Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

SR and one-way speed of light tests

  1. Aug 7, 2003 #1
    One-way speed of light test.
    I would be grateful for any views you might have on what is a simple test to measure lights speed one-way on the surface of the Earth. The purpose of this test is to prove that the speed of light is affected by the ether wind and as a consequence Einstein's SR is wrong. This test is unique in that it eliminates clock synchronization errors and I believe it has never been done!

    Let u be the speed at which the Earth moves through the ether, its orbit speed of 30000m/s.

    Two clocks A and B are placed on the Earth's surface 10km apart. When the clocks line up in the direction of the Earth's orbit, a laser fires from A to B. Clock A records the start time and B records the stop time. The time difference between the two clocks is

    td(1) = sync error + 10000/(c-u) seconds

    Assuming a small synchronization error exists between them.

    12 hours later the clocks line up again parallel to the Earth's orbit, but this time in the opposite direction. The laser again fires from A to B and the time difference recorded is td(2), where

    td(2) = sync error + 10000/(c+u) seconds

    The two readings are then compared.

    Now if SR is true the difference between the times should be zero, as the speed of light is unaffected by the ether - if it exists.

    But I believe that the speed of the Earth through the ether will result in light travelling at different speeds and td(1)-td(2) = 6.7nS

    This is a simple test and some may consider it irrelevant, as there are many two-way tests that supports SR. But think very carefully, if two-way tests cannot detect the ether then a one-way test must be the answer and to my knowledge no such one-way speed of light test has been done!
  2. jcsd
  3. Aug 8, 2003 #2


    User Avatar

    Staff: Mentor

    Must? You are assuming the result before doing the test. Perhaps tests we have done don't show the result you are looking for because your theory is wrong?

    A one-way test of the speed of light is done every time someone flips on a gps reciever.
  4. Aug 8, 2003 #3
    How can that be Russ? Whenever anyone turns on a gps, they never, nor could they, see if there is a time difference between outgoing and income signals from a satelite. It should be noted that the solar system, and galaxy are both moving through space too, so you'd have to take their movement into account while doing the experiment. However, I don't see how a one way test would make a difference. If an ether exists, the experiments show that light doesn't notice.
  5. Aug 8, 2003 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I work with an engineer who is in the process of setting up a femtosec pulse length laser, now all you have to do is provide me with a reason why the speed of light would aveage c but not be c in both directions on a trip to a mirror and back.

    Please provide me with the argument. I could possibly convince them to do a one way speed of light experiment if there is a concieveable reason we should take valuable equipment and personell time to do such an experiment.

    By the way the way I read your reasoning, Michelson & Morely did your experiment over 100yrs ago. In spite of having suffciet resolultion to measure the variations due to what you are citing, it was not found.

    Again,What is the physics behind the variations you speak of? We need a theory to test, what is your prediction of the magnetude of the variation and in what direction should we look?
  6. Aug 8, 2003 #5
    Michelson-Morley didn't exactly do the tests to prove this wrong. It was a two-way test, as each leg of the interferometer had mirrors on the end, reflecting the light back from each leg to a common point.

    But, they rotated the whole apparatus, at least 90 degrees, and so there can't be a directional difference in the speed of light.

    But .. they only rotated it in a plane, not on every axis of a sphere. I'm not suggesting there could be differences on different axis .. only enough wiggle room left to say it could be. Have to have a whole lot of conspiracies lined up, more than SR gives us, to show positive results here. i.e., if the direction effected the speed, it would also have to effect the wavelength.
  7. Aug 9, 2003 #6
    If I had access to two atomic clocks, a lasers and light gate triggers I would seriously do this experiment. But I don't work in physics. I have spent alot of my time and money developing a new theory (wisp unification theory - http://www.kevin.harkess.btinternet.co.uk ) that shows this test will give a positive difference result of 6.7nS over 10km.
    If anyone wants to use my theory as a reason to justify doing this, they're welcome. The theory is free to download.
    Chapter 7 - section 7.7 shows why the two-way MM experiment failed. It's due to a thing I call jiggle.
    The GPS data travelling back and forth has too many variables to determine any differences in journey times. I think that signals are delayed by approx 20mS each way and internet traffic gets another 80mS delay. I don't believe anyone has conducted a true one-way GPS test. GPS also use Einstein's clock synchronization method, which is biased in favour of SR.
    This one-way test has the advantage of cancelling any clock sync delays, which is important.
  8. Aug 9, 2003 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Since it appears that the purpose here is to promote your own theory, Off to TD this thread goes.
  9. Aug 9, 2003 #8


    User Avatar

    Staff: Mentor

    GPS is one way - they gps recievers are recievers only. They work by recieving signals from 4 or more satellites and comparing the time stamps on the signals to calculate the distance from each. If the speed of light varied, it wouldn't work.

    In addition, since the speed of light IS constant and this affects the flow of TIME, the gps system takes this effect into account.

    I don't understand. You acknowledge that GPS accounts for/uses relativity, but you're saying that since it uses relativity it can't prove relativity? Thats how science works!!

    Or maybe you just don't understand how it works. Too many variables? Listen, GPS WORKS. If there were too many variables to do the needed calculations to make the necessary relativistic corrections, GPS WOULD NOT WORK.

    And again: GPS signals are ONE WAY ONLY. The recievers do not transmit any information back to the satellites.

    Civilan GPS is accurate to about 10m. Light travels 10m in 1/30,000,000 second. The transit time from a satellite 300km away is 1/1,000sec. So if the speed of light were to vary by more than .003%, then that would need to be taken into account. It doesn't and it isn't.
    Last edited: Aug 9, 2003
  10. Aug 10, 2003 #9
    I find this intriguing. I read a tiny bit of this so called wisp theory. So, relativity was wrong because it violates commonsense? I see. Paradoxes exist? Whoa, I must have missed those. But what's this...you say certain things, like the speed of lights (and you add transverse forces of some sort) are unaffected by motion and are absolute regardless of who is observing. But doesn't this mean then that what you are trying to test is pointless? Why, yes, yes it does beacuse light would be uneffected by this so called ether.

    By the way, someone said they tested it only on a plane. That's all you needed. The plane was the general one that earth's orbit traverses, and hence the one that would create the most effect if it existed.
  11. Aug 10, 2003 #10
    The reference to my theory was to supply "intergral" with some support material to give a convincing argument to do the test; it's not there to distract this thread.
    I agree that on basic calculations "russ_watters" figures show that if the speed of light is constant the results seem OK.
    All things being taken into account there seems to be no reason to doubt that the figures work out.
    But it can be shown that the GPS system will also work if the Earth's motion through the ether affects lights relative speed.
    GPS satellite orbits are determined by their orbital period (approx 11hrs 58mins), their height (approx 26600km - 4 Earth radii), their orbital speed (about 3.9km/s), and the orbital eccentricity (typically less than 1%). Data from T.V.Flanders "What the Global Positioning System Tells Us about Relativity".

    Lets assume that a GPS was setup using clocks synchronized to an absolute reference clock. And assume that the exact orbit positions of the satellites are known.
    If the speed of the Earth through the ether affected lights relative speed, we would still be able to have a working system.

    Now if someone applied SR with Einstein's clock synchronization, we would see immediately that things were not right because the clocks are out.

    But if we did not record the satellites exact positions, we could say that if we assume the satellite orbits are shifted forwards by 2.6km in the direction of the Earth's orbital motion, then SR works out OK.

    The difference between the two systems is only a question of orbital shift. It is not related to orbital period, as a shift will not affect this value.
    I doubt that the exact positions of the satellites have been verified by means of an optical check. Two Earth based telescopes would need to make simultaneous measurements to within fractions of an arc second to pinpoint the satellites exact position.
    This data would then have to be compared with data predicted by GPS.

    The point I am making is that the GPS system doesn't prove that the speed of light is constant. The GPS is able to track positions on the Earth to about 30cm, but there are many ways that this can be achieved even if the true orbital positions were all shifted by 2.6km.

    What are the advantages of doing this Earth based one-way test.
    1. It has never been done before. Whatever the result, it will - if done properly - be recorded in history as the first test of its kind.
    2. It may change science if the difference result is positive.
    3. It is a simple test that will either strengthen SR are destroy it.

    I am still learning about GPS. The more I look into it the more I'm convinced that GPS is not a test for the speed of light.
  12. Aug 10, 2003 #11


    User Avatar

    Staff: Mentor

    In order to make the system give positions accurate to 10m, the position of each satellite must be known to better than 10m precision.
    Clocks on GPS satellites DO vary according to SR.

    This is as far as I'm going to go, wisp. You have a TON to learn about Relativity and its implications. I really recommend you read a book on it.
  13. Aug 11, 2003 #12
    Thanks for the reply. I have read may books on special relativity and don't consider it a difficult subject. However, I'm not going to persue this GPS discussion futher. I will wait until someone does this test and see what the results are.

  14. Aug 11, 2003 #13
    Your experiment is essentially not going to reveal anything new. The MM experiment would have detected ANY drift effect due to the ether.


    One should read that and study the diagram even to see why your one way test as you call it was more or less included in the MM experiment (in other words, the light did bounce back from the direction it came. IF there was any ether difference, a distinct interference pattern would have occured).
  15. Aug 11, 2003 #14
    Hmm, ANY drift? What if the EMF actually (what word do I use??) bent/modified/effected/drifted the ether? Or, the ether existed but was effected in such a way that makes it look like C is constant?

    Hard to believe? Yes, but look at something else. Now adays we all talk about the electromagnetic field. Just how is that different than an ether? I don't see it as any different. I see it as just calling an ether by another name, maybe with a few different properties.

    The MM experiment didn't prove there wasn't an ether; only that an ether wasn't needed to formulate a theory behind observations.
  16. Aug 11, 2003 #15
    not only have they checked them optically, they actually built several observatories across the globe to monitor them 24/7/365 for exactly that only!
    read this for short: http://www.trimble.com/gps/how.html
  17. Aug 11, 2003 #16


    User Avatar

    Staff: Mentor

    Yeah, you have no idea what an electromagnetic field is. Then again, a rose by another name and with different properties would be - a petunia.

  18. Aug 11, 2003 #17


    User Avatar

    You mean T Von Flandern? I thought that name was familiar.

    You're in luck, there is a wealth of information documenting Flandern's errors regarding relativity. Here's an article about his claims about GPS. And how they are complete rubbish.

  19. Aug 11, 2003 #18
    Seems you woke up a bit cranky from your nap.
  20. Aug 11, 2003 #19
    Cranky, but accurate. The electromagnetic field is by NO means anything like an ether.
  21. Aug 11, 2003 #20


    User Avatar

    Staff: Mentor

    Great link - much more in depth about relativistic effects on GPS than I had ever seen.
    Rough day + instinctive low tolerance for crap.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?