My time machine needs to know when it is

Question

I've built a time machine and it works in unmanned testing... sorta. I tell it to go into the past, it disappears, and then comes back to the present. The problem is that it isn't terribly accurate. I need to know how far back it's going in order to calibrate the "Time Selector" dial.

I need a device that can determine the time across millions of years. So far I've used a radio clock hooked up to the onboard computer, but I want to send the machine back more than a few years. I have a pretty big compartment (6x6x7 feet) to fit the device. I'm still in the unmanned testing phase, so everything has to be automated. I can put sensors outside the vehicle (star detectors, antennas, etc). My time machine stays in the past on a 24 hour timer. Assume there is no multiverse (it makes testing harder). And assume that all of my testing is being performed in a desert area of the American southwest.

Answer 1

If you can deploy a good quality, wide-angle telescope and take photos of the night sky over a period of more than a few days, you can determine the time down to seconds over a wide range of history. It would probably be good in the range of a million years pastward or futureward.

The technique would be, first, to get a basic orientation by finding recognizable stars. We know what stars are bright and close enough to have significant proper motion (PM) and what stars are bright (probably better than 8th magnitude) and far enough away to have little PM. This gives us a set of bright stars whose proper motion we know well enough to have known tracks through the sky both forward and backward from today. We spend significant computer cycles to find the set of bright stars in the photos which are consistent with those motions. This gives us a date that's good to within 1000 years or so. Perhaps more importantly, it gives us a solid coordinate system, since some of the sources are extragalactic and don't move significantly in a million years.

We can cross-check this with the measured locations of the poles over the range of dates where the precession is predictable.

Given that, we then can do a fitting of the observed motion of the Moon and planets to that period. The motion of the planets should tie the date down to a few days, and the motion of the Moon ties it down to a few minutes.

Sources of error: (1) It's much easier if you can be sure of getting extragalactic point sources. Some of the LMC sources would be great if you're in the southern hemisphere. There's probably some decent sources in Andromeda, though it may not be realistic to have a telescope big enough to resolve them. Best would be if you could get down to magnitude 13 and pick up some quasars such as 3C273. (This ought not to be hard.)

(2) I don't have a solid number for the rate at which planetary motion prediction become too unreliable. Given that we don't see chaos for a hundred million years into the future, I'm estimating that position calculations will be accurate for at least a million years.

(3) The device needs to be stable over the period of observation and have a view of the night sky.

Answer 2

Instead of having the device determine the time it is sent to, have it leave something in place, that you can find in the present to determine exactly when in the past it traveled. This is best if your location is very isolated (or underground in a stable rock formation) to minimize the chance of disturbance during the intervening million years.

The simplest I can think of is to leave some known amount of long term radioisotopes and use a detector in the present to determine the amount of decay, something similar to how carbon dating works. You should choose isotopes with appropriate half-lives to allow you to very accurately calibrate how long the sample has aged.

If you want multiple trips, you should ensure that the isotopes are well sealed in a container (fused in glass or other long term storage method) to prevent contaminating the whole area, so you can measure multiple samples.

Answer 3

You need to know when you are sending stuff, not how much time they stayed in the past (from their perspective). So sending clocks is a no-no.

One way you could do it is by collecting air samples from the past. When it comes back, compare the concentration of atmospheric gases against those of Antarctica's ice cores. The oldest ones are 2.7 million years old.

If you need to go further back, you may wish to measure the distance from the Earth to the Moon with a radio transceiver. The Moon is getting farther away from the Earth at a steady, known pace (3.8 cm/year). You may need to wait a full lunar month to get your distances right - if you wish your machine to spend less time in the past, you can also measure the time an Earth day takes in the past. Eath days are getting steadily longer with time (2 milliseconds per century, also mentioned in the last link). You just need to measure the time between two consecutives sunrises.

Answer 4

Assuming you can't take a star fix for whatever reason (during some of these time periods, Earth's surface could be pretty hazardous, and I'm not sure what precautions your time machine uses to avoid popping up in the middle of a rock, or a cave, or a sea of lava), your probe should still be able to take measurements of the atmosphere and magnetic field around it. Then when it gets back to your time, you'll have to compare that data to known values from Earth's past.

The good news is that we carry out the same process on fossils and rock strata all the time, in the study of stratigraphy. All the reference data you would need is well-studied. The bad news is that it's much less accurate than an astronomical approach, probably on the order of tens to hundreds of thousands of years. However, it has the advantage of allowing a much tougher probe compared to a delicate telescope.

You could use the two techniques together, if you felt the need. Stratigraphic fixes would let you safely determine a very rough timeframe, which will tell you roughly what geography to expect. For instance, during some past eras, much of North America was covered by a shallow inland sea - not exactly ideal for astronomy. Knowing when your probe will pop up allows you to choose a location you know will be suitable for the more delicate equipment. It also gives you a good feel for what the atmosphere will be like, so your telescope probe can take that into account.

Answer 5

I'd say go back as far as you could possibly want to and leave an atomic clock (or several for redundancy).

When it wears out / breaks replace it with another. Eventually you'll have a continuous line of clocks from the dawn of time

Once you have a line of clocks right up to modern time, you can tell exactly when you are down to the microsecond.

You can leave it on land, bury it or have it floating around in space or even on the moon.

Answer 6

Similar to measuring star positions but not reliant on any particular set of stars/location. Use the expansion of the universe.

This should provide you with a known rate which you could plug into your time selector using hind- or fore- casting algorithms.

You have a time travel device, I'm assuming you have something powerful and portable enough to measure or record two distant parts of the universe, to get a reading.

Answer 7

So, here's a suggestion which should cover a wide swath of destinations. NASA recently developed a method of locating any spacecraft to within a few miles within the solar system, by using the position of known pulsars. This is great for location in space (perhaps well outside of just the solar system), but the catch is that they are, gradually, changing their frequency. The Crab Pulsar, as an example, is spinning down at a rate of 3.7e-10 Hz/s. By identifying these pulsars, comparing their relative frequencies with those in the previous temporal position, and mapping their position (given that their velocities are also well known), you could get a reasonably precise location from hard-to-miss stellar bodies.

What would be even better is to combine the data regarding their spin-down (or spin-up, if you're going into the past) with trajectory distances and perhaps additional clues such as broader star movement and geomagnetic data, minimizing error in the result by taking a weighted average of known error. This would all be without the need for an internal device, beyond simple precision sensors, and could get you pretty good precision. The only points at which it would fail would be extremes where the pulsars weren't formed yet (which is pretty exhaustingly far out), or in locations far enough from known space where the pulsars could not easily be found in the sky.

...I would suggest filling the remaining space with a comfy couch and an entertainment system for the trip.

Answer 8

Since the earth and the sun move and rotate through space, too, I'm afraid you'd have to use a space-time machine, leaving you only with the ability to test whether you have hit the space-time coordinates you're aiming for.

One test-run would then look like this:

• Calculate where in the universe the earth would be at the time you choose.
• Send your space-time machine to these space-time coordinates.
• The machine tries to find earth and measures the distance to it while triangulating its own position using known stars.
• After returning, the data gives you your error margin.

Of course, you'd have to account for accidents like "there is molten lava outside the time machine" or "the time machine is a mile high up in the air and about to crash" or "the time machine has a vastly different velocity vector than the earth".

Or, you get around all that and assume your time machine is gravity-bound the whole time it is in transit.

Answer 9

Genetic analysis of ground life form

First, sample the DNA of as many commun ground dweling fungis (or any eukaryotes) as you can.

If you have no access to the night sky, you could dig to find the same species (more exactly their ancestor) and measure the number of change, especially in the Non-coding DNA. This can give you some estimation of the time between your departure and "now"

This method can work while traveling in the past or travelling the future. But I am afraid it can not tell you if you travelled 1000 years in the past OR 1000 years in the future.

Safe travel to you!

Answer 10

Use Radioactive Decay

Similar to Josh's answer, If you are capable of building a time machine, you should be able to build a small automated robotic device that can simply move out of the time machine, take a soil sample, then return to the time machine and perform analysis on the soil using spectroscopy.

Almost all soil around the world will have a small amount of natural uranium that you can analyze to check the ratios of U-238 and U-235. This is one method for determining the age of the earth, and it should do well to determine how far back your time machine went when comparing to modern ratios.

Answer 11

1) Send your time machine from the present to random times in the past when there were already radio broadcasts.

2) Measure the used fuel every time.

3) Wait until one of the radio broadcasts tells you the time for determining the exact time you have arrived in.

4) Calculate the relationship of time travelled and fuel used.

5) Once you have a working equation, test your time machine with travelling between past times to test whether the equation holds up if your point of origin isn't the present, and rework the equation if needed.

Answer 12

Are there any trees nearby? Dendrochronology could be used on species like bristlecone pine that live thousands of years. Beyond that, sediment layers in mountain ranges. You could potentially have a record of tree rings going back 10,000 years if the research has been patched together.

My main concern with time travel though is that you would move into the future or the past at the exact spot in spacetime, but the earth and solar system relative to that spot would be millions of miles away, or more. Maybe you have to aim the time ship in the direction of where the earth is at the destination’s time.

Answer 13

Continental drift

I think a cheap way of doing it, yet giving a more or less precise reading, is to look at the position of the continents. They are moving slowly, and we have a rather good idea of where they were at which time, as well as their speed of travel.

You could send an high altitude balloon to get pictures and process them when the machine gets back. Try to have a system that sends the pictures from the balloon to your machine, it's easier than to retrieve the balloon.

Answer 14

Use the star map for short time, use the expansion of universe for longer time

For a few million years

Other answer deal with it pretty well. You can see the star map at night, you can compare air sample with air trapped in Antartica (works only for travel in the past) or human radio emision (work only in near future)

From 100 million years to a few billion years: red shift of distant pulsar

Even if you cannot predict the exact trajectory of star/galaxies on the very long run, you'll still be able to locate a few pulsar.

Even after several billion years, the frequency of their pulse should not change much.

And if you have a doubt, their relative position should also be similar. Most distant object will not move by several degree even in 10 billion years, so a pattern of 5 pulsar will still look the same( Worst case scenario, there are only 4 left).

So, you'll be able to spot a few pulsar. You know the wave lenth they emit nowadays. By measuring their wave length somewhere in the past/future tells you how much the universe did contract/expand.

Crunch a few numbers and it will give you the date.

Edit: This works better if you travel in the future as in a distant past, the pulsar may not exist yet.

For eternity: Cosmic microwave background

Curently, the Cosmic microwave background has a temperature of 2.72548±0.00057 K

This decrease with time as universe expands.

So a few computation will tell you the date.

Problem: to measure this for earth surface, with current technology you'd need a 10m wide instrument. And the dificulty increase when you travel in the future.

Answer 15

1) Air pollution.

I believe the composition of trapped atmospheric gases in Greenland ice cores have be used to detect minute changes in Earth's atmosphere over decades, centuries and millennia, This has been used to chart the rise and fall of various industries over the ages as they pollute the atmosphere.

So your time calibration package should include an automatic air sampler to bring back an atmospheric sample you can test in your laboratory to see how well it fits in with the record of Atmospheric composition changes in past years, decades, centuries and millennia. If the time machine goes back to a time within the range of scientific studies of atmospheric composition from ice core samples you should know the date within a few years or decades.

2) If you have some sort of automatic robotic optical and/or radio telescopes in your time calibration package, they can go outside and measure the important optical and radio sources during the daytime and the nighttime.

The Moon can always been seen when above the horizon, day or night, except when hidden by clouds. The direction to the Moon, and the phase of the Moon, can be measured very accurately. The phases of the Moon change over a period of 29.6306 days.

The planets Mercury, Venus, Mars, Jupiter, and Saturn would always be visible to the naked eye when above the horizon during nighttime, and sometimes would be among the brightest objects in the sky - Venus, Mars, and Jupiter can sometimes be seen during the day.

If the telescope can tell the difference between a planet and a star, it can identify those planets that are above the horizon. So the directions to the planets, and their phases, will show the positions of the planets relative to the Earth, at the time the observations are made.

For each body in the solar system, there is a synodic period, the period between the times when it's angle relative to Earth is the same. For Mercury the synodic period is 115.88 days, for Venus 583.9 days, for Mars 779.9 days, for Jupiter 398.9 days, and for Saturn 378.1 days.

So if the automatic telescopes are able to find the directions to three or four planets, that will show a relatively rare configuration. A length of time that three or four synodic different periods could fit in as integers and not fractions would be a very long period of time in which each synodic period was repeated many times over. And then if that is combined with the phase of the Moon with its own cycle, that becomes much rarer.

Then there are the apparent positions of the four Galilean moons of Jupiter, which would be easy to detect. They have their own synodic periods in relation to each other. That will make the day that your time machine spends in the past even more unique.

So astronomical observations should make it easy to identify the 24 hour period that your time machine spent in the past or the future, for as long as astronomical orbital simulations are accurate.

3) For a much longer time span, the Sun takes about 200,000,000 years to make a full orbit around the galactic center. The center of the galaxy is the first astronomical radio source ever discovered, so it should be easy for a radio telescope to find the direction to it, if it is above the horizon. Several external galaxies, such as the large and Small Magellanic Clouds, M31 in Andromeda, and M33 in Triangulum, are bright enough to be seen with the naked eye and should be easy for a small telescope to detect. The galaxies Centaurus A and M87 are also famous radio sources that should be easy to detect with a small radio telescope.

So it should be relatively easy to detect the directions to both the galactic center and one or more external galaxies, and thus determine the position of the solar system in it's orbit around the galactic center, and the nearest past or future eras that could be.

4) Send a Mars rover type robot out to collect rock samples (different rock samples in different eras, of course, since you don't want time paradoxes). Most minerals contain trace amounts of radioactive isotopes which decay into other isotopes at a rate that can be calculated. You can test the samples in your lab for the ratios of various isotopes, and if the samples are collected from far enough in the past you should be able to calculate the approximate date.

And there are many similar astronomical and geophysical methods that you can used to detect the date on various long, medium, or short, timescales, and to calibrate various methods against each other to improve accuracy.

Answer 16

The radius of the Sun shrinks by 74 centimeters per year.

The fusion reaction of the Sun converts mass into energy and overtime it loses mass. You could measure the diameter of the Sun to determine the current year.

This approach has the advantage that the time machine can appear in any moment in history and calculate the current year. Assuming the rate of change in the size of the Sun is constant. It becomes a linear calculation of how much time has changed.

I doubt you'll be able to calculate the exact year, but if you have the technology to build a time machine you can surely build a device to accurately measure the diameter of the Sun.

Here is a reference to Sun shrinkage from Stanford University.

http://solar-center.stanford.edu/FAQ/Qshrink.html

Also, if you take a photograph of the Moon's surface. This image defines a timestamp in history. Comparing two images of the Moon's surface will tell you which image came first. As new impacts overlap old impacts. The more rapidly the surface of the Moon changes the further back in time you've gone. This isn't going to help you over small distances, but if you don't know how far your time machine has travelled, then taking an image of the Moon would be a quick way of knowing if it was a long distance.

Answer 17

I want to extrapolate on an earlier answer and comment:

Use an atomic clock.

Set the "Time Selector" dial to whereever you want to randomly send the machine, then have the automated systems deposit the atomic clock in the past. Recover the time machine. Turn the dial by, say, 1/100 of a degree forward (or adjust the dial however you can adjust it while measuring the change). Go back and read the time on the clock.

If the clock is old and rusted by the second jump, change the dial so it's only 1/1000 of a degree forward and try again. Keep using smaller calibrations until you get an usable reading from the clock. Now you have two data points; preferrably far apart in time.

Then, keep sending back the time machine to random times between your two endpoints. If you keep gathering samples, you will eventually have enough data points to approximate an equation to describe how much time a turn of the dial represents (is it linear, or is it some other strange model?). Find where this equation intersects with the present day, and this will allow you to find the numerical year that your time machine landed; given a good enough approximation, you could theoretically pinpoint your calibration to the nanosecond.

This method has the advantage of not needing to use measurements from uncertain sources (no sky readings, no DNA samples of external organisms, etc.), but only if you can overcome the obstacle of landing in the same place over a large amount of time (preferably the same desert).

Answer 18

1. Measure the luminosity and the size of the sun, and thereby derive its age (and in turn, via comparison with current-day values for the same measurements, find a temporal position).

2. Similar to the answer that mentions measuring the rate of expansion for the universe, you can measure the redshift of gravitational waves & compare to current-day values.

3. Again similar to the above, measure the background radiation and calculate time based on the CMB "trajectory".

For a craft of any kind, space-faring or otherwise, option 1 (out of the ones mentioned here) is probably the most feasible in respect to size of necessary equipment - and can probably easily be done with equipment you can also use for some of the other answers here, like the top-voted answer that suggests imaging the nightsky, meaning you can do several different measurement types and compare the results.

Answer 19

I would recommend tracking the movement of key stars in the sky. Pick the correct ones, and you save yourself the need to track them all while still forming a comprehensive picture of the night sky that should not appear perfectly identical between two given times. This does limit the device to being used under an open sky, primarily at night: for obvious reasons, this isn't going to work inside a cave or a building. You're going to need a good telescope on your time machine as well, and probably regular inspections to check its calibration and accuracy (if it's unable to detect and repair this automatically).

Or you can just use an atomic clock, if simply having precise timekeeping is enough in the absence of outside references. Those are accurate to within a few seconds every million years, and I don't think your eventual time-travellers are going to be too concerned about being off by a couple of minutes.

Answer 20

The most accurate way of doing this would be the Stars, but many people have suggested this so its best to see their answers, but i'll offer an alternate suggestion

They'll be a device within this time machine which when more power or [insert plot device] is passed through it controls the amount of time traveled, for personal reasons i'll call this the Flux Capacitor FC for short.

If you don't know what this device is you need to rig up several sensors to every component and run the below tests for each part just to ensure you have the right component.

Once you know which part the Flux Capacitor is, then run the below tests.

Test 1: Measure the power to the FC upon embarkation very precisely with several sensors, and take star measurements. and place a radioactive isotope on the ground. return to present

Test 2: repeat test 1 to ensure that the correct sensor readings, if Stars match test from Test 1 (taking planet rotation into account of course) then you have a good basis to believe that your testing is accurate so far and you measuring the relevant FC settings is correct, and that your machine has the ability to accurately repeat a destination. in theory the radioactive isotope should read the same levels of emitted radiation as when you left it there as it would have only been x amount of time between transits

Test 2.5: in theory it would be advisable to run test 2 several more times to be 100% certain

Test 3: Reduce the FC's power slightly and embark once again, take star measurements once again. but also scan for the radio isotope, if found measure the radiation emitted from it, it should be less, on the assumption that less power into the Time Machine/FC would mean a shorter distance back in time traveled. measure how much less and this should give you an idea of how much power difference changes the amount of time traveled.

Test 4: Repeat this transit at the same power level. always double and triple check your findings with multiple identical tests using the same power levels

Test 4.5: return to previous power levels and run test to ensure that you return to the previously established moment in time

Test 5: further reduce the power into the FC and perform the same test as Test 3, again this should give you a good idea of power change

Test 6: increase the power past the initial starting point to and area your previous readings would suggest is say... 1000 years before Test 1 landed. and then place the Radioactive isotope down again and then return to the present

Test 7, return to Test 1 levels and detect the isotope from an already established moment in time, confirm settings are as predicted and adjust for any changes

Again repeat repeat repeat, these tests should give you an idea as to whether the the power requires are exponential/logarithmic as you want further back, or incremental, once you know this then its fairly simple maths to figure out, I want to go X years into the past i need Y amount of power, this allows you to know where you are going

its better to know where you're going, because then you can use Star mapping as others have suggested to confirm you are where you expect to be.

Just do us all a favour and bring a hazmat suit with you. not for you, but so you don't bring a modern illness into the past.

Answer 21

Deploy a robot that travels to the past and buries a known quantity of a radioactive substance in the soil (in a properly shielded and tamper-evident container, to avoid detection and trouble with the EPA and such). In the present, dig up the container and measure how much of the substance has decayed. Depending on which element you choose, you should be able to determine when the sample was buried -and thus, how far back the machine went- to a fair degree of accuracy. You might even bury several samples of different elements, just in case strange time-travel effects might alter some of them but not others.

Answer 22

Assumptions

If you want to calibrate it, I assume you already have a way to make it go further or closer, and just need to know by how much. For example, you have a dial that goes up and down, but you have no idea if one rotation is a year or a million years.

I also assume that you are using a fixed timeline (stable time loop) model for time travel, because you specifically mention how there is no multiverse, and changing the past would make paradoxes by the dozen.

First Step

First, you find an object that has been in place for, let's say, five thousand years. There are such objects on the earth, and while you might have a few legal issues bringing the time machine there if it is big, you should be able to work around it. A great example would be Stonehenge. You setup the machine to scan for a specific Stonehenge rock, or maybe the entire pattern, and do some trial and error to find a dial setting where Stonehenge exists.

If your machine has the potential to go back as far as the big bang or even before, you're in trouble regardless, because the odds of you reaching the earth are so small you might as well give up now. If your range is around the age of the earth, you will need at most 908 600 attempts, so on average 454300 attempts, to find a time period where Stonehenge exists (Stonehenge has existed for 5000 years, and the earth has existed for 4 543 000 000 years).

The trick

Now it might already seem like way too many attempts. But remember, if you can automate scanning for Stonehenge, you can definitely automate trying again by changing the dial a bit. And since your machine is a time machine, it can always come back, after scanning or even after 24h in the past, to one millisecond after it left, therefore completing all those 500 000 scans (or so) in 500 seconds, or just a few minutes.

Bonus: You can even add to your story the fact that your machine's attempts throughout the last million years are what caused people to make Stonehenge in the first place, because a strange device kept spawning around that place and acting weird, then disappearing. This can either be a fun fact or a plot point.

Next Steps

Once you've found at least a dial setting containing Stonehenge, you can check the number of attempts that have been required, and do a rough estimate of your dial's step size. Now you can do the same operation, but with an object that has existed for just a few years, but is much easier to scan for. Take the largest tree behind a house that has been in your family for a few generations. Explain grandma's irational alien sightings at the same time, and do the exact same process, with a much smaller increment on the dial. This should take about the same number of tries, but it is much less public and problematic.

Once you've found the tree, you can again roughly estimate (but with much better precision) the dial's increment size with your two measures. You can just repeat with a one month old object that doesn't move. If there is none, you can spare a month making it, just make sure to ignore the random time machines that show up during that month of waiting.

Figuring it out

You repeat until you have a few very specific measures on the dial. I suggest more than two, just in case the progression isn't linear but something weirder like square, exponential or sin. About ten good precise measures, once you found a range of dial that is easily testable like an hour, should give you all the data you need.

The potential problems

• If your machine can also go in the future, you might "miss" the 5000 year window by a lot if your increment is a million years. If Stonehenge doesn't last that long in our future, then you might test billions of years in the future, never finding anything.

  Fix: If you don't find anything after billions of attempts, just start over with a smaller increment

• After a certain number of attempts, your machine may not come back. Maybe it got destroyed during the scan in the past, or worse, captured. Since no one in the past had time-travel inhibiting tech, you can assume it got at least disabled completely, maybe by people, animals, climate or environment.

  Fix: Add a self-destruction mechanism with a dead man's switch. If the running code either fails or stops running at any point, wait 1 minute, then blow up. Make sure the blow up is enormous enough to annihilate everything, to avoid giving modern tech to our ancestors. It doesn't matter if you kill one of a hundred of them since it's a stable time loop. Also make sure you have the machine write a log every time it pops back in the present, so you know exactly where it was so you can resume testing from there with the next one. Have very precise plans to build it again if required.

• Your machine doesn't compensate for earth's rotation, movement, the solar system's movement, the galaxy's velocity, etc

  Fix: Yeah you're screwed. Handwave something and compensate for earth velocity.

Answer 23

Are you only looking for ways to test based on the local environment? And are you trying to avoid more handwavium than a time machine already requires?

Assuming yours is the only time machine, you could conceivably set yourself up a "time beacon". Set up a device in your lab that syncs with a similar device in your craft (quantum entanglement sounds sexy without actually meaning anything). By comparing the settings of the two devices, you could determine not only distance in miles from your point of origin, but in years.

Such devices have been used in other stories. The Legion of Super-Heroes stories has a time beacon set up in the 31st century.

Answer 24

There is a much easier and safer way.. Program it to send and respond to the same signal. Send it back 100years to remain in place a few minutes and then return. Let us say you have it well calibrated to 10years but do not know exactly how well, which is why million year excursions would multiply your error and result in drift. So have the machine repeatedly send itself back 10 years 10 times. If the machine does not find itself after say 24 hours, return and assume the drift is under 10 years. Recalibrate to lengthen its concept of time and repeat. If the machine finds itself, note the delay before the first machine appears and calibrate accordingly. Repeat for 1000 years with 10 steps of one hundred. Do this then for 10000, 100000 and then one million. BONUS:The method works for future calibration too.

Answer 25

If you can move it somewhere else, try to go in a place where there is some ancient monument, for instance, the most ancient and known one is Barnenenz, in France. If your time machine can find it, it means it's 4850 BC or later, if it can't find it, it's earlier.

Do some attempts in a binary search fashion and you will have your time machine calibrated to that specific date (6868 years).

I know it's not a long time span but it's a start.

To get longer time calibration , you can place your time machine in the position where a dinosaur fossil has been discovered (possibly an easy one to find), this will give you a way longer time span. If the fossil is not there it means the dinosaur hasn't died yet 😉

Answer 26

Rather than using external cues, as others suggest, you could simply fit it with an odometer*.

I.e. you don't know exactly where* it's going to end up, but it could be capable of recording how far* it has travelled. You set the dial to 1 million years, it goes, and on it's return the odometer reads 2.5 million years (there and back again), you know it was out by 0.25 million (plus or minus 24 hours if the odometer records time elapsed in the old fashioned way). A few taps with the spanner later and you're in business.

This was pretty much the solution that H.G Wells used in The Time Machine, he controlled the machine with a lever, relying only on the 'odometer' to tell him how far he had travelled.

*temporally speaking, not geographically

Answer 27

You take a geological sample and do an analysis of uranium isotopes in the sample, using alpha b, mass spectrometry or maybe some other technique to obtain the isotope ratio between U-238 and U-235. As the halflife of U-235 is shorter than that of U-238, this ratio is slowly increasing. - this will easily tell you how many million years you are back, maybe within 100000 as well (I have not done the math ;-) ) - if you want to compare times a few millions ore more of years back, you may even be able to use ratios involving 236-U or 233-U - with much shorter halflifes, they will give a much better resolution 10000 / 1000 years, but they can not help you to obtain an absolute point in time.

Answer 28

If you consider that gravity and the magnetic field are not constant through time, the machine can just accurately measure Earth's values and do the proper calculations to exactly determine the time it ended up in. There is no need for other external measuring tools.

Answer 29

Fundemental problem

And assume that all of my testing is being performed in a desert area of the American southwest.

There is a fundamental problem that most time travel designers oversee. They all seem to assume that planet earth's position is at the center of the universe.

You are traveling through space-time. Let's say you assume to keep your position in space coordinates and you travel back or forward for only a few minutes, you are going to miss earth big time.

Wikipedia:

Earth's orbital speed averages about 30 km/s (108,000 km/h; 67,000 mph), which is fast enough to cover the planet's diameter in 7 minutes and the distance to the Moon in 4 hours.

Taking into account the speed of the solar system itself (Again wikipedia):

The Solar System is traveling at an average speed of 828,000 km/h (230 km/s) or 514,000 mph (143 mi/s) within its trajectory around the galactic center.

Now, this could go on taking all movements and expansion of the universe in account, but I will stop here. Point is: you will need a pretty good coordinate calculation system before you send your time machine. This can be done, if you know exactly what time you are sending it to. But there appears the paradox: you want to send it back in time to calibrate it's "Time dail". It is therefore impossible to put it anywhere on the planet.

Problem becomes the solution

So, don't try to land in earth. Make sure that the design is capable of withstanding outer space. Send it to exactly the same space coordinates, which will put it somewhere in deep space, with the best view of the galaxy you can ever get.

Now, it can do wavelength identification of our own solar system and other known stars or even remote galaxies. And based on the absolute or relative differences, you can accurately calculate the time.

Risk calculation

Before sending back your machine, try to estimate its inaccuracy. Based on the galactic model, try to keep the probability small it ends up in another star that since traveled to the same coordinates. Changes are really small, but the calculation is not difficult vs loosing your precious machine.

More runs

Do more runs, to different time frames. Try do to confirmation measurements that the galactic model is indeed correct. Each run your "Time dail" should become more accurate. Once you have enough confidence, start tuning the coordination system.

Do the same range of trips, but closer to earth. Assess earth's position and own rotational speed. Try to create your own calibrated movement model, not based on calculation of modern science, but what is now actually measurable.

In the final test you send it in the same place in the south-western American desert. If the machine returns in one piece (eg you did not put it in earth's crust or let it plunge down from a reasonable height) your machine is ready for human travel!

Answer 30

Turn off the space unit in your time-space machine (in order to avoid catastrophic failure, I'm assuming it is necessarily a time-space machine, but we don't need that part here).

Travel back some "moderate" unknown amount of time (couple of hundred years, or some thousand, whichever), that will place the vessel somewhere in space. Drop off a radio beacon, go back.

Receive the signal. If you don't receive anything, something went wrong (beacon destroyed, out of power, waaaaaayyyy too far into past?) - try again.

Either you need a timestamp encoded in the signal (then you need to correct for the speed of light, but luckily this is pretty constant, and well-known), or you can use Doppler's Effect (then you'd need 3 beacons minimum), whichever. You could use both, and cross-validate.

Calculate the distance between "then" and "now". Earth's speed while tumbling around Sun through Milky Way is known. There you go.