Tangherlini transformations and absolute simultaneity: a rotor model

  • Level: High School 
  • Thread starter AlexOsipovich
  • Start date
21 replies · 911 views · last reply
AlexOsipovich
Messages
17
Reaction score
3
TL;DR
Comparing rotating rotor kinematics via Tangherlini transforms (absolute simultaneity, 1958) and standard STR.
To be clear on the framework: I am using the transformations introduced by F. R. Tangherlini in his 1958 PhD dissertation (dissertation), specifically the framework detailed in Chapter 3: 'Velocity of light and synchronization of clocks under the Absolute Lorentz Transformation'. In this framework, absolute simultaneity is preserved, and the one-way velocity of light becomes anisotropic while the out-and-back speed remains invariant. The goal is to see how this kinematical choice affects the coordinate time of a rotating rotor in the absolute frame.

I have made some adjustments to the model and would appreciate your input on the symmetry of its half-turns.

The Setup:
• A laboratory moves through the AFR at a constant velocity V = 370 km/s along the X-axis.
• Inside the lab, a rotor of radius R = 1 m spins uniformly with a tangential speed u = 600 m/s.
• Two stationary triggers (Sensor 1 and Sensor 2) are placed on the stator diametrically opposite each other along the X-axis.
• Two clocks (Clock A and Clock B) are mounted on the rotor rim 180° apart:
• Clock A runs only during the first half-turn (from Sensor 1 to Sensor 2, moving against the V direction).
• Clock B runs only during the second half-turn (from Sensor 2 back to Sensor 1, moving along the V direction).

My Question:
Could you please clarify whether the half-turns of the rotor will be equal in duration according to AFR coordinate time, if calculated using Tangherlini transformations?

My preliminary calculations show a stark contrast: in Tangherlini's framework (absolute simultaneity), the sensors trigger simultaneously and coordinate time is symmetric, yet proper time differs. In standard STR (relativity of simultaneity), the sensors trigger at different times, exactly compensating the time dilation to yield zero proper time difference.

I have a Python simulation and mathematical derivation for both cases. I can share the full code or equations in this thread if you are interested in reviewing the math. I'm looking forward to your feedback on the conceptual differences.
 
I'm not going to download a doc from a random Google drive, and the only references to Tangherlini transforms I can find online are from fringe sources or worse.

That said, you seem to be describing a coordinate transform that picks one normal SR frame, and generates other "inertial frames" by picking the x coordinate of the co-moving SR frame but sticking with the time coordinate of the frame you first thought of. Note that the results are not frames, except for the original. They're non-orthogonal coordinate systems, and the description of physics in these systems has a lot of traps for the unwary.

To start with, a free-to-rotate rotor will rotate with constant tip velocity in an inertial frame, but a varying velocity in a non-orthogonal coordinate system. So if you mean that you set the tangential velocity of your rotor as ##u=|d\vec{x}'/dt'|=\mathrm{const}##, that's probably where you went wrong - you're describing a rotor subject to a periodically varying torque.

The Right Way to do this kind of analysis is to write down your laws of motion in tensorial form and then get the answer out of that. In this case, though, you can bodge it by writing down the motion of your rotor in its Einstein rest frame, transforming to an Einstein frame where it's moving, then declaring that to be the privileged frane and using the Tangherlini transforms to get a description in the rotor's rest "frame" in his more complicated coordinate system.
 
Ibix said:
I'm not going to download a doc from a random Google drive, and the only references to Tangherlini transforms I can find online are from fringe sources or worse.

My bad on the Google Drive link! It's actually Frank Tangherlini's 1958 Stanford PhD thesis ("The Velocity of Light in Uniformly Moving Frames", supervised by Sidney Drell), which was also reprinted in The Abraham Zelmanov Journal (Vol. 2, 2009).
I found a copy hosted on the SciSpace academic repository here:
https://scispace.com/pdf/the-velocity-of-light-in-uniformly-moving-frames-1sw3kem72z.pdf
 
AlexOsipovich said:
My bad on the Google Drive link! It's actually Frank Tangherlini's 1958 Stanford PhD thesis ("The Velocity of Light in Uniformly Moving Frames", supervised by Sidney Drell), which was also reprinted in The Abraham Zelmanov Journal (Vol. 2, 2009).
I found a copy hosted on the SciSpace academic repository here:
https://scispace.com/pdf/the-velocity-of-light-in-uniformly-moving-frames-1sw3kem72z.pdf
Life's too short!
 
PeroK said:
Life's too short!

Human life is indeed too short, particularly compared to an AI. OP reported.
 
So the metric in equation 3.19 is (in spacelike positive signature) $$ds^2 = -dt'^2+2 v_i \ dx'^i dt' + dx'^i \ dx'^j - v_i v_y \ dx'^i dx'^j$$

The Anderson metric is $$ds^2 = -dt^2+ dx^2 + dy^2 + dz^2 - 2 \kappa \ dt \ dx - \kappa^2 \ dx^2$$

So, by inspection we can see that the Tangerlini transform is a special case of Anderson's synchronization. Specifically for the case ##\kappa = -v##.

AlexOsipovich said:
Two clocks (Clock A and Clock B) are mounted on the rotor rim 180° apart:
How are these clocks synchronized?

AlexOsipovich said:
Two clocks (Clock A and Clock B) are mounted on the rotor rim 180° apart:
Specifically where? There are multiple positions that are 180 degrees apart.

AlexOsipovich said:
a rotor of radius R = 1 m spins uniformly
About which axis is it spinning?
 
The "Tangherlini transform"
##x'=\gamma(x-vt)##
##t'=t/\gamma ##
seems to be the same as the one given by Mansouri-Sexl ("relativity without relativity", p. 503): Mansouri R.; Sexl R.U. (1977). "A test theory of special relativity. I: Simultaneity and clock synchronization". Gen. Rel. Gravit. 8 (7): 497–513

See also p. 143 (eq. 62+63) in Anderson, R.; Vetharaniam, I.; Stedman, G.E. (March 1998). "Conventionality of synchronisation, gauge dependence and test theories of relativity". Physics Reports. 295 (3–4): 93–180"
 
  • Like
Likes   Reactions: Sagittarius A-Star, Ibix and Dale
Histspec said:
The "Tangherlini transform"
##x'=\gamma(x-vt)##
##t'=t/\gamma ##
seems to be the same as the one given by Mansouri-Sexl
Ah, that explains why Anderson did not cite Tangherlini. He spent a great deal of time discussing Mansouri-Sexl instead.
 
Ibix said:
To start with, a free-to-rotate rotor will rotate with constant tip velocity in an inertial frame, but a varying velocity in a non-orthogonal coordinate system. So if you mean that you set the tangential velocity of your rotor as ##u=|d\vec{x}'/dt'|=\mathrm{const}##, that's probably where you went wrong - you're describing a rotor subject to a periodically varying torque.
Thank you for pointing out the error, the problem lay in how I determined the rotation speed - to determine it, I used clocks synchronized according to the Einstein method.

After recalculation, the half-turns of the disk in the AFR result in different durations.
Dale said:
So, by inspection we can see that the Tangerlini transform is a special case of Anderson's synchronization. Specifically for the case κ=−v.
I haven’t seen Anderson’s work, unfortunately; I only studied Tangherlini’s dissertation.

However, I noticed an interesting feature: the modified Newton's law does not determine the elliptical trajectory itself - rather, it governs the speed of the point mass along that path. The elliptical geometry is dictated purely by the Lorentz contraction of the crystal lattice, which arises from the redistribution of electromagnetic forces in motion.

1794901608490.webp


Could you please explain how this law gives rise to an elliptical trajectory?
 
Histspec said:
The "Tangherlini transform"
##x'=\gamma(x-vt)##
##t'=t/\gamma ##
seems to be the same as the one given by Mansouri-Sexl ("relativity without relativity", p. 503): Mansouri R.; Sexl R.U. (1977). "A test theory of special relativity. I: Simultaneity and clock synchronization". Gen. Rel. Gravit. 8 (7): 497–513

See also p. 143 (eq. 62+63) in Anderson, R.; Vetharaniam, I.; Stedman, G.E. (March 1998). "Conventionality of synchronisation, gauge dependence and test theories of relativity". Physics Reports. 295 (3–4): 93–180"
Yes, those are the exact same transformations! Mansouri and Sexl (1977) later popularized them within their test theory framework.
However, as far as I know, Frank Tangherlini originally derived these exact equations in his 1958 Stanford PhD dissertation ("The Velocity of Light in Uniformly Moving Frames").
 
This is an older PF thread about the Tangherlini transformation:

 
AlexOsipovich said:
• Two stationary triggers (Sensor 1 and Sensor 2) are placed on the stator diametrically opposite each other along the X-axis.
• Two clocks (Clock A and Clock B) are mounted on the rotor rim 180° apart:
• Clock A runs only during the first half-turn (from Sensor 1 to Sensor 2, moving against the V direction).
• Clock B runs only during the second half-turn (from Sensor 2 back to Sensor 1, moving along the V direction).
Wouldn't clock B move from Sensor 2 to Sensor 1 in the same half-turn, in which clock A moves from Sensor 1 to Sensor 2?
 
Sagittarius A-Star said:
Wouldn't clock B move from Sensor 2 to Sensor 1 in the same half-turn, in which clock A moves from Sensor 1 to Sensor 2?
Clock A moves along the upper arc (from Sensor 1 to Sensor 2) at the same time as Clock B moves along the lower arc (from Sensor 2 to Sensor 1).

It all depends on the chosen reference frame and how the clocks in that frame are synchronized:
• Under Einstein synchronization (in the lab frame where the stator is at rest), both half-turns take the exact same amount of time.
• Under Tangherlini synchronization (which enforces absolute simultaneity), the events at Sensor 1 and Sensor 2 are no longer simultaneous in the moving frame, so the measured durations of the two half-turns ("with the wind" vs. "against the wind") will differ.
 
  • Like
Likes   Reactions: Sagittarius A-Star
AlexOsipovich said:
• Under Einstein synchronization (in the lab frame where the stator is at rest), both half-turns take the exact same amount of time.
• Under Tangherlini synchronization (which enforces absolute simultaneity), the events at Sensor 1 and Sensor 2 are no longer simultaneous in the moving frame, so the measured durations of the two half-turns ("with the wind" vs. "against the wind") will differ.
Einstein synchronization works only for clocks at rest in the reference frame.
Do you mean an Einstein synchronization of clocks A and B with reference to the rotating frame with light pulses forth and back along the rim of the disk?

Selleri used the Tangherlini transformation to explain the Sagnac effect in the rotating frame. He defined the one-way-speed of light to be anisotropic along the rim in the rotating frame and to be isotropic in the lab frame.

You describe three frames:
  1. the stationary frame
  2. the moving lab frame
  3. the frame rotating in the lab

Between which two of those frames do you transform with the Tangherlini transformation, and which of the two frames is defined to have the isotropic one-way-speed of light?
 
Sagittarius A-Star said:
Einstein synchronization works only for clocks at rest in the reference frame.
Do you mean an Einstein synchronization of clocks A and B with reference to the rotating frame with light pulses forth and back along the rim of the disk?
No, I meant the synchronization of clocks in the laboratory reference frame.

I intended to use clocks A and B as timers to measure the time elapsed during a half-rotation. Clock A would measure only the upper half-rotation, while clock B would measure the lower one.
They are not synchronized with each other.

Sagittarius A-Star said:
You describe three frames:
  1. the stationary frame
  2. the moving lab frame
  3. the frame rotating in the lab

Between which two of those frames do you transform with the Tangherlini transformation, and which of the two frames is defined to have the isotropic one-way-speed of light?
I wanted to perform the inverse Tangherlini transformations (2 ---> 1) from the inertial laboratory frame to a hypothetical absolute reference frame that is at rest relative to the ether.

The laboratory reference frame moves relative to a hypothetical ether, consequently, the physical speed of light (one-way speed) in this frame is anisotropic.
 
AlexOsipovich said:
The laboratory reference frame moves relative to a hypothetical ether, consequently, the physical speed of light (one-way speed) in this frame is anisotropic.
If the one-way-speed is isotropic or not with reference to an inertial frame is a matter of definition. It depends on how the clock synchonization is defined. It is not "physical".

Source:
https://en.wikipedia.org/wiki/One-way_speed_of_light
 
Could you please help me look into another issue related to the previous topic(link)?
The team dismissed it, citing a modified version of Newton’s law. At first, I thought they were right, but today I ran some simulations based on this topic, and now I’m no longer certain that they were correct.


AlexOsipovich said:
However, I noticed an interesting feature: the modified Newton's law does not determine the elliptical trajectory itself - rather, it governs the speed of the point mass along that path. The elliptical geometry is dictated purely by the Lorentz contraction of the crystal lattice, which arises from the redistribution of electromagnetic forces in motion.

1794901608490.webp


Could you please explain how this law gives rise to an elliptical trajectory?
 
Sagittarius A-Star said:
If the one-way-speed is isotropic or not with reference to an inertial frame is a matter of definition. It depends on how the clock synchonization is defined. It is not "physical".

Source:
https://en.wikipedia.org/wiki/One-way_speed_of_light
If we consider the issue strictly within the framework of special relativity, then you are correct.
I am exploring this model within the framework of special relativity and LET.

According to Lorentz Ether Theory (LET), the laboratory frame moves relative to the ether—the medium of light propagation, which means that the physical one-way speed of light in the laboratory inertial frame is anisotropic.
 
Last edited:
Sagittarius A-Star said:
The formula from the team is correct in SR. You see it, if you look at the spatial components of the 4-force and the 4-acceleration.

https://en.wikipedia.org/wiki/Four-force#In_special_relativity

https://en.wikipedia.org/wiki/Four-acceleration#Four-acceleration_in_inertial_coordinates
Adding to this, note that in an Einstein frame a free-rotating rotor has a simple value for ##\vec{f}## in its rest frame - a constant modulus and directed towards the rotation axis. This gives you a simple shortcut to writing down the path in this frame.

Not so in Tangherlini "frames". The "force" (coordinate force?) will be time varying and probably doesn't point at the axis in general. I don't immediately see a shortcut to the answer - you need a model of the way the rotor behaves and generates forces.

The only way I can think of to do this is to use the Einstein rest frame of the rotor and write down the four-force and four-velocity of the rotor tip, transform them into your arbitrary "ether frame", then use the Tangherlini transforms (you'll need the Jacobean matrix) to transform them into the Tangherlini "frame". Then you can extract the three force, three velocity, and three acceleration and confirm that they match Dale's formula.
 
  • Like
Likes   Reactions: Sagittarius A-Star
AlexOsipovich said:
The team dismissed it, citing a modified version of Newton’s law.
I'm not sure what you think was "dismissed". The force law that you were given is correct in SR. But nobody else except you was claiming that there was an "elliptical trajectory", nor did anyone either confirm or contradict that claim. It's simply something that no one but you has made any claim about at all.