Tom Van Flandren: Gravity faster than Light

[Grey]

Now see, what you say here seems an impossibility. How can the clocks be "corrected" for relative speed with respect to the "ground" AND with respect to each other (multiple other satellites)? No single correction could possibly compensate for all the relative speeds. I'm sure each GPS satellite had a correction applied that assumes a relative speed with respect to a common reference frame (I would guess the earth). TVF wrote that SR does not allow this, but Lorentzian Relativity is based on such a notion. So I thought the GPS system actually verified LR not SR.

My understanding of SR (weak though it is, and yes I understand that is not an argument!) precludes the application of one correction to a clock, to keep it synchrounous with multiple other clocks, each moving at a different speed relative to the first. That's the whole problem with the twin paradox, after all.

But of course, the satellites don't have to be corrected with respect to each other. That is, in using the satellites to fix your position and time, they don't communicate with each other; each sends a signal that is correlated by your receiver. So each satellite merely (well, it took the complex calculations of general and special relativity to get the results, so maybe it wasn't so mere ) has to be corrected with respect to clocks on the ground. You don't really care if satellite 10 thinks satellite 11 has experienced time dilation, as long as both satellite 10 and 11 have been adjusted so that they agree with your clock. Van Flandern is simply mistaken when he suggests that special and general relativity won't permit this.

As for the issue of gravity waves. In GR the motion of a mass will generate waves in the field that propagate at c. (understood also that in GR, velocity dependent terms will compensate for motion aberration when it comes time to calculate the force vector). So in GR the force is propagated by the same field that propagates waves.

TVFs Meta Model is completely different. The particles carrying the force (and momentum) necessary to push masses around travel at very high speed (on order 10^10 c, lower bound). They have a large mean distance between collisions, which keeps backscattering effects minimal until you get to distances like 2000 light years (i.e. the force falls off with square of distance relatively close to masses, then linearly when very far from masses). The Meta Model does NOT propose that this medium of fast particles propagates waves, or at least, if it does, they will not be observable anytime soon.

Not be observed any time soon is one issue. But if there's a force that propagates at less than infinite speeds, then a time-variant distribution of the source will produce a time-variant field, i.e. waves.

Grey and others, I don't think we have reached understanding on the nut of my problem, the instantaneous correction of position demanded by maxwell's equations (or GR, in the case of gravity), when objects accelerate. Your explanations - even discounting infinite accelerations, as Grey has tried - do not help me. I read that line in Carlip's paper over and over again, and it is clear the equations require something that I cannot believe has been observed, or will ever be observed. I am at wits end, I am sorry.

If it's any consolation, I'm also frustrated at my inability to make it clear that such a propagation delay has indeed been observed. Not only is it directly apparent in the process of electromagnetic radiation, it's a basic part of Maxwell's equations governing all of electrodynamics. If it weren't true, electricity and magnetism just wouldn't work the way they do, and we'd notice because our cool toys that rely on such things wouldn't work.

I'm going to show you what's going on in the process of electromagnetic radiation in slow motion, and perhaps that will help you see what's happening. First, we have you. You're at a transmitter, which has a switch. If you flip the switch up, current flows upward, and positive charge accumulates at the top (negative charge accumulates at the bootm). If you flip the switch down, current flows down and positive charge accumulates at the bottom.

There are a couple results of this. When current is flowing, it produces a magnetic field. If the current is flowing up, the magnetic field will be anticlockwise, and if the current is flowing down, the field lines will point clockwise. If we're reasonably far away so that we don't see the whole field, we'll just see a magnetic field pointing to the right when the current is flowing upward, and left when current is flowing downward.

The other thing that this contraption generates is an electric field. Having a positive charge and negative charge separated by space is an electric dipole, with field lines starting on the positive charge and ending on the negative charge. Again, if I'm relatively far away and facing the dipole, what I'll see is an electric field pointing downward if the current is flowing up (i.e. positive charge is accumulating on the top), and pointing upward if the current is flowing down. Someone check me if I managed to get all my directions right, but it should make a difference as far as the demonstration goes.

You also have three observers to take measurements to help you out. On is stationed fairly close to you, so there's no appreciable transmission delay. One is stationed one light second away, and one is two light seconds away. They're equipped with suitable equipment to detect electric and magnetic fields, and we'll ignore the fact that the signal strength would be much smaller for the observers further out.

For the pedantic, not that the images of what's happening at the transmitter, as well as what the observers see, should really be stacked ontop of each other, or drawn with the magnetic field lines pointing into or out of the page. However, either of those would be a lot harder to see, so please imagine taking the observer 0 image and putting it in front of the transmitter, facing it, observer 1 goes behind observer 0, but one light second further away, and observer 2 is behind observer 1, another light second away.

So, we begin at time t=0, with the current flowing upward, assuming that it's been doing so for some time. Here's the picture we see.

Time t=0

   Trans    Obs 0       Obs 1       Obs 2

   ++
  +  ^
   + |        |           |           |
     |        |           |           |
     |     ---+-->     ---+-->     ---+-->
     |        |           |           |
   - |        v           v           v
  -- |
    -

Now after a second, you decide to flip the switch. Here's what it looks like.

Time t=1

   Trans    Obs 0       Obs 1       Obs 2

   --
  -  |
   - |        ^           |           |
     |        |           |           |
     |     <--+---     ---+-->     ---+-->
     |        |           |           |
   + |        |           v           v
  ++ v
    +

Note that as soon as you flip the switch, the electric field around the transmitter changes. Observer 0, who is close, sees this right away. To observers 1 and 2 though, the information that this change has happened hasn't reached them yet. As far as they can tell, you haven't moved the switch, and current is still flowing upward. The electric and magnetic fields they see are exactly what they would see if you hadn't moved the switch. We wait another second, and here's the picture.

Time t=2

   Trans    Obs 0       Obs 1       Obs 2

   --
  -  |
   - |        ^           ^           |
     |        |           |           |
     |     <--+---     <--+---     ---+-->
     |        |           |           |
   + |        |           |           v
  ++ v
    +

Now observer 1 sees the change, and his detector immediately switches to the new readings, showing that the electric and magnetic fields have changed. Observer 2 is still oblivious, though, and has no idea that you've done anything.

Time t=3

   Trans    Obs 0       Obs 1       Obs 2

   --
  -  |
   - |        ^           ^           ^
     |        |           |           |
     |     <--+---     <--+---     <--+---
     |        |           |           |
   + |        |           |           |
  ++ v
    +

Finally, after a two second propagation delay, observer 2 notices the change. You flip the switch the other way.

Time t=4

   Trans    Obs 0       Obs 1       Obs 2

   ++
  +  ^
   + |        |           ^           ^
     |        |           |           |
     |     ---+-->     <--+---     <--+---
     |        |           |           |
   - |        v           |           |
  -- |
    -

Now after one second, you flip it back down again.

Time t=5

   Trans    Obs 0       Obs 1       Obs 2

   --
  -  |
   - |        ^           |           ^
     |        |           |           |
     |     <--+---     ---+-->     <--+---
     |        |           |           |
   + |        |           v           |
  ++ v
    +

Notice again that observer 0's measurements reflect the field which is currently seen at the transmitter. Observer 1, though, sees what was going on at the transmitter a second ago. Again, observer 1 can't tell that you just had the switch changed for a second, because the measurements there are exactly the same as if you'd flipped the switch up and left it there. The field at observer 1's position is based on what the field would be like if the current were still flowing upward.

So there's this weird propagation delay, where the electric and magnetic fields aren't based on what the charge and current distribution is now, it's based on what the charge and current distribution would be if they were still doing what they were doing a second ago, for observer 1, or two seconds ago, for observer 2.

If this weren't the case, if the field were updated instantaneously, then there would be no delay in the propagation of radio signals. When something was broadcast, you'd hear it instantly, and it's trivial to show that this just isn't the case.

[Boris]

Grey, I appreciate all the effort you are putting into explaining propagation delay to me. Isn't it clear that I understand propagation delay? If I didn't understand propagation delay, then lack of aberration in force vectors wouldn't be an issue for me - propagation delay is what causes aberration in a force vector between two moving charges (or masses).

Let's take it back several posts (like ten days now, or it seems). If there is propagation delay, then there should be an aberration in force vectors between two charges that are moving with respect to one another. Maxwell's equations, however, provide velocity dependent terms that just cancel the aberration. This is described in Carlip's paper.

Okay I can go with that. The trouble is that it doesn't work very cleanly for accelerating charges. Carlip points this out too, where he says:

...if a uniformly moving charge suddenly stops at position z(s0), the field at a distant location x will continue to point toward its extrapolated position - even though the charge never actually reaches that position

This has NOTHING to do with my understanding of propagation delay per se. Grey, you are not addressing the problem that I have with Carlip. This will be the third time I've quoted the problem statement in Carlip's paper, and as far as I can tell, either you don't understand what Carlip is saying here, or I don't know what else to say!

Carlip is saying that for some time (the propagation delay time, actually) the corrected, extrapolated force vector will point to a location that the other particle will never reach - a completely fictitious position. The force will point at an empty place in space! Then after the delay is passed for the updated information to reach the detector, after the field "update" arrives, then "Oh!" the vector will snap to the correct position of the other particle.

The example given assumes a particle suddenly accelerating (infinitely accelerating, perhaps), but the "error" should be evident even if there is a finite acceleration. ANY acceleration. It just looks like the equations will make the correct predictions just to first order, but not beyond.

Now I will stick to TVF's proposal, that a theory which posits transmission of the force at a much higher speed, where aberration is naturally very small, will not have this problem.

This does not affect wave speed or propagation delays for waves. Just as gravity waves are treated with a different medium from gravity force transmission, so will electromagnetic waves be treated differently from the electric force. It makes everything simpler, and sofar as I can tell, your voluminous efforts notwithstanding, describes reality more accurately.

[Grey]

Let's take it back several posts (like ten days now, or it seems). If there is propagation delay, then there should be an aberration in force vectors between two charges that are moving with respect to one another. Maxwell's equations, however, provide velocity dependent terms that just cancel the aberration. This is described in Carlip's paper.

Okay I can go with that.

So do you believe that there are velocity dependent terms that will cancel the aberration? This is actually required by Lorentz invariance, if you accept that. (Lorentz invariance just says that if I have two chrged particles moving uniformly with respect to each other, I can decide particle 1 is moving and particle 2 is stationary, or vice versa. If particle 1 is stationary, then clearly the field it produces must point directly toward the particle. However, if I decide that particle 1 is the moving particle, then there must be a velocity dependent term that cancels aberration, or the field would point somewhere else.)

The trouble is that it doesn't work very cleanly for accelerating charges. Carlip points this out too, where he says:

...if a uniformly moving charge suddenly stops at position z(s0), the field at a distant location x will continue to point toward its extrapolated position - even though the charge never actually reaches that position

This has NOTHING to do with my understanding of propagation delay per se. Grey, you are not addressing the problem that I have with Carlip. This will be the third time I've quoted the problem statement in Carlip's paper, and as far as I can tell, either you don't understand what Carlip is saying here, or I don't know what else to say!

My apologies for being frustrating; it hasn't been deliberate. It really does work fine for accelerating charges, but I'm not sure I can explain it more clearly without getting into the quantitative calculations, which I'd hope to avoid. Partly just because a qualitative demonstration would be more convincing. And it is tied closely to the propagation delay, because without that, there would be no delay when a particle changes motion, and hence no propagation delay in electromagnetic radiation.

You seem to accept that there is a propagation delay in the transmission of the electromagnetic field. If you also believe that there is no such thing as absolute motion, then we also know that there has to be a compensating velcoity dependent term (or else we'd be able to tell which particle was "really" moving in the above example). However, if the motion changes, information about the new velocity takes just as long to propagate as any other signal would.

Carlip is saying that for some time (the propagation delay time, actually) the corrected, extrapolated force vector will point to a location that the other particle will never reach - a completely fictitious position. The force will point at an empty place in space! Then after the delay is passed for the updated information to reach the detector, after the field "update" arrives, then "Oh!" the vector will snap to the correct position of the other particle.

That's exactly correct, and that is what happens during a radiation process. I've been in touch with Dr. Carlip, actually, to see if he happens to know of a direct experimental result that doesn't involve radiation. His first suggestion was EM radiation, of course, and I gave some of the reasons you were unsatisfied with that. Hopefuly, he'll know of a demonstration of the existence of the velocity dependent term that would be more clear to you.

This does not affect wave speed or propagation delays for waves. Just as gravity waves are treated with a different medium from gravity force transmission, so will electromagnetic waves be treated differently from the electric force. It makes everything simpler, and sofar as I can tell, your voluminous efforts notwithstanding, describes reality more accurately.

Except that, as I've stated, and time varying field will produce waves that move at the propagation speed for that field. If you move the field generatingobject one way, the field will move, and that disturbance will propagate out at some speed. If you move it back, that disturbance will likewise propagate outward. If I do so regularly, I'll see an oscillating field at some distant observing station, and the propagation delay has to be exactly the same as that for the field, since that's what's "waving". That's exactly what we see when we measure electromagnetic radiation: electric and magnetic fields that vary with time. Radiation is nothing more nor less than a variation in the field itself.

I'll wait to post further until I hear more from Dr. Carlip, and perhaps we'll be able to find evidence that would be easier for you to accept.

[Boris]

Well, I would be very honored! I didn't think my question was worthy of such an audience. I am very curious to see what you come up with.

thanks!

Boris