The only way for the clock to run faster is for time to run faster on the satellite.
Are you sure?
Yes. It's the only viable explanation we've got. And it's been put through the wringer like no other, and passed all tests. Feel free to prove yourself smarter than Einstein and all the relativists that have studied the subject in the past hundred years, and propose another
viable explanation which explains
all the observed facts that General Relativity does.
Someone may come up with a different theory. But it's going to have to predict all the same things that GR does except for extremes of subatomic sizes and/or black-hole-strength gravitational fields. At velocities and gravitational fields we encounter in our solar system, the predictions of GR are 100% true and confirmed.
That's the only thing that can affect the frequency of cesium transitions.
Then how do they adjust the frequency to sync the satellite atomic clock's to the stationary earth atomic clock's?
They don't adjust the frequency. They can't control that. They change the number of transitions that pass before the clock registers one second. An atomic clock on Earth counts 9,192,631,770 transitions before it registers one more second. A GPS satellite clock on Earth counts 9,192,631,770 - 26,316 = 9,192,605,454 transitions (26,316 = 1/38*10
-6, the difference between satellite time and Earth time) before it registers one second. In orbit, the clock continues to count 9,192,605,454 transitions before it registers one second…
and these 9,192,605,454 transitions take the same amount of time as 9,192,631,770 transitions on the Earth-based clock because time is running faster on the satellite.
(How it's actually done is more complex, involving PLLs and computers and all sorts of fancy circuitry, but what's actually done is equivalent to what I described above.)
Which is done once each day.
The clocks are reset once each day because there are other effects that make them drift. These are much smaller than the main relativistic corrections. If the relativistic corrections were not applied the clocks would have to be reset every few minutes. And that's exactly what they are doing each day; resetting the clocks, which doesn't affect the transition frequency. Just as setting the time on your alarm clock at home doesn't affect the frequency of the quartz crystal that it uses to measure time.
I am not that smart to propose a theory but how about this
one
I'm not impressed that you can find a uneducated nutjob to back up your claims; you can find anything on the Intertubes. It's up to you to demonstrate that his analysis is valid (Hint: it's trivial to show that it ain't, 'cause the satellites are continuously accelerating but he's trying to analyze the situation using only SR. QED.)
That is not what relativity says will happen
Mostly, it is. The gobbledygook about "Time remain unaffected…" is just that: usupported gobbledygook. We have experience measuring the lifetime of muons far from the surface of the Earth, and it's up to you to demonstrate that our experience is somehow inadequate. The twins do not wind up the same age; I like
http://mentock.home.mindspring.com/twins.htm as a simple explanation of why.
Just a crazy question I have. If the duration of a second is determined by dividing up the duration of a rotation of the earth on it axis, wouldn't the earth have to slow down to make time slow down for the satellite clock?
The duration of a second once had something to do with the rotation of the Earth, but it has not had anything to do with the rotation of the Earth for decades. The duration of a second is 9,192,631,770 cycles of radiation corresponding to the transition between two energy levels of the cesium-133 atom (at Earth's surface) and has been since 1967.
However, the second is just a unit we choose to measure time; time itself (really spacetime) is not dependent on the tools we use to measure it. Just like length exists whether we measure it with a wooden ruler or a steel tape. Even if the second were defined as 1/5,284,000 of the time it takes for the Earth to rotate on its axis, time would still slow down on the satellite clock without changing the Earth's rotation rate
because time passes differently on the satellite clock relative to the Earth clock whether the Earth clock is a cesium atomic clock or a rotating Earth or your alarm clock at home or anything that measures time.
We do have to add leap seconds to the atomic clocks because the earth's rotation is slowing down.
Yup. So what? We like to have Mean Solar Time (based on the Earth's rotation) close to Universal Coordinated Time (based on atomic clocks). But that's just an arbitrary preference we've chosen. Leap seconds are not necessary to measuring time, and could be discarded without affecting the fact that the time runs faster on the satellite than it does on Earth.
Whether I like or dislike GR has nothing to do with it being right or wrong. It doesn't make any difference what I think. It is either right or it is wrong. There are things that are wrong.
List 'em. The "paper" you linked to doesn't help you.
Your assertions don't make it right either
Nobody's assertions do. But I'm not going to try to reproduce the equations and figures and explanations in this medium Here's some links that support my
explanations.
GPS and Relativityhttp://tycho.usno.navy.mil/ptti/1996/Vol%2028_16.pdfDealing with Creationism in Astronomy: Scott Rebuttal. I. GPS & RelativityChangesHome | Navigation Center
I also recommend
Global Positioning System: Theory and Practice, but it's a little expensive.
™ Version 4.2