if you want to figure out a local clock that's the sun you don't need a sundial if you all you want is sort of crude approximations of time you have Sunrise you have Sunset you have noon you can pretty much schedule a day by that if you have more complex day you might want to divide up into smaller increments hours say this is a sundial this is a um a horizontal Sundial so it's a flat surface this is called the gnomen this is what casts the Shadow on your Sundial it's always called a no man if it casts a shadow on a sundial and this Edge is tilted up from the surface that edge is actually aligned to be parallel to the rotation axis of the Earth and that's important that defines how you're going to cast the Shadows you can see that this one is claiming that it's almost one o'clock now this is however an inherently local time if you move someplace else on the earth to a different longitude at the same instant of time if you know we can allow me to coordinate time in ancient times the shadow will be at different places so it's inherently local it's a readout that depends on where you are um it however doesn't do it perfectly at various times of the year your Sundial will lie to you if you had a perfect reference to compare it against and this is actually what you will see the phenomena is called an analemma and it's sort of the answer to the question of it's a constant time of day why is the Sun not where I expect it to be the reason is because the Earth's orbit is not perfectly circular and we're not spinning normal to the Earth's orbit so the Earth's axis is tilted and then as the earth goes around the Sun the orbit is elliptical these two have necessary physical uh consequences that cause the sun to at different times of the year Trace out this figure this is a picture of what would happen this is a simulation there are some gorgeous gorgeous solar photography pictures online I however do not have permission to use them so I couldn't use them so this is a simulation this is all you know scientists faking things on the sky this is um where the sun would be this line is sort of the average location so that would be the average location you would draw on your Sundial but there's going to be errors and here's my one plot for the for the whole talk this is another way to express this is called the equation of time and almost everything I've been talking about here is referenced in these these displays you've seen back here so there actually is there's a clock that self-references for the equation of time back in the display not a clock a picture of a clock in a book so there are times of the year this is as you March through the year so this is that's 360 days so this is sort of one year from here to here and there's part of the Year where the Sundial is behind and part of the year when the sun dials ahead so if you want to build a really a better Sundial you correct using this information of the equation of time and it often looks like beautiful animals drawn on your Sundial or your gnomen looks like an analemma um the topic of sundials can go on for an hour talk by itself so I'm going to give you one other different type of Sundial this is a vertical Sundial and these things are gorgeous this is a tower in Norwich England this is a vertical Sundial this is the gnomen again it's aligned parallel to the Earth's rotation axis now let's this is a north or south facing dial this is an east or west facing dial and this is a gnome one that once again is aligned or parallel to the Earth's axis but now it's not sticking out of the page anymore it's parallel to this plate so the lines it draws are pretty much straight lines going across in fact on the bottom here we have patterns of as you go from a south facing dial which looks like that one to a North uh to a west facing dial which looks sort of like this one you get all these beautiful intermediate patterns as well sundials are like I said wonderful interesting topic um the next big step past sundials are pendulum pendulum clocks are very regular oscillators and they have some interesting features one of which is that they are Universal so they only depend on the local value of gravity and the length of the pendulum that that's all they can be powered by weight additionally which makes them unattended good oscillators uh Galileo is credited with realizing that they would be good oscillators not actually building good oscillators based on pendula that was uh Kristen Huygens and that's also referenced back in the in the display Galileo was claimed to have timed chandeliers in church and he actually used his pulse to time the chandeliers so it tells you that first of all he was probably very calm and second he was bored at the time so this is also interesting that it's a once again it's another clock comparison measurement that he made um turns out a pendulum isn't perfect and you can detect with a pendulum the Bulge in the earth the earth isn't round and the difference at sort of sea level on the equator between gravity at the equator and gravity elsewhere you can measure with a pendulum this is actually a gorgeous pendulum called refloor clock that's inside of a glass vacuum vessel there's a great book that talks not only about pendulum clocks but lots of other history of clocks and Time by Joel and Barnett times pendulum this is a picture of an escapement so I mentioned you it's very difficult to take a mechanical clock and get the information out and once again there's beautiful work on escapements back in the in the exhibit and this is a picture of a very complicated escapement called a grasshopper escape and I won't go into the details but basically this wheel is always trying to rotate and this escapement is stopping it and then it's going to grab with the other side and it's going to release it and it's going to let it move one tooth and it can do this in a way without perturbing the pendulum this is the pendulum that goes down to a bob that's cut off on the bottom of the screen and this in fact is the tick and talk you hear in a pendulum clock it's the escapement is something whacking against the escapement gear in your pendulum clock so now I'm going to go on to the first big technological and social problem that was solved by clocks and this is the longitude problem I'm going to skip to the bottom of the slide first you should come back here on October 11 and you should hear David Sobel talk about the longitude problem I'm going to talk to you for four slides about this and it's a great story so not only should you come in here or talk you should buy her book or check out her book it's called longitude if you love the book as much as I do you will buy the Illustrated longitude which is also a collaboration with Will Andrews who used to be the scientific instruments curator at the Harvard Library and it has unbelievable illustrations and wonderful explanations of the illustrations in it as well okay enough plugging so the longitude problem the longitude problem boils down to where equals when um finding your latitude and and we're actually talking about C here because I get pendulum on ground pretty good you know the ground isn't moving you know where down is pendulum depend on knowing where down is at C much much harder so latitude is relatively easy think about the pole star you can go find where the pole star is and you can measure angle down from the pole star and now you know your latitude your longitude however if you're going to go look at stars you need to know how much the Earth has turned out from underneath the Stars you need to know what time it is so the way people used to navigate is you would go down to a fixed latitude where you knew the island was out in the middle of the ocean and then you would say I'm sure I'm on the west side of the island and you sail East at constant latitude until you run into the island that's the way they would navigate um unfortunately your ability to figure out how close the island in is really is really terrible you're dead reckoning and you may run into the island during a storm at night in which case you're probably going to run into the island um horrible horrible loss of life loss of goods and some very famous crashes in England induces England to offer a prize in 19 in 1714 England offers a prize of 20 000 pounds that's huge I did a rough calculation it's about 50 million dollars today this is huge um so there's the launch group right now it's 20 million dollars sorry 20 minutes 50 million dollars 20 000 pounds for a half a degree on a great circle that's 30 miles that tells you how bad they were at this this is really crude and this is going to change the world so 30 miles huge that's only two minutes of time however and it must be practical and useful the test for the clock based Solutions were going to be that you had to sail to and from the West Indies Indies this takes six weeks and you must not air by more than three seconds per day all I did was I divided two minutes by six weeks to get three seconds per day this is really hard because there's going to be a horrible environment it's going to get cold it's probably going to get wet and you don't know where down is there are many competing astronomical methods and it was actually kind of a joke that a mechanical method like a clock could possibly solve this problem the astronomers are the timekeepers and they're sure that you're they're going to solve the problem with lunar eclipse methods or the phase of the rotation of the Jovian satellites so it's actually a real long shot that a basically uneducated clockmaker named John Harrison pulls it off he built several clocks over many years 1730 to 1772. one interesting note is over this time the longitude board becomes the first scientific funding agency keeping these efforts alive in various researchers along the way I mean this guy John Harris is a genius he invents many things along the way including the bi-metallic strip which is a clever Arrangement that allows you to have self-temperature compensating spring systems he invents the cage roller bearing as part of making his clocks better the story includes politics Intrigue scientific battle scientific misconduct and more um I'm not going to get into it so H4 the fourth Harrison clock finally passes the test it does it in two trials uh the first was in Jamaica the second to Barbados the first one in 1761 to 1762 is better than two minutes he's forced to do it again second time in 1764 also passes the test uh he finally ends up getting the prize in 1773. there's lots of like I said politics and Intrigue in the meantime after H5 also passes the test and King George III intervenes turns out King George III was a science and technology buff and was interested in the problem K1 K1 is the first Kendall clock it is a copy of H1 H4 which is this beautiful clock here it's sales with cook in the Pacific and was was um regarded by cook as the most valuable member of his crew so this this solves a huge huge problem this clock lives in the Greenwich museum is five inches across it's not a pocket watch it's actually a fairly large watch so that's the first big huge technological problem solved by clocks um here's another here's a convention problem we're going to get to now is time zones local times and rail this is actually going to be a Time problem that's brought on by technology in some sense now on land you have access to local time you have the sun you have access to a really good local time so let's take the case of England England is only minutes wide if you think about how wide it is east to west if you were to take London as ground proof for the time in England and that's what they tended to do it's sort of 15 minutes One Direction is seven the other so this is not an issue if your time transfer method is someone walking from one side of England to the other you'll just keep local time as you go first problems arise when you start having fast Coach Services because fast Coach Services want to tell you their service the way they're going to sell you their services they're going to tell you they're more on time than their competitor or they get you there faster and the comparisons are not well defined which time are you using for your start in your stop time between Norwich and London this really comes to a head with the rail travel with rail travel you have serious physical problems if you have a clock coming from norat shedding One Direction a clock coming from London any other direction and they're keeping a different time and you want to make sure they're not going to head down the same piece of rail at the same time so yes lots of trains run into each other and lots of people die um turns out that what then starts to happen is there's a phenomenon of keeping rail time and local time this actually leads to the production of dual face watches in England so you can keep track of what your local time is and what the rail time is by about 1848 the English rail gets together to decide this is a problem they're going to standardize on London time it takes until 1855 when most local time is stamped out in England it's replaced by Greenwich Mean Time now this is a relatively easy problem they have one time zone what about the us we're a lot wider than one hour um in the U.S the local railroads are the reason why we have problems and we with this local time and coordination of local time turns out the rail companies in the U.S keep their own time every company decides it's going to keep his time independently and how they choose tends to be the founding city of the rail line so the Baltimore and Washington Rail Line keeps Baltimore time and the Ohio and Ohio River we keep Ohio time so this is great quote here I'll let you read it but you know you can show up in a train station and it's quite possible you will have to deal with four separate times what your watch says from where you came from because that was your local time the rail time of the train you came in on the rail train time of the train that you're trying to catch and then the city is going to keep a different time because they're going to have their own local time um this is a serious problem now there was a talk here that you should have gone to on September 13th where this was covered there actually is interesting information that you can get to from the Linda Hall Library website that goes into more of this as well I read your website um so there are people that propose solutions for this so different time zone Solutions not coincidentally most of these people are railroad men Charles Dowd is the first person to really Champion having four time zones in the U.S this is in the 1860s initially his Meridian that he chose was DC meridian's going to be where you your reference point then he sort of wises up and picks Greenwich initially the implementation though is really clunky the way he wants to implement it is a table of literally 8 000 separate Corrections for what your local time is compared to The Standard Time if this is not good um Standard Time finally gets adopted it was actually championed vigorously by someone named Stanford Fleming this is also a railroadman unfortunately his efforts were destined it looked like to die until they were picked up by William Allen and William Allen campaigns for a standard time and the way he gets it done is he convinces the observatories to just change over so he doesn't go to the local time he just everyone's getting their time locally off of the observatories observatories look at the stars and they have time balls that they drop the time ball at New Year's Eve that's actually a really slow time ball time balls drop fast but that's what they're supposed to do you set off a cannon often as a warning you go and you look at the time ball and goes boom and you set your watch so he convinces all the observatories to move to Standard Time so on November 18 1883 at noon all the observatories dropped their time ball reference to Standard time and it's game over because now we have a Time transfer problem everyone trusts them everyone trust them is truth it just propagates National legal adoption has to wait till 1918 as part of daylight savings time initiative and that's actually a war effort initiative worldwide there's an international Meridian conference in 1884 and the proceedings of the international Meridian conversation 84 is in the room behind you as well they meet to discuss the coordination of world time they agreed to use Greenwich now the exact year that the um the country implements that varies and in fact the U.S is later than 1884 because that comes in a legal definition it has to pass governmental issues some countries take a long time and some of those stories are actually interesting one that's kind of humorous is France France doesn't like England being the definition they prefer Paris and in fact they insist on Paris and they Define Paris mean time it is however shifted to agree to Greenwich Mean Time okay um Liberia actually Waits until 72 to join until then they were resolutely 44 minutes and 30 seconds different from everyone else in the time zone another interesting tidbit China has one time zone they used to have time zones they don't anymore they have one time zone and there actually are local tolerated definitions of local time there is one area of China that just unofficially wink and Nod operates two hours off of Beijing time Hong Kong and Macau are special cases they're special administrative regions they have their own time um so those are sort of two big historical touch points now I'm going to go back through clock Technologies we're going to we're going to keep moving forward with better and better clocks through history this is the Pinnacle of pendulum clocks this is the short clock um I've mentioned that disturbing your oscillator is a big problem this is an effort to do the best possible job of having an undisturbed pendulum this metal can is evacuated it has a pendulum inside that's the master oscillator it has electronic feedback linking it between this master oscillator and this slave oscillator here the slave oscillator has an escapement on it the escapement is also electronic the advantage of using electronic escapements are you have very little back action on the pendulum these are good to hundreds of microseconds error for one day forward and prediction that's phenomenal for a a pendulum oscillator in fact these are used at usno from 1930 to 1945 that's really late that's really recent they're replaced by Quartz once usno we'll answer that in about four slides quartz clock quartz crystals that's the next technology that is the technology that displaces the pendulum quartz is piezoelectric Piezo means pressure electric means Electric so if you squish apa's electric Crystal you get an electric charge if you put electric charge on it it squishes um that's a really interesting property another interesting property is it's a single beautiful single Crystal that is very stiff so it will ring like a bell bells are good oscillators that's a sign of a good oscillator so take your disc of quartz and you ping it and it will ring very very pure tone the nice bit about a piezoelectric Bell is that I can ring it electronically and I can sense it electronically so this allows me to build an all electronic oscillating oscillating clock it's this electricity as a property is discovered in um 1880 the first oscillator is built in um just for 1920 and the first practical clocks are about 1927. but these really kill off the pendulum clock now we're going to move on atomic clocks so the thing that displaces knocks the quartz crystal off its Throne is the atomic clock now radar from World War II turns out to be the enabling technology radar requires microwave oscillators so we get the microwave technology out of places like the radiation Lab at MIT and that this is going to allow us to probe atoms we have a complicated slide here and this is a reference and this is an atom here and this is some stuff to tie it together so we have an oscillator let's say it's a quartz crystal and I'm going to multiply it up to microwaves this is the this frequency multiplier change is stuff that I got from the from radar World War II now I have microwaves I'm going to shine them on atoms think of atoms like little bells if you shout at them at the right frequency they will ring sympathetically and we can sense that using a whole bag of atomic physics tricks they teach us in grad school so Atomic physics happens here and we're going to be able to to sense whether or not we're shouting at the atom with the correct microwave frequency we get that information out and this just means steer this means correct we're going to correct this coarse oscillator so it agrees with the atoms and then we count the quartz oscillator that's an atomic clock all you do is you measure an atom's internal frequency they happen to be up in the microwave and you count them that's the atomic clock now everyone is a building atomic clock one thing I would like to to put out in public here they're not radioactive it has nothing to do with nuclear Decay it has nothing to do with radioactivity has nothing to nuclear power we're talking about light bulb energies here they're low power electron changes not Nuclear Physics changes in the atom so atomic clocks are not only better than coarse they're enormously better than quartz this leads to the downfall of the earth um well the downfall of the earth is our preferred timing mechanism so now we have much better clocks in the Earth and the Earth has been defining the seconds so the compromised solution up till now has been to define something called the mean solar day so you can define a couple ways you know a second is and I should have said a second is not a needle or day above those items a second is 1 over 86 400 of a day or it is one over about 31 million plus a bunch of other significant digits of the tropical year 1900 that's the mean solar day and that is the truth that is the second up until 1967. um yeah 1967 we moved from the mean solar day to something based on atoms and we choose cesium why do you season uh cesium has a nice high frequency ground state micro transition um ignore what a ground state that's a red herring never mind um it has an oscillating frequency inside of the atom that oscillates about 9.2 gigahertz 9.2 billion times per second um it's a heavy atom which for practical purposes means when I build an atomic beam it moves slowly so the atom stays around a lot longer inside this apparatus because it moves slowly the only reason I give you that detail is that that's the big lever I'm going to use for my Atomic fountains later um it's so a heavy atom nice high frequency the really beautiful thing about atoms is everyone seesium atom is the same they say it's the same here and in France and in Brazil it's the same thing so the world agrees that 91 92 631 770 oscillations of the ground state of those cesium-133 isotope is the second that's it that's your second now we've all agreed um that's the way it works we Define truth right we've defined our convention we've all agreed so now what about realizing that realizing the second is a the job of building a primary frequency standard a primary frequency standard is not my job I work for the Naval Observatory it really isn't um it's nist's job so why because the second is a base SI unit remember SI unit is the system international the the international units we use for describing everything including the meter right yeah right so defining these units is the job of nist that's the National Institute of Standards technology they have two campuses one's in Gaithersburg just north of DC one's in Boulder Colorado the Boulder Colorado places where they do the time and frequency stuff um so they are charged with maintaining the SI units for the nation because every nation is interested in knowing what the units are um so they're going to build a primary frequency standard so they make an oscillator based on the cesium-133 atoms got to be season 133 because then otherwise isn't the second now here are all the caveats here's what's hard unperturbed by Electric magnetic fields at sea level and at rest so you have to either correct four and zero out these perturbations or measure them and correct for them some of those seem kind of silly at sea level it turns out due to special relativity if I take a clock and it's sitting on the floor here and it's ticking a certain rate and I raise it one meter it just changed its frequency by a part in 10 to the 16. okay part number 16 who care right well these primary future standards are defined to a part 10 to the 15 they're actually better there are several parts until the minus 16. you need to know the elevation of your Atomic frequency standard to a meter or so if you're if you're going to be in the game um so institutions like nists Define the frequencies and they they build these devices these physical devices and they have calibrated frequencies good to better than a part in 10 to the 15. they're some of the most precisely known physical quantities full stop um now I've said a bunch of stuff about precisely defined and I haven't said the word accurate although this is an accuracy thing they're doing they're really realizing a quantity so I'm going to spend a slide on accuracy and precision um accuracy is how how well you have realized the measurement of a thing the actual value of a thing Precision is no matter how bad your measurement is is it the same bad measurement every time they're not mutually exclusive though the best way to illustrate this however is for those of us that do target shooting um we have a Target here and we're going to shoot at it so this pattern right here this is very precise it's very repeatable but it's the same wrong answer every time so it is precise but it's not accurate accurate would be centered this is not precise and it's also not accurate this is very accurate that could be a primary standard it just wouldn't be a very precise one you'd have to wait a long time to figure out the answer this one is both accurate and precise so this is the vision you should have when someone talks to you about an accurate measurement of something or a very precise thing um in general it's fair to say that nist is worried about accuracy and the Navy's worried about precision as long as everyone agrees and has the same very precise answer everything the Navy and everyone in the Navy services for time all the dod we can all operate so we care about Precision I won't talk about stability but basically stability is how much the target's moving out from underneath you um okay so now we've defined what time it is based on atoms with all agreed how we're going to do this how we're going to coordinate well we're going to coordinate through institutions that have been set up by the Treaty of the meter Treaty of the meter was signed in 1889. this is an international agreement about how we're going to deal with units and things and artifact standards the bipm is the international keeper of these units it's a little Enclave outside of Paris um that stands for Bureau International Bureau of weights and measures and the words are all French but the order is wrong for French and the lore is that the the compromise was that the acronym would make sense in nobody's language and and it's plausible if I tell you one other thing your dues as a nation to the bipm are a scaled version of your U.N dues it is that political thing um so what we've done is we've all agreed to send our clock data to the bipm so usno owns a bunch of clocks we'll measure them we'll send them over there we'll get to how we send them later the bipm then tells us what time it was and that's called Universal Coordinated Time and that's not a typo it's what time it was it's not what time it is we get the data back in one month batches a half a month later so on about the 15th of the month we're going to get a bunch of data so about the 15th of October we're going to get data telling us what time it was throughout the month of September honestly it's the way it works so if you want to know what time it is now you have to have your own clock that you're reporting to the bipm and you have to guess what time is going to be um so if you see UTC sub K where K is replaced by some institution's name that's that institution's best guess at what time it will be and UTC us and always our guess and once again I'm going to Define usno in a minute so we now have enough time also to take a quick detour there's gonna be a detour here in a detour in a moment on into leap seconds you guys have probably heard of leap seconds this is when we either on the transition um the first of the first month or the first the sixth month we add a second or subtract a second so why are we doing this uh it's the answer to a question the question is what do you do when two time scales disagree I told you that physical clocks always move apart from each other the two time scales that are disagreeing are the Earth's rotation and atomic time so the only answer is we've got to force one to agree with the other that's it if we want them to coordinate it's much easier to add a second to the world's atomic time than it is to change the phase of the earth so we decide to add a second or subtract a second from UTC um so the Earth is slowing down is the problem and it's the moon's fault this is really cool it's also our fault I'll get to that in a moment um it's the moon's fault the moon has exerts tidal forces on the earth and that's what sloshes the oceans around it also distorts the entire structure of the Earth there are solid earth Tides rolling past us and they're about a meter high um this is dissipative this isn't a lossy process so you dissipate energy in squishing the Earth and having this big Earth tide roll around and that changes the spin rate of the system it slows everything down so we add or subtract seconds to our time scale to make the earth spin rate the earth spin rate angle is called the ephemerity the ephemeris excuse me um to agree this is sometimes defended as a practice of keeping the sun overhead at noon to which anyone who is in a business like me will give it a big raspberry and say because you know you saw that in a Lemma right you saw the errors of the sun die you saw the errors where the sun is overhead So my answer is if the Sun is overhead at noon almost nowhere and at most four times per year right and your Anna Lemma is minutes wide um I can refer you to the U.S military's opinion on leap seconds which is that they should not continue that is that's policy I will tell you that you could look that up um so that's what's up with leap seconds now this is one of the more egregious examples of steering a really good time scale to a bad one okay now we're going to continue detouring in some sense uh detour in the sense that I'm not now going through history so much um I'm sort of giving some context of implementation and where I'm from and what I do so observatories in the usnl now from the start observatories have been at the center of time keep because the earth and the stars have been our best clock throughout history remember when we went over to standardized time we did that by co-opting all the observatories and getting them to collaborate and change the standard time and then everyone followed um everything from Stonehenge and sundials have been observatories we started having increasingly centralized efforts and the U.S Naval Observatory is the Department of defense's centralized effort at being an observatory and a timekeeper this is a picture of our building our institution this is actually our time ball we have a Time ball um we don't drop it regularly it was built for the special event that was you know 1989 to 2000 and for those of us that are actually time Keepers 2000 2001 which is when it actually happened right the account from one um think about it for a moment so that's a telescope Dome and that's a Time ball and we've dropped it a couple times and that's the main Administrative Building um so what do we do at usno we actually do more than tell the time we tell the time and we tell folks what time it is because if you only know what time it is you don't tell people it's useless we measure where the stars are and we make star charts and we tell people where the stars are and where they're going to be we measure where the Earth is pointed and where it's spinning where this angle is and how it's wobbling and tell people about that and how to predict that all of these are navigation products all of them are navigation products and that's why the Navy is interested and that's why we're in the business so now I'm going to focus on time service let's talk about time so we Define the time for the dod it's a definition of Truth for the dod whatever UTC usno says it is that is the time um it's a joint Chief instruction so UTC Uso is a master clock for the dod and all other DOD organizations and because of how ubiquitous our time is we end up being the de facto time for lots of other endeavors as well we figure out what time it is by measuring a whole bunch of physical clocks roughly a hundred there are three different construction types of clocks different types of clocks are better have different properties and you do better by averaging the optimal combination of the Apple types of clocks this is a really hard math problem and a hard physics problem and we're not getting into it but the interesting issue here the takeaway is times have voted quantity even at the observatory we're going to vote among a bunch of clocks and by vote I mean we're going to have a really smart mathematician right in a hard algorithm and then we distribute this to the users now more on how we do the distribution but the big punch line is think GPS GPS is a military system if you ask GPS what time it is they're going to tell you what we told them um okay so now I'm going to go to a topic around four slides they're near and dear to my heart it's sort of what my group does and and how we're pushing time keeping so this is sort of the end of the road for currently deployed really good clocks um it's called an atomic Fountain clock it's a microwave clock just like these primary standards I talked to you about in fact the current U.S primary frequency standards use this same technology and describe to you except they use cesium133 instead of Rubidium 87. there are a bunch of geeky technical details why I use rubidium 87 I can build a better precise clock with everybody maybe 7 that I can build um with season 133 and since I don't care about defining the second I don't I'm not constrained to use cesium so I pick rubidium its oscillation frequency is lower 6.8 billion cycles per second the key to all of this is laser Cooling laser cooling is Nifty see I avoided saying cool that's nice um so laser cooling is the practice of shining lasers at atoms in a vacuum chamber and shoving them around manipulating them this is an interesting and revolutionary enough idea that it was worth the Nobel Prize in 1997 to three folks um Bill Phillips from nist in Gaithersburg Claude kontanuji from France and Stephen Chu from Stanford who is our current Secretary of Energy um so the way that you shove atoms around with light first of all you want to use a laser beam because it's all the same color and it's all coming in the same direction so you have an atom and it absorbs a photon now a photon is the smallest chunk of light and there's lots of them in laser beam so when the atom absorbs the photon it absorbs the energy and it changes its internal State and we're just going to ignore that right now so it changes internal energy but it also recoils from getting the momentum from the photon photons carry momentum it's exactly like me sitting on a sled in Winter and someone nailing the snowball and I recoil it's exactly that process so the photon the atom now has recoiled a little bit not a whole lot of momentum in a photon it then remix the photon and I can then absorb another Photon now it remits the photon in a random Direction so I'm going to pull the physicist trick and average that to zero it doesn't sound like that would be very efficient or very useful because you put your hand in front of the projector and doesn't get shoved very hard by the photons but atoms are really light individual atoms and I can repeat this process doing the right Atomic physics games about a million times per second you work out the math and I can accelerate my atoms at a thousand G's you can really move them around then I'm going to play other Atomic physics games and I'm going to have laser beams come in from multiple directions and by playing games with the tunings the laser light and Doppler shifts I can make sure that the atom always absorbs photons from the laser beam that is directed so as to slow the atom down now I do it in three dimensions this is what's called an optical molasses it looks very viscous to the atoms I can cool the atoms down I can get them basically not moving I can get them so not moving that they're moving in about a couple centimeters per second now that doesn't sound very impressive centimeters per second but the nitrogen molecules in this room are walking into all of us at hundreds of miles an hour so it's really slow if I convert this to a temperature is a couple millionths of a degree above absolute zero it's Nifty stuff it was worth a Nobel Prize so what we do is we grab a ball about them all this happens inside a vacuum chamber so we grab a ball of Rubidium atoms and I can make them nice and cold and then I mentioned that I can accelerate them at a thousand G's so tossing them is easy I can toss them a little bit by messing with the light and I can keep them cold in their kind of Center or Mass frame so now I toss this cold ball of atoms up through a microwave cavity and they go up and they come back down and there's lots of my hand waving here at this point right so I'm gonna I'm doing what physicists tend to call using intuitive models using ball and stick bottles um it's lying right so I'm gonna lie very creatively and try and convey as much information as possible so the atoms go up through this microwave County and come back down and I can arrange things such that it looks like I'm asking the atom what his frequency is the whole time and I'm comparing to this microwave frequency then it's going to come back down I'm going to read out whether or not the um my microwave frequency is too high too low just right as arbitrated by the atoms by looking at how much light I see scattered with another laser that the atoms cross so I'm looking at the items for a long time that's a fundamental win that fundamental win allows me to love much much better clock um so that's an atomic Fountain clock this is hello there we go this is a cutaway of our vacuum chamber so this is a little more realistic um drawing of it so this is where we collect the atoms we toss them upwards as a micro cavity they float up they come back down we detect with some laser beams we actually use two laser beams see I was lying and um it's all inside of a vacuum chamber we have inside the vacuumed here because we don't want background gas atoms slapping into the atoms and knocking them out of the whole experiment like I said nitrogen molecules hundreds of miles an hour not good for having the atoms stick around we toss about 30 centimeters into the microwave cavity we detect with lasers and we repeat this whole process the grab the atoms cool them to a millionth of degree toss them measure them do all this sort of stuff every 1.2 seconds um continuing the theme of you have to isolate isolate isolate if you're building a clock we isolate everything to the systems we've already isolated putting in the side of a vacuum can that's important we're going to wrap them in magnetic Shields this is showing four layers of magnetic shields on the outside we regulate the temperature with the room we regulate the humidity because that's important and that allows us to have a very very stable clock I'm going to say I'm going to explain this from one I decided black body shifts black body shifts are the response of the atoms to the background radiation background thermal radiation um something else about us in this room not only 100 mile an hour molecules hitting us we're being bathed in radiation and light that's centered around 10 microns that's the light that thermal imaging cameras use to see things at night it's about 10 microns long whereas the light in the room that we're seeing is about a micron long about a half micron um this light that's all around us that's bathing the atoms even inside the vacuum chamber the walls the vacuum chamber are bathing the atoms in this light that's good for about two parts and 10 to the minus 14 frequency shift which is massive on our in this game so that's the black body so if you need a regular temperature okay this is a picture of an atomic Fountain so this is the outer magnetic Shield the physics package here we affectionately call this the water heater and then the only cool Atomic physics happens in there two racks of equipment and we have several of them running and have been for over a year and a half four of them running over a year and a half at the observatory and these are showing stabilities relative to each other at about the part in 10 to the 16 level that's nanoseconds per year nanosecond 10 to the minus nine ten minus nine seconds per year and the moment we're going to talk about why we care about nanoseconds um I'm getting closer now I just talked about a whole bunch of atomic physics and um this would be a little more applicable if there were more young folks in the audience so carry this to All the Young Folks you know this is a short advertisement um I just talked a bunch about Atomic physics and stuff but I'm not coming from a university so I'm a physicist I did my undergraduate and my PhD um in experimental Atomic physics and getting a PhD was not the road to riches I have a brother that's a lawyer um but I get to work on really interesting problems really interesting work and I get to play with unbelievable toys I mean really cool toys and very few of us that go down this road of being an undergraduate and getting their PhD in physics end up as professors but that's okay but realize that going in that that's what you're doing okay time transfer back in the talk um we're almost done so time transfers the act of coordinating comparing times so this has come up before you've seen this throughout you already know about time transfer the trains and their passengers were a Time transfer mechanism that drove us to standardized time and and have time zones the longitude clocks or time transfer mechanisms they transfer time from where you were at a known longitude to another time and that time transfer allowed you to determine to interpret the error as your longitude time balls you've seen astronomical observations at different observatories our time transfer mechanisms so yes or no we do time transfer as well we have a voice announcer and call that number it will tell you what time it is it'll be good to about a tenth of a second that's about how well you can pick you can listen to a phone and um and you'll hit a button and such we get about four million callers a year slightly more impressive is Network time protocol that's computers asking other computer servers like ones we have at the observatory uh what time it is that will time up a computer about 10 milliseconds so 100th of a second we get about 18 000 requests per second big Peaks on the hour huge Peaks after a leap second this averages to about a half a trillion requests per year GPS is one of our most ubiquitous time transfer mechanisms we tell GPS what time it is GPS tells you what time it thinks it is it's very close to what time we think it is and in fact if you have a sophisticated timing receiver encoded in the message from GPS are the corrections to allow you to get even closer to what time it is for us what time we say and finally there's two-way satellite time transfer that's our highest test time transfer mechanism we set up a communication satellite dish we bounce off a communication satellite over to you with your um Telecom dish and we do experiments back and forth in the same direction at the same time and we cancel out delays and we get things down to about a nanosecond there are very few High test users in the military that care about that okay now the time transfer it turns out is everywhere so you're surrounded by time transfer all the time time transfers as simple as telling someone what time it is or setting your watch sitting here watching guess what you have to decide what you trust household electronics are always transferring time to you your holdover oscillator is kind of remembering what time it is for those of us who remember VCRs yeah um your computer probably checks via ntp your cell phone is a Time transfer mechanism all cell towers have GPS receivers many self Towers have atomic clocks in them that's how they officially hand off calls as you're walking around the city they're driving around the city your GPS receiver is an Exquisite time transfer devices so it's everywhere around you so I mentioned GPS a couple times so everyone's favorite application is GPS um GPS is a constellation of satellites fully functional as 24 satellites it's it's uh six orbital planes four satellites per oil plane this is a little Jiff of you know you sitting on the earth rotating around and the satellites you can see this is actually telling you how many satellites are visible at any given time um the GPS constellation is a whole bunch of satellites all of which have atomic clocks on them multiple atomic clocks and they're all shouting the time at you they're all shouting tick at you and your GPS receiver is measuring the different delays from the different satellites the different ticks so the differences are forming rulers where the units are delayed now the way they're broadcasting all this information is microwave signals micro signals are electromagnetic radiation which is light which travels at the speed of light so a nanosecond is a foot light travels this far enough in a nanosecond so that's why we care about defining things in nanoseconds at the observatory are these sorts of applications and that's how GPS works is literally your GPS receiver which is everywhere now right measuring all of these little rulers to all of these satellites that are whizzing around over our heads all of which are shouting tick at you so time to round up circling back around here are the takeaways I guess clocks or oscillators hook to counters it's that simple and time what time it is is largely a human convention and we build artifacts to implement those conventions but it's largely a human convention clocks have evolved enormously throughout history and a lot of the drive to improve clocks or many of the things that have come out of clocks have been important technological or social problems and the high Precision type keeping touches all of us every day in many ways and with that I would like to thank you and I would be happy to entertain your questions [Applause] if you have a question I'll come by with a microphone recently we had an extremely large earthquake and the question is do you get involved with how that might change the shape of the earth and time rotation uh the question sorry we have microphone the questions I need to repeat it um we do have a department called Earth orientation those are the folks that determine where the Earth is pointed they did look at that um you I think the takeaway was you might have been able to see a small short-term perturbation but there's nothing long-term in the clocks that are defined by the Earth's spin rate and angle and orientation much larger effects are surprisingly things like weather the the distribution of water in the atmosphere changes the spin rate enormously those are much larger Dr ekstrom thank you for a wonderful presentation I noticed that almost everything you said is global Centric how do what's the convention for establishing the time on the International Space Station that's just going around and also on the Curiosity Rover on Mars where obviously you have to be concerned about time thank you um I'm going to have to plead ignorance on some of this uh what you've asked about the International Space Station at a crude level my guess is at the second level you can just pick a location and you can do the Transformations the Transformations aren't that hard at a very high Precision level I've actually been involved with with analyzing possible clock experiments to the ISS um it's a very subtle problem because the transfer the reference frames Transformations are not well defined because this rotating frame in the in the minkowski Transformations don't hold for rotating systems you're smiling you knew the answer um uh so at the highest Precision level it's a subtle hard problem that you can solve at about the part and 10 to the 16 level um maybe part 1017 level um from what I know that that's a couple years old data for the Curiosity Rover I honestly have no idea what time reference they're using we have a question over here maybe okay okay so someone has mentioned that they're they're using the Martian day and that sounds completely possible yeah and answer with any Authority I can just repeat what you told me any other questions oh I was wondering if you'd be able to say anything about how standardization of time affects the standardization of other physical quantities there are other SI units that are based that you can base off of time um those were a bit of a moving Target but for instance the meter is defined using base units of time or frequency it used to be determined in terms of a Krypton line um I believe that's been overcome by events and I know that you can get to physical measurements through time and frequency based measurements that end up improving uh limits on physical quantities and sometimes you can back those back into base units um but really the meter is the only one I can give you with any sort of authority and that comes at the speed of light which is defined I've got two questions the first has to do with daylight savings time is it worth all the time and trouble that causes us every six months and the second one is I've heard Neil deGrasse Tyson twice say that global positioning satellites do not have the leap seconds added to them and thus they're like 15 seconds off from a fish from official time but I haven't noticed that in whenever I check times is that true or not um GPS time I'll ask them in um we'll see first global positioning system global positioning system keeps its own time scale it's called GPS time uh it added leap seconds early on in its life and no longer does it is indeed many seconds off of Coordinated Time your GPS receiver is just correcting for you and um sorry the first one again daylight savings time um oh military systems ignore that I mean we you changed that for muster so the military answer is you know nothing we do with the observatory reflects Daylight Savings Time besides our voice announcer our voice answer will say Standard time or whatever um in some sense you're asking a human convenience versus political will question um for the daylight savings time personally I find Daylight Savings Time advantageous but I don't live in an agrarian area nor do I have an agrarian job where it becomes a real pain so in some sense I can't answer that I mean those are clearly the trade-offs are sort of agrarian activities where that daylight might be better used at a different time of day or you know lighting and heating considerations where it becomes an economic benefit which is why we had the recent change doctor um if you fly and you're on Zulu time and I didn't hear you mentioned Zulu tonight Zulu I thought was all around the globe whether you're Chinese American British Australian Zulu comment um that would be a standardized time so it would be like GMT or what was GMT what is now UTC um so often The Way Way times are written is you actually will reference the offset so I think what you're asking is you know how do you keep track of the offsets or why don't we just have a common time all the way around the globe and in some sense we do so it will be it was UTC and if I'm referencing a local time here the convenience time I will reference is you know it's it's 9 13. but then there'll be an addendum to it which is UTC plus seven so I'm just hiding the math so in effect we do but our local conventions for convenience and the fact that when I I flew from DC to here and I expect it to be light at noon um we just added an offset to make it convenient for human convenience but underlying that is the standardized time I'm not sure if that answered you about this Dr extra more here on your left we have time for two more questions how was it established that the day was divided into 86 400 units as opposed to something like 100 000 or something else wow um I think I knew that at one point if if I had to if I were told I I have homework and I have to go come up with that tomorrow I would go to David Ewing Duncan's book on calendars and I would go there to look I don't know your answer but that's where I would look I'm sorry digital watch um I have a very nice uh quartz crystal based watch with hands on it it also does have a digital display for doing other things on it but it's I use the hands [Laughter] and we'll end with that one thank you and thank you for attending tonight's lecture please join us October 11th Davis Sobel on longitude and we will also have a book signing that evening following the lecture courtesy rainy day books and please stop by and uh visit the on-time exhibition again which will be on view through March 2013. thank you and good night

Channel: Linda Hall Library

Views: 4,192

Keywords: time, clocks, timescales, physics, U.S. Naval Observatory, science, engineering, technology, astronomy

Id: 7VCuvnLCGC8

Length: 69min 7sec (4147 seconds)

Published: Thu May 11 2023