Page 4 of 4 FirstFirst ... 234
Results 91 to 119 of 119

Thread: Gedanken Experiment

  1. #91
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by 01101001 View Post
    Perhaps. OK. But, may I then say that this doesn't sound in any way like an actual thought experiment, which poses a "what-if" (a scenario, the experimental setup) and logically (the thought, gedanken) proceeds to a "therefore" (the results, an understanding). Hence, my confusion with the use of the term, the title.
    Ok, what if light really does travel at c relative only to its source, and what if it really is reset as it passes through any medium? What if this test set up is able to see a difference in the velocity of light between approaching and receding stars? Wouldn’t that be an interesting result?

    Quote Originally Posted by 01101001 View Post
    But, carry on. Have fun with whatever it is.
    Thanks, I will.

    Quote Originally Posted by 01101001 View Post
    Thanks for the explanation.
    You are welcome.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  2. #92
    Join Date
    May 2006
    Posts
    5,512
    Quote Originally Posted by MentalAvenger View Post
    No, actually I want to measure the velocity of light precisely. Finding our IF there is a difference would be of little help without knowing how much that difference is.
    You wouldn't need the difference, just the ratio of times. Whatever the actual speed of light is when stationary wouldn't matter, you could just call it c, and if it takes twice as long to receive a pulse when in motion, then light travels at .5 c for that particular relative velocity.

    Anyway, measuring light from stars has too many variables. Do we know how fast the star is moving toward us or away? Do we know the aberration from cross-wise movement? How accurately do we know the distance to that star? There are too many uncontrollables.

    So whether you want to measure the actual speed of light or not, I think a combination of yours with snowflakeuniverse's idea is best. I guess I missed reading the whole second page before posting. I must have gotten to the end of the first and thought that was it.

    So we have two rockets in space and two identically long cables, a clock on each rocket, with a sensor on one and transmitter on the other. The length of the cords are measured by running them through a machine that will measure them precisely a few times until we are satisfied as to the margin of error for their length. One cord is attached between the rockets in space and they then travel away from each other until it is straightened. That same cord transmits the time from the clock on the transmitting rocket to the clock on the other. Running the two cords to each other wouldn't work, though, because each will keep reading a lesser time than the other due to the transmission time through the cords and adjust themselves accordingly so that both clocks just continually wind down. But we don't even have to know what that transmission time is in order to account for it. Just run the second cord from the receiving clock back to the same receiving clock and adjust the receiving clock until the time read coming from both cords over the same distance is the same. Once we are satisfied with that, we can send a continual beam of light between the two to make sure the readings for the distance do not change, whatever those readings are. This has nothing to do with the experiment, however; just a precautionary measure. We can now begin the experiment by transmitting pulses to the receiving rocket and marking the time on each clock when transmitted and received. When we are done, we bring the two clocks together and find the difference between their times. That will be the time of transmission through the cables. Finally, adding that to the time of transmission of the pulses, we find the time the light took to cover a distance equal to the length of the cables, and we know the speed of light when stationary.

    We can then go back and experiment with relative velocities and accelerations in a similar manner, but by letting the cord between them run freely out of one of the rockets, measuring the length as it goes. The free cable would be the same length as the entire length of the cable between them plus any additional that has not been run out yet. In other words, the entire length of the cable between the clocks. The experiment might be even more precise, though, if it were performed with extendable solid tracks instead.

    Well, that was fun.

    Really, though, the beam of light that makes sure the distance remains the same is enough to know they are stationary relative to each other as long as the reading does not change, and we can add the time of transmission according to the difference in whatever time we are reading after the clocks have already been synchronized, so the cords would probably just be more of a burden than anything. Of course, again, we couldn't measure the distance or the speed of light precisely this way, just the ratio to c locally, but c has already been measured very precisely locally, so I personally would just like to know the ratio for the relative velocities and accelerations of such an experiment.
    Last edited by grav; 2007-Jan-26 at 06:59 AM.

  3. #93
    Join Date
    Oct 2001
    Posts
    6,275
    I see a possible fundamental problem with MentalAvenger's experiment, or any such experiment. The problem is that a null result is inconclusive.

    If the experiment did pick up a consistent difference when measuring c from two different sources (approaching and receding) there would clearly be new Physics involved. But if there were no such difference, it would not disconfirm the hypothesis.

    Why? Because the basic notion is that light "changes speed" if it is reflected, refracted, or absorbed/reemitted, all of which are processes that occur when light interacts with matter. The experiment assumes that the light from those distant stars has no interaction with matter anywhere between the source and the experiment's sensors.

    But this is an unfounded assumption. The interstellar medium is not a perfect vacuum. The solar system is even less so. It may very well be that all starlight that reaches the Moon would have interaction with some nearby matter that could "reset" its speed to a value that reflects local velocities.

    Even if you could guarantee that pristine light made it as far as the apparatus, there's still the problem of interaction with the edge of the shutter as it opens. It's the leading edge of the light beam you're measuring at the sensors, as you go from "no light" to "light". But those leading-edge photons are the very ones that may be interacting with the material of the shutter as they "sneak past the edge" -- and we know that photons are affected by edges of material objects (think single-slit experiments and the resulting interference patterns).

    Thus the leading edge of your signal, which you're depending on to be pristine, is necessarily affected by its close encounter with a nearby object: the shutter. Once it's fully open, you do get unadulterated starlight; but that's too late. Those photons could have different speeds and you'd never know it.

    There's also the problem of stray starlight entering the detectors, and whether the techniques you'd use to prevent that would affect the incoming "desired" photons as well. But I've rambled on long enough.
    Last edited by Donnie B.; 2007-Jan-26 at 01:50 PM. Reason: To clarify which experiment I was discussing.

  4. #94
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by grav View Post
    You wouldn't need the difference, just the ratio of times. Whatever the actual speed of light is when stationary wouldn't matter, you could just call it c, and if it takes twice as long to receive a pulse when in motion, then light travels at .5 c for that particular relative velocity.
    Not so. The actual differences would be far less than .5c, probably more like .01c. There would be no way to know what the actual differential in object velocity is. You cannot use what you are attempting to measure as a basis for what you are measuring.

    Quote Originally Posted by grav View Post
    Anyway, measuring light from stars has too many variables. Do we know how fast the star is moving toward us or away? Do we know the aberration from cross-wise movement? How accurately do we know the distance to that star? There are too many uncontrollables.
    The object would be to take readings on many different stars, and compare the results with conventional determinations of their radial velocity. Of course, if the experiment finds that the speed of light is the same for all objects, then we have another problem. Either TOR is correct, or the light coming from distant stars encounter enough interstellar matter to reset the velocity to local normal.

    Quote Originally Posted by grav View Post
    So whether you want to measure the actual speed of light or not, I think a combination of yours with snowflakeuniverse's idea is best.
    Thank you for your suggestion.

    Quote Originally Posted by grav View Post
    So we have two rockets in space and two identically long cables, a clock on each rocket, with a sensor on one and transmitter on the other. The length of the cords are measured by running them through a machine that will measure them precisely a few times until we are satisfied as to the margin of error for their length.
    There is still too much room for error in that method. The physical length of the cables is only one factor. The processing time through the whole system needs to be zeroed. In any case, your suggestion enters too many more circuits and devices into the loop. Each additional device decreases the overall accuracy. That is why I use such a simple setup.

    Quote Originally Posted by grav View Post
    We can then go back and experiment with relative velocities and accelerations in a similar manner, but by letting the cord between them run freely out of one of the rockets, measuring the length as it goes.
    The amount of error here goes off the scale.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  5. #95
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by Donnie B. View Post
    I see a possible fundamental problem with MentalAvenger's experiment, or any such experiment. The problem is that a null result is inconclusive.

    If the experiment did pick up a consistent difference when measuring c from two different sources (approaching and receding) there would clearly be new Physics involved. But if there were no such difference, it would not disconfirm the hypothesis.
    Absolutely correct. I could not come up with an experiment where a null result was not possible.

    Quote Originally Posted by Donnie B. View Post
    Why? Because the basic notion is that light "changes speed" if it is reflected, refracted, or absorbed/reemitted, all of which are processes that occur when light interacts with matter. The experiment assumes that the light from those distant stars has no interaction with matter anywhere between the source and the experiment's sensors.
    An astute observation, and quite correct. It would require an independent experiment to test for that. IMO, we need to measure particle density and composition, not just past the Heliopause, but also out just beyond the Bow Shock, out where the Sun’s magnetosphere might be creating a region of higher density of particles.

    Quote Originally Posted by Donnie B. View Post
    Even if you could guarantee that pristine light made it as far as the apparatus, there's still the problem of interaction with the edge of the shutter as it opens. It's the leading edge of the light beam you're measuring at the sensors, as you go from "no light" to "light". But those leading-edge photons are the very ones that may be interacting with the material of the shutter as they "sneak past the edge" -- and we know that photons are affected by edges of material objects (think single-slit experiments and the resulting interference patterns).
    I understand the point, but the resetting effect does not appear to be caused by passing near something. All the information I have seen indicates it only happens when light is either transmitted through a medium or reflected off a medium. In both cases, it appears that the light is absorbed and reemitted.

    Quote Originally Posted by Donnie B. View Post
    Thus the leading edge of your signal, which you're depending on to be pristine, is necessarily affected by its close encounter with a nearby object: the shutter. Once it's fully open, you do get unadulterated starlight; but that's too late. Those photons could have different speeds and you'd never know it.
    I don’t think so, but it appears that it is worth designing another experiment to test for that.

    Quote Originally Posted by Donnie B. View Post
    There's also the problem of stray starlight entering the detectors, and whether the techniques you'd use to prevent that would affect the incoming "desired" photons as well. But I've rambled on long enough.
    The threshold of the detectors can be selected to respond only to the light from the target star. Stars can be selected which do not have any nearby (laterally) stars of sufficient brightness to interfere. The particle density on the Moon does not appear to be great enough to scatter light sufficiently to interfere, and the small opening in the shutter, and the close alignment of the sensors, doesn’t allow much more than the target starlight through. Good point, though.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  6. #96
    Join Date
    May 2006
    Posts
    5,512
    Quote Originally Posted by MentalAvenger
    Quote:
    Originally Posted by grav
    You wouldn't need the difference, just the ratio of times. Whatever the actual speed of light is when stationary wouldn't matter, you could just call it c, and if it takes twice as long to receive a pulse when in motion, then light travels at .5 c for that particular relative velocity.

    Not so. The actual differences would be far less than .5c, probably more like .01c. There would be no way to know what the actual differential in object velocity is. You cannot use what you are attempting to measure as a basis for what you are measuring.
    Sure it would. The ratio for the speed one would measure would just be some distance divided by the time of transmission. So if we measure the time when stationary as t, then if it takes twice as long with some relative velocity, then the local speed of light for that particular velocity would just be v=d/(2t). So even if we don't know the original speed of light or the distance travelled, we would still know the speed of light is measured at half of that of the stationary value. So we would then know what relative velocity causes the speed of light to be measured locally at half of the value for when stationary. By locally, I just mean what we would measure using our own rulers and clocks.

    Now, if the relative velocity or acceleration between the receiver and transmitter causes an additional real time dilation, then yes, the measure must be re-adjusted. We would know this, though, when we brought the two clocks back together and compare times to see if they differ. That would cause some complications if they do. But it would also tell us about that as well. The relative velocity or acceleration used here is also not the same as the relevant velocity or acceleration for what one would actually observe for the other, but just that we would expect in flat space-time, although we could then learn what the relevant values would be as well.

  7. #97
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    I can’t figure out what test setup you would be using for your claims to be valid. And I don’t see how it applies to what I was attempting to determine with my experiment.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  8. #98
    Join Date
    Mar 2004
    Posts
    13,440

    Errors

    As I'm sure all readers of this thread know, the Moon has been used, by astronomers, for quite a long time, to estimate the diameter of stars which it occults, and to estimate limb darkening (and to detect unresolved binary companions).

    The limits of this technique, other than that a star must be occulted by the Moon!, include collecting enough photons from the star and resolving their arrival time finely enough.

    These limits have obvious implications for the MA experiment; some examples:

    * geometry - if the star being observed is not a true point source, then the geometry of the system (shutter and two sensors) will affect the outcome (e.g. was what triggered the first detector the same as what triggered the second?)

    * sensitivity of detectors - the faster the sensors respond to the arrival/detection of photons (from the star), the more finely arrival time can be resolved. Further, good time resolution is required to ensure that the two detectors will be triggered by the same 'event'. Once you have selected sensors with quantum efficiencies close to 100%, the only way to improve time resolution is to collect more photons ... using mirrors or lenses (or both)

    * the sensors' photon energy response curve - if faint stars are used, unresolved binaries, of different colours, may create errors by having triggers that are not repeatable (e.g. due to geometry differences)

    Some additional sources of error:

    * cables - while it may be possible to say that the two are 'the same length' (within a very small error) while they are in the lab, the deployment of the cables may cause their lengths to change, by amounts that would likely be difficult to estimate. Simply swapping the cables and repeating the measurement may help, but if the changes in length are due to factors specific to the environment of each cable - a sharper bend in one than the other say, or a difference in the temperature profile - this may be hard to estimate

    * triggering of the cable pulses - ideally, the delay between arrival of photons at the sensor and triggering of the pulse down the cable could be estimated quite accurately, through side-by-side testing in the lab (before deployment). However, just as the environments of the two cables will not be same, neither will the environments of the two sensor-cable heads. In situ calibration and swapping the units would reduce this error (or at least allow it to be estimated), but there will always be a residual, which may be difficult to estimate.

    So, MentalAvenger, why not try to make some estimates of the likely accuracy of your experiment, using realistic inputs?

  9. #99
    Join Date
    Mar 2004
    Posts
    13,440
    Quote Originally Posted by MentalAvenger View Post
    [snip]
    Quote Originally Posted by Nereid
    Using light to measure light, eh? How would avoid all systematic errors if you went down that path? How could you estimate the systematic errors involved if you used light to measure distance?
    If you care to enlighten me on what you think those systemic errors might be, I will address your concern.

    [snip]

    Remember, no lenses or mirrors in this setup.
    Perhaps I just don't understand ... how would you use light (or radar) to measure the distance between shutter and each of the sensors?

    Would you use lenses and/or mirrors?
    I think my test procedure, including the unique all-system alignment, eliminates virtually all potential sources of errors.
    As I just noted, there are quite a few sources of error ... the real question is: "how large, potentially, could each be?"

    Would you care to have a go at estimating the size of all sources of error identified so far in this thread?

  10. #100
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Nerid,
    Thank you for your careful analysis. You have indeed highlighted some points that need attention. I really appreciate the time you took, and I will try to address those issues.
    Quote Originally Posted by Nereid View Post
    These limits have obvious implications for the MA experiment; some examples:
    * geometry - if the star being observed is not a true point source, then the geometry of the system (shutter and two sensors) will affect the outcome (e.g. was what triggered the first detector the same as what triggered the second?)
    Choosing stars with the same apparent magnitude, far enough away to appear to the system to be point sources should solve this problem.

    Quote Originally Posted by Nereid View Post
    * sensitivity of detectors - the faster the sensors respond to the arrival/detection of photons (from the star), the more finely arrival time can be resolved. Further, good time resolution is required to ensure that the two detectors will be triggered by the same 'event'. Once you have selected sensors with quantum efficiencies close to 100%, the only way to improve time resolution is to collect more photons ... using mirrors or lenses (or both)
    Correct. I assume using state-of-the-art sensors whose characteristics are optimal for this test. Low threshold and high sensitivity would be required. Unlike the reference image I posted, the sensors would be aligned side by side, rather than one above the other. The shutter would move vertically. The final field test should confirm simultaneous triggering alignment.

    Quote Originally Posted by Nereid View Post
    * the sensors' photon energy response curve - if faint stars are used, unresolved binaries, of different colours, may create errors by having triggers that are not repeatable (e.g. due to geometry differences)
    Correct. Again, the selection of stars would be critical to the test.

    Quote Originally Posted by Nereid View Post
    * cables - while it may be possible to say that the two are 'the same length' (within a very small error) while they are in the lab, the deployment of the cables may cause their lengths to change, by amounts that would likely be difficult to estimate. Simply swapping the cables and repeating the measurement may help, but if the changes in length are due to factors specific to the environment of each cable - a sharper bend in one than the other say, or a difference in the temperature profile - this may be hard to estimate
    It may not have been obvious, but I believe my testing procedure eliminates all those errors. Of course the cables will be assembled in a clean environment, hopefully on the Moon, but on Earth if necessary. The assembly would include sealed connections at all points, and all units would be sealed. The assembly would include the Comparator, the cables, and the sensors, all connected together. Once assembled, the connections are never broken.

    The test is to bring the two sensors together, side by side, and to trigger them with a light that simulates the faint light from a star as near as possible. For purposes of the test, it would be logical to choose stars of the same apparent magnitude. When the simulated starlight achieves a zero error between the two sensors, then we can assume that any differences in circuitry, materials, or assembly all cancel out. This is one of the most important steps.

    Once at the site of the Experiment, the test will be run again. After the Comparator, sensors, and cables have normalized to Lunar nighttime temperature, the sensors will again be brought together and placed side by side. A simulated light will be shined on both at once, using the shutter to create the pulse. If there are any differences, either a trim component can be used to zero the system, or the difference can be noted and factored out later. The only trick will be supplying a point source of light. In this part of the test, focusing with lenses is not a problem.

    I don’t think there will be any problems due to the cables being looped around to be together if sharp bends are avoided. There should be no difference to the system once they are removed to their final locations.

    Quote Originally Posted by Nereid View Post
    * triggering of the cable pulses - ideally, the delay between arrival of photons at the sensor and triggering of the pulse down the cable could be estimated quite accurately, through side-by-side testing in the lab (before deployment). However, just as the environments of the two cables will not be same, neither will the environments of the two sensor-cable heads. In situ calibration and swapping the units would reduce this error (or at least allow it to be estimated), but there will always be a residual, which may be difficult to estimate.
    As explained above, that was the point of my in situ side-by-side final test.

    Quote Originally Posted by Nereid View Post
    So, MentalAvenger, why not try to make some estimates of the likely accuracy of your experiment, using realistic inputs?
    It has been too long since I actually designed or built any electronic circuits. That was back in the mid-70’s to mid’80’s, and I am not up to speed on the specifications of current components. Any estimate I could make would be a guess and therefore of no value.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  11. #101
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by Nereid View Post
    Perhaps I just don't understand ... how would you use light (or radar) to measure the distance between shutter and each of the sensors?
    AFAIK, radar does not suffer from the same “resetting” that light does. That is an interesting anomaly since they are merely different sections of the same electromagnetic spectrum. Using radar is one way to determine the exact distance between the two sensors, which is the only critical distance. However, since radar is a longer wavelength, resolution will be reduced.

    Using light for determining the distance between the two sensors should not violate the conditions of the Experiment as long as it is a one way trip from the source to both sensors. Since it is a local source, we should be able to use 299,792,458 m/s (or 1.802,265,898 MF/mf) as the standard.

    Quote Originally Posted by Nereid View Post
    Would you use lenses and/or mirrors? As I just noted, there are quite a few sources of error ... the real question is: "how large, potentially, could each be?"
    As long as the lens is only at the source, it should not violate the conditions of the Experiment.

    Quote Originally Posted by Nereid View Post
    Would you care to have a go at estimating the size of all sources of error identified so far in this thread?
    Again, I don’t think I can.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  12. #102
    Join Date
    Apr 2005
    Posts
    11,545
    Quote Originally Posted by MentalAvenger View Post
    AFAIK, radar does not suffer from the same “resetting” that light does.
    As far as we know, light doesn't either.

    The two experiments mentioned earlier show that.

  13. #103
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by hhEb09'1 View Post
    As far as we know, light doesn't either.

    The two experiments mentioned earlier show that.
    Eric Cornell, Carl Wieman, Deborah Jin, and Dana Anderson, amongst others, would probably disagree. It is their experiments with Bose-Einstein Condensate which show this effect. It explains why light “slows down” when passing through a medium. In fact, that is what started me thinking about this situation.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  14. #104
    Join Date
    Mar 2004
    Posts
    13,440

    simple question - shutter and sensor

    A (vertical) shutter opens, allowing light from a distant (point) source through.

    A sensor detects that light.

    If you plot the signal from the sensor as a function of time, what does it look like?

    The shutter closes, blocking light from a distant (point) source.

    If you plot the signal from the sensor as a function of time, what does it look like?

  15. #105
    Join Date
    Apr 2005
    Posts
    11,545
    Quote Originally Posted by MentalAvenger View Post
    It explains why light “slows down” when passing through a medium.
    Ah, sorry, I thought we were talking about the resetting that goes on from the emission of light from stars. I see what you mean.

    I meant that the two experiments show that the speed of the emitting body does not affect the speed of light over the intervening space.

  16. #106
    Join Date
    Oct 2001
    Posts
    6,275
    Why would the photons used by radar not be subject to the same "resetting" phenomenon that the photons of light are (possibly) subject to? (I don't necessarily think it would invalidate using radar to determine the distance between two mutually stationary sensors, but I'm curious why you think there's a difference.)

  17. #107
    Join Date
    Apr 2005
    Posts
    11,545
    Quote Originally Posted by Donnie B. View Post
    Why would the photons used by radar not be subject to the same "resetting" phenomenon that the photons of light are (possibly) subject to? (I don't necessarily think it would invalidate using radar to determine the distance between two mutually stationary sensors, but I'm curious why you think there's a difference.)
    I think it's related to the refractive index increasing with freqency--radar is just lower frequency.

  18. #108
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by Nereid View Post
    A (vertical) shutter opens, allowing light from a distant (point) source through.
    A sensor detects that light.
    If you plot the signal from the sensor as a function of time, what does it look like?
    If it really is a point source, the signal from the sensor would rise as a function of the response time of the sensor itself. The input would be off, then on. I am not familiar enough with the properties of starlight to make an informed evaluation of it, but I suspect that starlight is a little more complex than being a point source. Maybe someone here knows more about that.

    Quote Originally Posted by Nereid View Post
    The shutter closes, blocking light from a distant (point) source.
    If you plot the signal from the sensor as a function of time, what does it look like?
    Irrelevant, we only need to look at the leading edge of the light pulse.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  19. #109
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by hhEb09'1 View Post
    Ah, sorry, I thought we were talking about the resetting that goes on from the emission of light from stars. I see what you mean.
    I was, actually. Regardless of the velocity of the light on the way here, once it passes through a lens or is reflected off a mirror, the velocity is set to c relative to that lens or mirror.

    Quote Originally Posted by hhEb09'1 View Post
    I meant that the two experiments show that the speed of the emitting body does not affect the speed of light over the intervening space.
    Relative to what?
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  20. #110
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by Donnie B. View Post
    Why would the photons used by radar not be subject to the same "resetting" phenomenon that the photons of light are (possibly) subject to? (I don't necessarily think it would invalidate using radar to determine the distance between two mutually stationary sensors, but I'm curious why you think there's a difference.)
    That brings up a whole set of different issues. Even if light is comprised of photons, I was not aware that radar was considered to be photons also. I’d rather not get into the duality issue either. AFAIK, only visible light has been shown to have that effect. I have no idea if radar would do the same, except that radar does not appear to propagate through substances the way light does. In any case, the longer wavelength means lower resolution.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  21. #111
    Join Date
    Mar 2004
    Posts
    13,440
    Quote Originally Posted by MentalAvenger View Post
    Quote Originally Posted by Nereid
    A (vertical) shutter opens, allowing light from a distant (point) source through.
    A sensor detects that light.
    If you plot the signal from the sensor as a function of time, what does it look like?
    If it really is a point source, the signal from the sensor would rise as a function of the response time of the sensor itself. The input would be off, then on. I am not familiar enough with the properties of starlight to make an informed evaluation of it, but I suspect that starlight is a little more complex than being a point source. Maybe someone here knows more about that.
    I did mention, earlier (I think), that your setup looks much like one technique for measuring the diameter of stars (and limb darkening, and detection of faint/close companions, and ...) - lunar occultation.

    Perhaps you could read up on it? And when you're done, tell us how your setup would address diffraction effects?
    The shutter closes, blocking light from a distant (point) source.
    If you plot the signal from the sensor as a function of time, what does it look like?
    Irrelevant, we only need to look at the leading edge of the light pulse.
    You might want to reconsider this, esp as a possible means of addressing some systematics (via symmetry - the 'signal' from the shutter closing is - or should be - symmetrical with the signal from the shutter opening).

  22. #112
    Join Date
    Apr 2005
    Posts
    11,545
    Quote Originally Posted by MentalAvenger View Post
    Relative to what?
    Most importantly, I guess as far as the experiment, relative to the earth

  23. #113
    Join Date
    Oct 2001
    Posts
    6,275
    Quote Originally Posted by MentalAvenger View Post
    That brings up a whole set of different issues. Even if light is comprised of photons, I was not aware that radar was considered to be photons also. I’d rather not get into the duality issue either. AFAIK, only visible light has been shown to have that effect. I have no idea if radar would do the same, except that radar does not appear to propagate through substances the way light does. In any case, the longer wavelength means lower resolution.
    I don't see how you can discuss an experiment that depends on electromagnetic radiation without "getting into the duality issue".

    Are you saying you don't accept the dual wave/particle nature of light? Are you suggesting that either light or radar is something other than e/m radiation? Because if you are, we have much bigger issues to discuss than your doubts about the constant speeed of light.

  24. #114
    Join Date
    Mar 2004
    Posts
    13,440

    Dispersion

    If the speed of light is frequency dependent (e.g. gamma travels slower than radio, or vice versa, or some more complicated relationship), then we should see the footprints all over the sky.

    For example:

    * a distant boom (nova, supernova, whatever) will be seen arriving at different times (e.g. gamma first, radio next century)

    * eclipsing binaries will have 'light curves' at different frequencies that are out of synch with one another

    * ditto pulsars, Cepheids, and other variables.

    As we have sampled an enormous range of environments - plasma density, temperature, ionisation; dust density; column length; etc - with sight lines to astronomical objects both near and far, and (with some notable exceptions) have seen no evidence whatsoever of any of these kinds of effects, I think it's a pretty safe conclusion that the speed of light is frequency independent.

    The notable exceptions (e.g. interstellar scintillation) are well understood, and have (apparently) nothing to do with an in vacuo frequency-dependent speed of light ...

  25. #115
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by Donnie B. View Post
    I don't see how you can discuss an experiment that depends on electromagnetic radiation without "getting into the duality issue".
    Well, this discussion has wandered off the original intent. Although I did say we might as well measure the speed of light as accurately as possible as long as we have the test setup, the original intent was only to see if there is a difference in light velocity between radially receding stars and radially approaching stars, and to see if that velocity coincided with their determined radial velocity.

    Quote Originally Posted by Donnie B. View Post
    Are you saying you don't accept the dual wave/particle nature of light? Are you suggesting that either light or radar is something other than e/m radiation? Because if you are, we have much bigger issues to discuss than your doubts about the constant speeed of light.
    I would call it apparent duality. But that is an issue for another discussion. Something like that could easily hijack the thread.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  26. #116
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by Nereid View Post
    If the speed of light is frequency dependent (e.g. gamma travels slower than radio, or vice versa, or some more complicated relationship), then we should see the footprints all over the sky.
    Since you didn’t reference your comments, I don’t know what triggered your statements. I am not aware of making any comments that would indicate that the velocity of light (in a vacuum) is frequency dependent.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  27. #117
    Join Date
    Mar 2004
    Posts
    13,440
    Quote Originally Posted by MentalAvenger View Post
    Since you didn’t reference your comments, I don’t know what triggered your statements. I am not aware of making any comments that would indicate that the velocity of light (in a vacuum) is frequency dependent.
    The history (in brief):
    - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    Nereid: Perhaps I just don't understand ... how would you use light (or radar) to measure the distance between shutter and each of the sensors?

    Would you use lenses and/or mirrors?

    MentalAvenger: AFAIK, radar does not suffer from the same “resetting” that light does. That is an interesting anomaly since they are merely different sections of the same electromagnetic spectrum.

    hhEb09'1: As far as we know, light doesn't either.

    The two experiments mentioned earlier show that.

    MentalAvenger: Eric Cornell, Carl Wieman, Deborah Jin, and Dana Anderson, amongst others, would probably disagree. It is their experiments with Bose-Einstein Condensate which show this effect. It explains why light “slows down” when passing through a medium. In fact, that is what started me thinking about this situation.

    Donnie B.: Why would the photons used by radar not be subject to the same "resetting" phenomenon that the photons of light are (possibly) subject to? (I don't necessarily think it would invalidate using radar to determine the distance between two mutually stationary sensors, but I'm curious why you think there's a difference.)

    hhEb09'1: I think it's related to the refractive index increasing with freqency--radar is just lower frequency.
    - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

    Or you can ask: what effect would 'resetting' have, on 'light', other than on its speed? If 'resetting' didn't affect its speed, why do the MA experiment at all? If radar is not 'reset', but light is, and if 'resetting' affects speed, how could we describe this, treating it as a special case of a more general phenomenon? Why, a frequency-dependent speed of light (in vacuo)!

    A similar conclusion can be reached by following the quantum road ('duality') - that the ISM (and IPM, and IGM) absorbs 'light' is well-known. That this absorption is frequency-dependent is also well-known (we can 'see' stars orbitting SgrA*, in the IR, but not in the optical or UV, for example). But this is also true at much finer frequency scales -the ISM is much more opaque to Halpha light (for example) than it is to light only 2 or 5 nm different. When light passes through the ISM, does it get 'reset' when it is absorbed (and re-emitted) by H atoms (as line radiation), but not otherwise?

  28. #118
    Join Date
    Apr 2004
    Location
    Wyoming
    Posts
    1,412
    Quote Originally Posted by Nereid View Post
    Or you can ask: what effect would 'resetting' have, on 'light', other than on its speed? If 'resetting' didn't affect its speed, why do the MA experiment at all? If radar is not 'reset', but light is, and if 'resetting' affects speed, how could we describe this, treating it as a special case of a more general phenomenon? Why, a frequency-dependent speed of light (in vacuo)!
    Ah, I understand. Thank you for the explanation. You are treating radar as another form of light. I was treating light as “light” in the classical sense, the visible or near visible portion of the EM spectrum. I wasn’t planning on sensors which would detect much outside the 300-800 nm range. The intent of the experiment was to confirm (or refute) the TOR claim that light travels a c relative to the observer, rather than relative to the emitting source, using a combination of conditions that has not been tried before.

    Quote Originally Posted by Nereid View Post
    A similar conclusion can be reached by following the quantum road ('duality') - that the ISM (and IPM, and IGM) absorbs 'light' is well-known. That this absorption is frequency-dependent is also well-known (we can 'see' stars orbitting SgrA*, in the IR, but not in the optical or UV, for example). But this is also true at much finer frequency scales -the ISM is much more opaque to Halpha light (for example) than it is to light only 2 or 5 nm different. When light passes through the ISM, does it get 'reset' when it is absorbed (and re-emitted) by H atoms (as line radiation), but not otherwise?
    If the ISM is so populated with atoms that none of the light can reach us from a distant (200ly+) without passing through at least one molecule locally, then the experiment will not likely produce any useful information. I’ll have to think about this some more to see if there is a way to make this experiment not subject to some of these problems.

    I’d like to take this opportunity to thank everyone for participating in this exercise and staying within the stated conditions. After a somewhat rocky start, this discussion has progressed quite well. This is the first forum on the internet where I have tried this experiment where it did not immediately get hijacked by people merely parroting TOR and refusing to try to discuss the experiment itself. I appreciate the courtesy and the participation. For the first time I am learning something.
    Speed of light = 1.802,265,898 MegaFurlongs / MicroFortnight

  29. #119
    Join Date
    Mar 2004
    Posts
    13,440
    Quote Originally Posted by MentalAvenger View Post
    Ah, I understand. Thank you for the explanation. You are treating radar as another form of light. I was treating light as “light” in the classical sense, the visible or near visible portion of the EM spectrum. I wasn’t planning on sensors which would detect much outside the 300-800 nm range. The intent of the experiment was to confirm (or refute) the TOR claim that light travels a c relative to the observer, rather than relative to the emitting source, using a combination of conditions that has not been tried before.
    I'm not sure why you would choose to be so restrictive ... that's not even one decade, and we have detected EM from ~10 TeV gammas to close to the plasma frequency cutoff in the (LF) radio ... that's a lot of decades!

    Is there anything particularly special about 300-800 nm (other than it's just a bit broader than what we Homo sapiens can 'see', with our eyes)?
    If the ISM is so populated with atoms that none of the light can reach us from a distant (200ly+) without passing through at least one molecule locally, then the experiment will not likely produce any useful information. I’ll have to think about this some more to see if there is a way to make this experiment not subject to some of these problems.

    I’d like to take this opportunity to thank everyone for participating in this exercise and staying within the stated conditions. After a somewhat rocky start, this discussion has progressed quite well. This is the first forum on the internet where I have tried this experiment where it did not immediately get hijacked by people merely parroting TOR and refusing to try to discuss the experiment itself. I appreciate the courtesy and the participation. For the first time I am learning something.
    I'll leave it to you to decide just what, in your idea, "passing through at least one molecule locally" means* ... other than to point out that the EUV Explorer didn't 'see' very many stars at all (all that Lyman continuum absorption and (re-)emission ...).

    *Why does it have to be molecules? Why not atoms? ions? electrons? neutrinos?? dark matter??

Similar Threads

  1. Try a Unique Experiment...
    By Jimbee68 in forum Off-Topic Babbling
    Replies: 33
    Last Post: 2010-Aug-11, 04:25 AM
  2. LIFE experiment
    By Argos in forum Life in Space
    Replies: 23
    Last Post: 2009-Jan-23, 10:05 PM
  3. Experiment
    By afterburner in forum Space/Astronomy Questions and Answers
    Replies: 37
    Last Post: 2006-Aug-24, 09:42 PM
  4. Relativity(experiment DFERN and experiment DFERN b)
    By tsolkas in forum Against the Mainstream
    Replies: 3
    Last Post: 2006-Jan-27, 02:57 PM
  5. GR and Experiment
    By Wiley in forum Astronomy
    Replies: 6
    Last Post: 2002-Apr-29, 11:01 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •