Is clock synchronization compulsory

In summary, the answer is "no", clocks synchronization is not compulsory for deriving the fundamental equations of special relativity.
  • #36
doppler and his relatives

MeJennifer said:
You completely mix up theory and experiment here. :smile:

Doppler effects are phenomena of nature. By a set of experiments we can conclude that the Lorentz transforms are in accordance with experiment, as is the case with the theory of relativity.

Interesting point of view. Do you think that performing a given experiment in a given inertial frame we could obtain results which are conflicting with Newton and Galileo. Say the relationship between the proper mass and the relativistic mass (accept please that concept). We obtain some experimental results which can be plotted (m/m(0) as a function of u/c) looking for a function which reproduces best the experimental results. Extrapolation of the experimental results shows that they are best expressed by m=m(0)/((1-uu/cc)^1/2. (1) Consider three inertial reference frame I,I' and I(0). I(0) is the rest frame of the particle that moves with speed u relative to I with velocity u' relative to I, I' moving with V relative to I. Apply (1) from I and I'. Eliminate m(0) between the eqaution you obtain and express its right side as a function of u' only via the addition law of relativistic velocities obtaining the transformation equations for momentum and mass. Of course the same approach can be used in the case of many other relativistic effects. Do you think that such an approach presents some pedagogical advantages? IMHO the addition law of relativistic velocities can be derived without using the LT.
 
Physics news on Phys.org
  • #37
bernhard.rothenstein said:
As I see you are not able to give up your aggressive style. Should ridiculous be used in a scientific discussion? I have tried to help you and as I see you put questions concerning theirs content without consulting them. That is the style Latins named stante pede answer. I would avoid such a style. If you read Zhang you will see that the way in which he derives the Doppler formula, with arbitrary incidence angles, shows clearly that it holds only when it relates infinitesimal periods dt=F(theta',V/c)dt' i.e. very high frequencies. The Doppler formula obtained by using phase invariance and LT leads to the same formula but with finite periods.

Unfortunately this is not true: the Einstein derivation applies to ANY frequency. I do not know what gave you the idea about "finite periods" (you probably mean "discrete" periods) but either way, it is not true.


IMHO the theta Doppler formula you find in the literature holds only in the case of very high frequencies!

Whatever gave you this idea?

Could you convince me that "Einstein formula is fully general, there are no special cases, applies for all angles,frequencies and relative speeds between source and observer." of course in a scientific language.

Sure, read the original Einstein paper, I gave you the link about 3 times, maybe it is time that you read the paper.

But we are very far from my initial question: Is clock synchronization compulsory?

Yes. For the 4-th time. Read Einstein's paper.
 
  • #38
bernhard.rothenstein said:
The formula that accounts for the Doppler Effect relates two proper time intervals measured by the observers of the two reference frames respectively

Not really, it relates the observed frequency to the source frequency. See here:

http://en.wikipedia.org/wiki/Relativistic_Doppler_effect#For_motion_in_an_arbitrary_direction


and can be derived using initialized clocks theirs synchronization being not compulsory.

Not really, the derivation uses the Lorentz transforms. The Lorentz transforms have been derived assuming the Einstein clock synchronisation prior to the derivation of the Doppler effect. Both derivations happen in this particular sequence in the same paper. Here is the link to it :

http://www.fourmilab.ch/etexts/einstein/specrel/www/
 
  • #39
bernhard.rothenstein said:
IMHO the addition law of relativistic velocities can be derived without using the LT.

Don't think so, check this out:

http://en.wikipedia.org/wiki/Addition_of_velocities_formula


First sentence, from the top:

"A velocity addition formula appears in the special theory of relativity as a consequence of the Lorentz transformations"

For good reason: it is derived by simple chain differentiation of the Lorentz formulas. The very same way Einstein did it more than 100 years ago:

http://www.fourmilab.ch/etexts/einstein/specrel/www/

See paragraph 5. This paper contains a wealth of information, should set all your misconceptions straight, once you bite the bullet and read it.
 
Last edited:
  • #40
Doppler

nakurusil said:
Not really, it relates the observed frequency to the source frequency
frequency=1/period




Not really, the derivation uses the Lorentz transforms. The Lorentz transforms have been derived assuming the Einstein clock synchronisation prior to the derivation of the Doppler effect. Both derivations happen in this particular sequence in the same paper. Here is the link to it :

http://www.fourmilab.ch/etexts/einstein/specrel/www/
Do you think that the order in which formulas which account for a relativistic effect has relevance as long as Einsten's postulates are respected Sending people to Einstein for whom I have a profound respect makes the Forum useless as well as the work of those who try to present his theory with human face and more palatable
 
  • #41
doppler

nakurusil said:
Unfortunately this is not true: the Einstein derivation applies to ANY frequency. I do not know what gave you the idea about "finite periods" (you probably mean "discrete" periods) but either way, it is not true.




Whatever gave you this idea?



Sure, read the original Einstein paper, I gave you the link about 3 times, maybe it is time that you read the paper.

Please do not send me to Einstein because I respect his theory and know backward since youth. I also respect those who give simple derivations in accordance with Einstein's postulates. They are a good exercise before to read ALBERT my teacher. Even if I know that you do not read the references I give you, please have a look at
R.Neutze, William Moreau, "Frequency measurement by uniformly accelerating observers," Phys.Letters A 179 (1993) 389-390
W.Moreau, "Nonlocality in frequency measurement of uniformly accelerating observers," Am.J.Phys. 60, 561 (1992)[/COLOR]
I think that we have nonlocality in the case f=F(theta,V/c)f' as well because between the reception of two successive wave crests theta and the radial component of the relative velocity change. So using in that case the concept of instantaneous velocity at the moment when the observer receives a wave crest, we obtain results that hold only in the case of very small periods. As far as I know in a Doppler Effect we have to compare the period at which the source emits two successive wave crests measured in its rest frame and the period at which the same wave crets are received by the moving observer measured in its rest frame. Both are proper periods.

Not motivated, short no and yes answers are not usefull nor for me but neither for the participants on the Forum. Had you a look at Zhang? I could send you a copy of the pages of interest as an attachment if you give an address. I think it is worth for you to follow the piece of advise gave to you Robphy #33.
 
  • #42
bernhard.rothenstein said:
Please do not send me to Einstein because I respect his theory and know backward since youth.

Then read his general derivation: it has nothing to do with any "crests", it has none of the restrictions that you are claiming, it is very straightforward.



I think that we have nonlocality in the case f=F(theta,V/c)f' as well because between the reception of two successive wave crests theta and the radial component of the relative velocity change. So using in that case the concept of instantaneous velocity at the moment when the observer receives a wave crest, we obtain results that hold only in the case of very small periods.

This is too bad. Because the formula derived by Einstein and agreed upon by everybody holds for ANY frequency. So the thing with "holds only for very small periods" sounds like you are misunderstanding some derivation or you stumbled on a bad one and you are holding it to being correct.


As far as I know in a Doppler Effect we have to compare the period at which the source emits two successive wave crests measured in its rest frame and the period at which the same wave crets are received by the moving observer measured in its rest frame. Both are proper periods.

As long as you do a proper derivation, yes. If you come up with a derivation that "holds only for very small periods", it is all bogus.


Had you a look at Zhang? I could send you a copy of the pages of interest as an attachment if you give an address.

Zhang is a very experienced physicist, is is unlikely that he made such a mistake. Did you consider the possibility that you are misinterpreting his writings? Why don't you scan the relevant page, turn it into a JPEG and attach it to the next post so we can all have a look at what you think is a "Doppler formula that holds only for very small periods"
 
Last edited:
  • #43
how correct is the "general" Doppler shift formula?

Zhang is a very experienced physicist, is is unlikely that he made such a mistake. Did you consider the possibility that you are misinterpreting his writings? Why don't you scan the relevant page, turn it into a JPEG and attach it to the next post so we can all have a look at what you think is a "Doppler formula that holds only for very small periods

As you suggested I present Zhang's derivation.
Consider a source of light located at the origin O of the I frame and a clock C located at that point. When C reads t(e) S emits a light signal along a direction that makes an angle w with the positive direction of the OX axis. At a time t(r) the light signal generates the event E(x=rcosw,y=rsinw,t(r)) and we have
t(r)=t(e)+r/c (1)
Differentiating (1) we obtain
dt(r)=dt(e)+dr/c (2)
which holds for each value of t(e) even for t(e)=0. Taking into account that by definition dr/dt(r)=Vcosw (3) represents the instantaneous radial component of an observer of the I' frame (2) leads to
dt(e)/dt(r)=1-(V/c)cosw (3)
(equation 2.10.17) in Zhang's derivation.) In (3) dt(e) represents a proper time interval whereas dt(r) represents a coordinate time interval. Taking into account the time dilation formula
dt'(r)=dt(r)/g(V) (4)
dt'(r) representing a proper time interval (3) becomes
dt(e)/dt'(r)=g(V)[1-(V/c)cosw. (5)
Arrived at that point Zhang considers that in the infinitezimal time dt(e) the source emits dn wave crests which are all received by the observer and defines the correspnding frequencies
dn/dt(e)=f(e) (5)
dn/dt(r)=f''(r) (6)
obtaining the final result
f'(r)=f(e)[1-(V/c)cosw]/g(V) (7)
equation (2.10.22) in Zhangs derivation.

I add to all that my own comments.
1. Equation (7) is the same as that derived from phase invariance and Lorentz transformations.
2.It holds only in the case of very small periods the velocity defined by (3)representing an instantaneous velocity.
3.The derivation involves the concept of wave crest which is not mentioned in the derivation to which you send me obstinately.
4.In the case of both derivations (7) holds only in the case of the first pair of received wave crests because for the second pair the angle w changes.
5.Considering the inverse transformation of (7) in which appears the angle w' and eliminating between them f(e) and f'(r) we obtain the formula which accounts for the aberration of light effect which IMHO uses initialized and not synchronized clocks.

6.Because dn=1 is a realistic value we see that the involved frequencies are infinite and so the associated photons would burn all they find in their way.
7.Leading to the same results the two derivations have the same physics behind them.
For more conformity please have a look at the original version of Zhang.

Please take a "time out" before an instantaneous answer and give punctual answers, inserted in my text taking into account that you consider that the Author is a serious one.
I would highly appreciate Albert's oppinion, but that is not possible, taking into account that he was open minded.
 
  • #44
Thank you for taking the effort to post Zhang's solution.

bernhard.rothenstein said:
Zhang is a very experienced physicist, is is unlikely that he made such a mistake. Did you consider the possibility that you are misinterpreting his writings? Why don't you scan the relevant page, turn it into a JPEG and attach it to the next post so we can all have a look at what you think is a "Doppler formula that holds only for very small periods

As you suggested I present Zhang's derivation.
Consider a source of light located at the origin O of the I frame and a clock C located at that point. When C reads t(e) S emits a light signal along a direction that makes an angle w with the positive direction of the OX axis. At a time t(r) the light signal generates the event E(x=rcosw,y=rsinw,t(r)) and we have
t(r)=t(e)+r/c (1)
Differentiating (1) we obtain
dt(r)=dt(e)+dr/c (2)
which holds for each value of t(e) even for t(e)=0. Taking into account that by definition dr/dt(r)=Vcosw (3) represents the instantaneous radial component of an observer of the I' frame (2) leads to
dt(e)/dt(r)=1-(V/c)cosw (3)
(equation 2.10.17) in Zhang's derivation.) In (3) dt(e) represents a proper time interval whereas dt(r) represents a coordinate time interval.

So far , so good.

Taking into account the time dilation formula
dt'(r)=dt(r)/g(V) (4)

The time dilation formula (4) is a DIRECT consequence of the Lorentz transform for time. So, Zhang's derivation is a consequence of the Lorentz transforms. I hope you realize that you just proved my point.



Now, to some secondary issues:

dt'(r) representing a proper time interval (3) becomes
dt(e)/dt'(r)=g(V)[1-(V/c)cosw. (5)
Arrived at that point Zhang considers that in the infinitezimal time dt(e) the source emits dn wave crests which are all received by the observer and defines the correspnding frequencies
dn/dt(e)=f(e) (5)
dn/dt(r)=f''(r) (6)
obtaining the final result
f'(r)=f(e)[1-(V/c)cosw]/g(V) (7)
equation (2.10.22) in Zhangs derivation.

So what gave you the idea that the formula applies only to very high frequencies? As an aside, how high do these frequencies have to be?


.
2.It holds only in the case of very small periods the velocity defined by (3)representing an instantaneous velocity.

Because he used differentiation?

3.The derivation involves the concept of wave crest which is not mentioned in the derivation to which you send me obstinately.

Yes, Einstein's standard derivation doesn't need any of the "crests". So?

4.In the case of both derivations (7) holds only in the case of the first pair of received wave crests because for the second pair the angle w changes.

This would be very bad, if your statement were true, Zhang's derivation would be invalid. So, either Zhang's derivation is worthless or you didn't understand it. Without a picture it is hard to tell but I am willing to bet, based on your other misunderstandings , that you didn't understand this one either.




5.Considering the inverse transformation of (7) in which appears the angle w' and eliminating between them f(e) and f'(r) we obtain the formula which accounts for the aberration of light effect which IMHO uses initialized and not synchronized clocks.

Just another misunderstanding: (7) has been obtained using the Lorentz transforms in expression (4). The derivation of the Lorentz transforms is based on Einstein's clock synchronisation.

6.Because dn=1 is a realistic value we see that the involved frequencies are infinite and so the associated photons would burn all they find in their way.

This is incomprehensible. Can you try again?
Ahh, I think I know what you are trying to say: do you really think that Zhang's formula doesn't apply for dn infinitely small? That there may be some magical lower bound to dn? dn is an infinitesimal quantity, therefore it is infinitely small.
 
Last edited:
  • #45
doppler again

nakurusil said:
Thank you for taking the effort to post Zhang's solution.



So far , so good.



The time dilation formula (4) is a DIRECT consequence of the Lorentz transform for time. So, Zhang's derivation is a consequence of the Lorentz transforms. I hope you realize that you just proved my point.

In our discussion it is not important that he uses the LT. What he shows is that the Doppler formula holds only at very small periods!

I told you for many times that many authors derive the time dilation formula without using the LT. They also derive the addition law without using them. The addition law of velocities leads directly to the LT. In your usual way to answer you told me that all that is trash and the journals which publish them have no scientific value. So that point is closed!



Now, to some secondary issues:



So what gave you the idea that the formula applies only to very high frequencies? As an aside, how high do these frequencies have to be?

I do not speak in terms of frequency but in terms of periods. The periods should be small enough in order to ensure that the velocity in the Doppler formula is an instantaneous one. That is the case in the classic derivation as well where the frequency in the phase is an instantaneous one and so not measurable.




Because he used differentiation?
Of course and because the involved periods in the Doppler shift formula are infinitezimal dt(e) and dt(r) and so theirs inverses are infinite.



Yes, Einstein's standard derivation doesn't need any of the "crests". So?

Without crests there is no Doppler Effect! Einstein's standard derivation applies the LT to a single point of the space through which the wave propagates considering that there the observer could receive two successive crests, making the very small period assumption.



This would be very bad, if your statement were true, Zhang's derivation would be invalid. So, either Zhang's derivation is worthless or you didn't understand it. Without a picture it is hard to tell but I am willing to bet, based on your other misunderstandings , that you didn't understand this one either.
Zhang does not present any picture. Sapienti sat! The aggressive style prevent me to answer because I could say that based on your misunderstandings... We are in the field of relativity.







Just another misunderstanding: (7) has been obtained using the Lorentz transforms in expression (4). The derivation of the Lorentz transforms is based on Einstein's clock synchronisation.
Here our oppinions diverge and so I do not insist. Please do not send me to the Classic because I could send you to modernists




This is incomprehensible. Can you try again?
Ahh, I think I know what you are trying to say: do you really think that Zhang's formula doesn't apply for dn infinitely small? That there may be some magical lower bound to dn? dn is an infinitesimal quantity, therefore it is infinitely small.

I quote Zhang " We now assume that the n-th and the (n+dn)-th crests are received at the point O (i.e. at the same point in space,locality assumption, my remark). If I remember well from analysis very small means that during it nothing changes!
As I have mentioned in the previous thread the Doppler formula and its inverse contain the angles w and w'. Handling them we obtain the aberration of light effect which is not associated with clock synchrnization involving only initialization. How do you explain that?
Consider a scenarion in which the observer moves with constant velocity parallel to the OX axis at y=constant, the source being located at the origin. IMHO the period he measures is continuosly changing a fact not allways mentioned in the literature. Not taking into account the nonlocality at high periods we make errors!





I have tried to present Zhang's derivation with my comments. It leads to the same formula as phase invariance and LT do. How do you explain that fact? I think that it is the result of the fact that the derivation you like, obscures some interesting pecualiarities of the Doppler Effect. I know some modernits who consider that the use of the LT obscures the physics behind the studied effects. Discussing with me please avoid a personal address without to put in question my ability to understand the stuff we discuss. I could suppose the same thing about you. We are in relativity! I will no longer answer your comments if you do not give up that unpolite and offending style, a fact I have mentioned for so many times.

I invite people on the Forum to participate to the discussion. The stuff is interesting and all of us have to learn from it!
 
  • #46
bernhard.rothenstein said:
nakurusil said:
The time dilation formula (4) is a DIRECT consequence of the Lorentz transform for time. So, Zhang's derivation is a consequence of the Lorentz transforms. I hope you realize that you just proved my point.

In our discussion it is not important that he uses the LT. What he shows is that the Doppler formula holds only at very small periods!

Of course this is not true, you are misinterpreting his work.

I told you for many times that many authors derive the time dilation formula without using the LT.

1. So, the one that you claimed not to be using LT (Zhang's) turned out to...be using LT. Try another paper, please, you failed on this attempt.

They also derive the addition law without using them.

2. Please provide one from a reputable source after you prove your first point. You haven't done it yet. (see point 1)

I do not speak in terms of frequency but in terms of periods. The periods should be small enough in order to ensure that the velocity in the Doppler formula is an instantaneous one. That is the case in the classic derivation as well where the frequency in the phase is an instantaneous one and so not measurable.

3. Why makes you persist in this nonsense? It is clearly non-physical.
Of course and because the involved periods in the Doppler shift formula are infinitezimal dt(e) and dt(r) and so theirs inverses are infinite.

4. You realize that this is wrong, why do you cling to it?
Without crests there is no Doppler Effect! Einstein's standard derivation applies the LT to a single point of the space through which the wave propagates considering that there the observer could receive two successive crests, making the very small period assumption.

5. There is no very small period assumption anywhere in the description of the relativistic Doppler effect. There isn't any in the Zhang description, it is all based on your misinterpretation of his derivation.
 
Last edited:
  • #47
bernhard.rothenstein said:
I will no longer answer your comments if you do not give up that unpolite and offending style, a fact I have mentioned for so many times.

I invite people on the Forum to participate to the discussion. The stuff is interesting and all of us have to learn from it!
Thanks Bernard.

By the way, I already lost my appetite in having any discussion with nakurusil due to his attitude here on PF.
 
  • #48
clock synchronization

MeJennifer said:
Thanks Bernard.

By the way, I already lost my appetite in having any discussion with nakurusil due to his attitude here on PF.

Please stay here. Probably together with others we will convince him to do not mix scientific discussions wih offending people.
 
  • #49
Clock Synchronization

I postpone my answer until others will express their oppinion concerning the stuff under discussion.
If you think that you can convince me and as I see others in a dictatorial style (You told me that the Forum is not mine. Because we are relativists I could say that it does not belong to you.) simply stating that I do not understand things which are obvious and without giving any prove then you are in error.
 
  • #50
Bernard

I doubt I’ll convince you, as you are an old fox set in your relativistic ways. But your posts are interesting and warrant courteous reply.

In response to whether you can derive the relativistic Doppler equation without using Lorentz Transformations, and without need of Einstein’s clock synchronization.
The answer is yes.
See equation set 9.6 and 9.7
http://uk.geocities.com/kevinharkess/wisp_ch_9/wisp_ch_9.html

The general Doppler equation (9.6) includes arbitrary angles for both source and observer with respect to ether. Clocks are initialized at origin t’= t=0, and thereafter do not need synchronizing.
The equation set produces a match for the relativity’s velocity addition formulae only if the observer and source angles are zero.

The equation set (9.7) matches relativity’s Doppler equation for an arbitrary angle when the observer’s speed through the ether is zero.

Regardless of whether you support ether concept, it’s fascinating that the derived general Doppler equation can mirror relativity this way.
 
Last edited by a moderator:
  • #51
wisp said:
Bernard

I doubt I’ll convince you, as you are an old fox set in your relativistic ways. But your posts are interesting and warrant courteous reply.

In response to whether you can derive the relativistic Doppler equation without using Lorentz Transformations, and without need of Einstein’s clock synchronization.
The answer is yes.
See equation set 9.6 and 9.7
http://uk.geocities.com/kevinharkess/wisp_ch_9/wisp_ch_9.html

This is just an old and tired crank theory,well known over the internet, we do not discuss crank theories in this forum.
 
Last edited by a moderator:
  • #52
ether and Doppler shift

wisp said:
Bernard

I doubt I’ll convince you, as you are an old fox set in your relativistic ways. But your posts are interesting and warrant courteous reply.

In response to whether you can derive the relativistic Doppler equation without using Lorentz Transformations, and without need of Einstein’s clock synchronization.
The answer is yes.
See equation set 9.6 and 9.7
http://uk.geocities.com/kevinharkess/wisp_ch_9/wisp_ch_9.html

The general Doppler equation (9.6) includes arbitrary angles for both source and observer with respect to ether. Clocks are initialized at origin t’= t=0, and thereafter do not need synchronizing.
The equation set produces a match for the relativity’s velocity addition formulae only if the observer and source angles are zero.

The equation set (9.7) matches relativity’s Doppler equation for an arbitrary angle when the observer’s speed through the ether is zero.

Regardless of whether you support ether concept, it’s fascinating that the derived general Doppler equation can mirror relativity this way.

Thank you for your message. I respect all my coleagues and my contact with "etherits" (in the history of my country "etherits' were the members of a social mouvement") convinced me that theirs results and those obtained by Einstein, are the same. It is importamt to underline that in the case of the Doppler Effect both theories lead compulsory to the same results because it does not involve clock synchronization. I think the same situation is compulsory in the case of the aberration of light effect. I am shure that the link you gave me will confirm that fact.
I have seen a very interesting "etherist" approach to the Doppler effect presented by Selleri showing that the formula he obtains and that obtained by Einstein are the same. Asking him why not to make use of Ockhams razor, using the simplest theory, he told me that there is place for everibody.
All the best.
 
Last edited by a moderator:
Back
Top