PolymerTim
Senior member
- Apr 29, 2002
- 383
- 0
- 0
Sorry to comment on a side-argument of the original poster, but it seems there is a lot of discussion about the "atomic clocks in a plane" experiment without anyone bothering to go look it up. My favorite part about being in a university is the electronic library so I took 5 minutes to read the paper and even in its short 4 pages, most of these concerns are addressed. I like this experiment since it avoids the whole issue of "light playing tricks" since the clocks start and end at the same point for comparison. To argue with this one, I think you would have to show that the trip somehow affected the vibrational frequency of the Cs without affecting time. I'm not really sure how you can separate those two, but it may not matter. If the trip affects all matter the way it does Cs vibrations then maybe that is the same thing as slowing down time.
Here's my quick summary from the paper for those who can't get it themselves.
From "Science Vol. 177 pg 166--170 (1972) (Experiment)." (My paraphrasing)
It is known that no two "real" cesium beam clocks keep precisely the same time (differing from each other from up to 1 microsec per day) and even are not completely constant since shot noise in the beam tubes cause short term fluctuations in rate. Intervals between these fluctuations can range from days to months and are unpredictable, but random. Realizing this complication, the researchers did not put one cesium beam clock on the commercial jet-liner, they put 4. In this way, any deviations in one clock from the others could be attributed to noise on that clock and factored out.
The difference in time between each of the 4 clocks and the reference ground clock was monitored for ~240 hours before being placed on a commercial jetliner and flown eastward around the earth for ~65 hours (41 in flight). Once back at the original location, the clocks were monitored for another 150 hours before being put on another jetliner flying westward for ~80 hours (49 in flight). Once at home again, the clocks were monitored for another ~100 hours. A plot of the time differences for each clock and their average is presented in the paper and shows that there is in fact a significant spread between them as expected. In this plot, time difference from ground clock (in ns) is plotted vs time (hours) covering the whole experiment. The slope of each curve for 25 hours immediately prior to flight for each clock and the mean is extrapolated linearly to the time point after the trip and compared to the first actual data point after the trip and the results reported are thus:
Clock serial # | dt(ns) east | dt(ns) west
120 | -57 | 277
361 | -74 | 284
408 | -55 | 266
447 | -51 | 266
Mean| -59 +/- 10 | 273 +/- 7
Pred| -40 +/- 23 | 275 +/- 21
Note that the predicted values shown at the bottom of the table are based off of expectations based on SR and GR. GR was used to add a kinetic term due to the ground clock and flying clocks being at different altitudes. Their final equation combines these effects into an integral over delta time for the entire known flight path, which was carefully recorded. As you can see, the agreement between the theory and the results is very convincing.
Error in the experiment was considered carefully. In addition to random errors, accounted for by multiple clocks, systematic errors were tested as well. Effects of temperature, pressure, magnetic fields, and impulse acceleration were tested on ground clocks and proven to be non-systematic if present at all.
Here's my quick summary from the paper for those who can't get it themselves.
From "Science Vol. 177 pg 166--170 (1972) (Experiment)." (My paraphrasing)
It is known that no two "real" cesium beam clocks keep precisely the same time (differing from each other from up to 1 microsec per day) and even are not completely constant since shot noise in the beam tubes cause short term fluctuations in rate. Intervals between these fluctuations can range from days to months and are unpredictable, but random. Realizing this complication, the researchers did not put one cesium beam clock on the commercial jet-liner, they put 4. In this way, any deviations in one clock from the others could be attributed to noise on that clock and factored out.
The difference in time between each of the 4 clocks and the reference ground clock was monitored for ~240 hours before being placed on a commercial jetliner and flown eastward around the earth for ~65 hours (41 in flight). Once back at the original location, the clocks were monitored for another 150 hours before being put on another jetliner flying westward for ~80 hours (49 in flight). Once at home again, the clocks were monitored for another ~100 hours. A plot of the time differences for each clock and their average is presented in the paper and shows that there is in fact a significant spread between them as expected. In this plot, time difference from ground clock (in ns) is plotted vs time (hours) covering the whole experiment. The slope of each curve for 25 hours immediately prior to flight for each clock and the mean is extrapolated linearly to the time point after the trip and compared to the first actual data point after the trip and the results reported are thus:
Clock serial # | dt(ns) east | dt(ns) west
120 | -57 | 277
361 | -74 | 284
408 | -55 | 266
447 | -51 | 266
Mean| -59 +/- 10 | 273 +/- 7
Pred| -40 +/- 23 | 275 +/- 21
Note that the predicted values shown at the bottom of the table are based off of expectations based on SR and GR. GR was used to add a kinetic term due to the ground clock and flying clocks being at different altitudes. Their final equation combines these effects into an integral over delta time for the entire known flight path, which was carefully recorded. As you can see, the agreement between the theory and the results is very convincing.
Error in the experiment was considered carefully. In addition to random errors, accounted for by multiple clocks, systematic errors were tested as well. Effects of temperature, pressure, magnetic fields, and impulse acceleration were tested on ground clocks and proven to be non-systematic if present at all.