Special Relativity and momentum

[SDG]

Yes. E=mc². That applies to the batteries too.

Where does that mass go? It doesn't teleport from the battery to the photon. It travels with the electrons which the battery put into a higher energy (and thus higher mass) state. These electrons then lose mass at the LED diode when emitting the photon. So the creation of the photon still involves an electron losing mass at the LED.

Close but not completely correct.
Let us assume the following circuit:
The plus/positive battery pole/clamp (1) is connected to switch input pole/clamp (2) with a wire.
The switch output pole/clamp (3) is connected to diode + anode (4) with a wire.
The minus/negative battery pole/clamp (5) is connected to - cathode (6) with a wire.

When the switch is off the voltage delta is between switch input pole (2) and - cathode (6).
When the switch is on the voltage delta is between + anode (4) and - cathode (6).
The energy from battery travels along the wire through EM field.
The voltage delta will excite an electron from the - cathode (6) tip that is available for the flow.
The first electron that would emit light is already at the - cathode (6) tip when the switch is turned on.
The first electron would not come from the battery.

Assuming only one photon emitted the light between on/off change then energy from the battery gave energy through the EM field to the electron at the - cathode (6) tip to become free, to cross the cavity.
The electron does 'Bremsstrahlung' at the end of cavity, the electron changes its kinetic energy for the photon.
In conclusion the energy came to the LED from the battery and the mass is missing from the battery at the center of the flashlight.
The mass is not missing at the tip.

To think about it differently.
There is our setup from the above.
We turn on the switch and we let it go for a day.
We turn off the switch.
Where is the 'mass hole'?
... in the battery, center of the flashlight, there is no 'mass hole' at the LED.
Why this would not apply for one electron and photon?

 

[SDG]

One of the simplest scenarios to do the calculations on is the case of a coaxial cable for the battery to whatever it’s running. You get a radial electric field between the high voltage side and the low voltage side, and you get a circumferential magnetic field from the currents through the wires. And you even get momentum being carried down the wires by the field. It’s interesting stuff, but well beyond SDG’s level of understanding.

And none of that changes the fact that the electron is losing mass at the point of emission.

Using shielded twisted pair cable eliminates significant momentum concern from the EM.

 

[W.D.Clinger]

The electron does 'Bremsstrahlung ' at the end of cavity, the electron changes its kinetic energy for the photon.

I assume the word I highlighted is a comical misspelling of electroluminescence.

In conclusion the energy came to the LED from the battery and the mass is missing from the battery at the center of the flashlight.
The mass is not missing at the tip.

Just before World War II, the United States built four South Dakota class battleships. Two have been preserved as museums, so you can walk around inside them and imagine the process described below.

Their main armament consisted of nine 16-inch guns, but let's consider what happens when only one of those guns is fired. With an armor-piercing round and full propellant, almost 1.5 tons of mass disappears through the muzzle.

During the next 30 seconds or so, a new load can be advanced by one station along the partially mechanized but labor-intensive path that starts deep within the heavily armored barbette beneath the gun turret. When the previous equilibrium is re-established, the gun is once again ready to fire.

When the gun is fired, mass disappears through the tip of the gun. A much slower process redistributes the remaining mass. One could reasonably say the overall net effect is to remove 1.5 tons from the bowels of the ship each time the gun is fired and reloaded.

On the other hand, one cannot reasonably deny the fact that 1.5 tons leave the muzzle of the gun each time it is fired. To deny that fact would be idiotic.

I invite SDG to contemplate this analogy between 16-inch guns and LEDs.

 

[SDG]

I assume the word I highlighted is a comical misspelling of electroluminescence.


Just before World War II, the United States built four South Dakota class battleships. Two have been preserved as museums, so you can walk around inside them and imagine the process described below.

Their main armanent consisted of nine 16-inch guns, but let's consider what happens when only one of those guns is fired. With an armor-piercing round and full propellant, almost 1.5 tons of mass disappears through the muzzle.

During the next 30 seconds or so, a new load can be advanced by one station along the partially mechanized but labor-intensive path that starts deep within the heavily armored barbette beneath the gun turret. When the previous equilibrium is re-established, the gun is once again ready to fire.

When the gun is fired, mass disappears through the tip of the gun. A much slower process redistributes the mass. One could reasonably say the overall net effect is to remove 1.5 tons from the bowels of the ship each time the gun is fired and reloaded.

On the other hand, one cannot reasonably deny the fact that 1.5 tons leave the muzzle of the gun each time it is fired. To deny that fact would be idiotic.

I invite SDG to contemplate this analogy between 16-inch guns and LEDs.

Did you have a chance to read the whole thread from the beginning?
Is there an x component of P_recoil observed from the outside frame?

 

[Ziggurat]

The electron does 'Bremsstrahlung' at the end of cavity, the electron changes its kinetic energy for the photon.

This is wrong. It’s wrong in the case of a semiconductor for the same reason it’s wrong for hydrogen. You really have no clue at all how quantum mechanics work. You’re just throwing around terms that you don’t actually understand.

In conclusion the energy came to the LED from the battery

Originally, yes. But not at the moment of photon creation. The energy (and thus mass) must be transferred to the LED before it can be converted into a photon. So the mass change at the moment of photon creation is still happening at the LED. That mass was transferred there prior to that moment isn’t relevant for the current discussion.

As for your contention of what happens over an extended period of time, that won’t change anything either. The transfer of mass from the battery to the LED will shift the flashlight downwards, so you still introduce angular momentum changes in the moving frame, without spinning the flashlight.

tl;dr is still you don’t understand what happens, you don’t understand angular momentum, and the flashlight will not spin in any reference frame. Your imagined paradox is merely your own failure of understanding. You are not smarter or more knowledgeable than the entire profession of physics. And this is all just a Dunning-Kruger effect.

 

[Ziggurat]

Using shielded twisted pair cable eliminates significant momentum concern from the EM.

No, it doesn’t. You have no clue how electromagnetism works, or how EM fields can carry momentum.

Twisted pair cables are used to reduce EM fields at a distance. But 1) they are worse at doing this than coaxial cables. They are used instead of coaxial cables because they are cheaper and more flexible, not because they are better at shielding. And 2), the really important part, twisted pair reduces fields at a distance, but NOT in the space between the wires. And that’s where the fields will carry momentum in such a case.

 

[SDG]

...

Originally, yes. But not at the moment of photon creation. The energy (and thus mass) must be transferred to the LED before it can be converted into a photon. So the mass change at the moment of photon creation is still happening at the LED. That mass was transferred there prior to that moment isn’t relevant for the current discussion.
...

This is from your post:

The former center of mass is not the new center of mass.

That's how you were defending there is no torque.
I just showed you that the center of mass is the same prior and after the photon emission.

As you say: "The energy (and thus mass) must be transferred to the LED before it can be converted into a photon."
... this is the reason the center of mass does not change.
We control the position of batteries, we can place the center of mass anywhere we want.

Your no torque defence with the varying center of mass position does not hold.

 

[Ziggurat]

That's how you were defending there is no torque.
I just showed you that the center of mass is the same prior and after the photon emission.

Again, this is wrong. You did not show this, because it isn’t true.

We control the position of batteries, we can place the center of mass anywhere we want.

As mass moves from the battery to the front of the flashlight, the center of gravity viewed externally will remain motionless but will move forwards relative to the battery. This means that the flashlight and battery will move backwards.

Then when the photon is emitted, the total mass decreases, but since mass is lost from the front, the center of gravity will move backwards. The center of gravity has returned to the battery, but the battery is already shifted from its original position, accounting for the change in angular momentum and thus torque in the moving frame, with no rotation.

You aren’t keeping track of everything that’s going on, and you are mistaking your own oversights for insight. You are way, way out of your depth.

 

[Dave Rogers]

Did you have a chance to read the whole thread from the beginning?
Is there an x component of P_recoil observed from the outside frame?

No, there isn't. The x component of velocity is unchanging, so there is no x component of impulse. The fundamental problem here is not that relativity is wrong, it's that you don't understand vectors.

Dave

 

[W.D.Clinger]

Did you have a chance to read the whole thread from the beginning?

Yes. It's been amusing.

Is there an x component of P_recoil observed from the outside frame?

The easiest way to analyze this is to use an inertial frame in which the flashlight is initially at rest and never has any velocity or momentum in the horizontal direction (which I assume to be your x axis). Having obtained that solution, its coordinates can be transformed to the coordinates of any other inertial frame. Such transformations are mere changes of spacetime coordinates, and as such cannot change the spacetime manifold itself. The flashlight is not rotating in one inertial frame, so it cannot be rotating in any other inertial frame.

Before someone asks about rotating frames, which can be admissible in general relativity: Rotating frames are not inertial. Besides, this thread is alleged to be about special relativity.

You appear to believe there is an inertial frame in which the flashlight is rotating, so it is obvious (to anyone who understands the basic mathematics) that you've made some kind of mistake. Throughout this thread, the only real question has been the nature of your mistake. We now know you have made quite a variety of mistakes.

 

[SDG]

...

As for your contention of what happens over an extended period of time, that won’t change anything either. The transfer of mass from the battery to the LED will shift the flashlight downwards, so you still introduce angular momentum changes in the moving frame, without spinning the flashlight.

...

That is the reason I said this

If we are worried about EM field being transferred along wire from the batteries to the LEDs then we can have the same setup, creating symmetry, towards the opposite side.
Having said that, the LEDs at the back would not emit photons out but towards each other into a 'triangle' receiver to eliminate torquing at the back.

in my post #61.

Based on this the y position of the center would not change at the moment of emission and the torquing/spinning would start, right?

 

[SDG]

Yes. It's been amusing.


The easiest way to analyze this is to use an inertial frame in which the flashlight is initially at rest and never has any velocity or momentum in the horizontal direction (which I assume to be your x axis). Having obtained that solution, its coordinates can be transformed to the coordinates of any other inertial frame. Such transformations are mere changes of spacetime coordinates, and as such cannot change the spacetime manifold itself. The flashlight is not rotating in one inertial frame, so it cannot be rotating in any other inertial frame.

Before someone asks about rotating frames, which can be admissible in general relativity: Rotating frames are not inertial. Besides, this thread is alleged to be about special relativity.

You appear to believe there is an inertial frame in which the flashlight is rotating, so it is obvious (to anyone who understands the basic mathematics) that you've made some kind of mistake. Throughout this thread, the only real question has been the nature of your mistake. We now know you have made quite a variety of mistakes.

That is the whole point, aberration of light creates the delta in recoil momentum.
The aberration gives it the x component.
One frame (the rest frame) predicts no rotation and all other frames predict some rotation.

 

[Ziggurat]

That is the whole point, aberration of light creates the delta in recoil momentum.
The aberration gives it the x component.
One frame (the rest frame) predicts no rotation and all other frames predict some rotation.

No, SDG. No frames predict rotation. You still don’t understand angular momentum. You still don’t understand how it changes with shifting center of gravity.

You are not nearly as smart as you think you are.

 

[Dave Rogers]

That is the whole point, aberration of light creates the delta in recoil momentum.
The aberration gives it the x component.
One frame (the rest frame) predicts no rotation and all other frames predict some rotation.

I asked you to explain that upthread and you ignored it. I suspect you misunderstand aberration as spectacularly as you misunderstand vectors and light emitting diodes - the very fact you're applying the concept to a single photon strongly suggests it - but your assertion is so meaningless, it's hard to fathom what you think you mean.

Dave

 

[slyjoe]

One of the simplest scenarios to do the calculations on is the case of a coaxial cable for the battery to whatever it’s running. You get a radial electric field between the high voltage side and the low voltage side, and you get a circumferential magnetic field from the currents through the wires. And you even get momentum being carried down the wires by the field. It’s interesting stuff, but well beyond SDG’s level of understanding.

And none of that changes the fact that the electron is losing mass at the point of emission.

I was directing my comment to SDG, but thanks anyway. As an EE, I worried about E and M fields, not so much momentum.

 

[Ziggurat]

I was directing my comment to SDG, but thanks anyway. As an EE, I worried about E and M fields, not so much momentum.

In most situations the momentum involved is too small to be detected, let alone matter, but it's not zero. Classically, it's identical to radiation pressure, and is even calculated the same way (proportional to the integral over space of the cross product between electric and magnetic fields). So you would be safe ignoring it for the same reasons that, say, structural engineers don't worry about radiation pressure on buildings.

 

[Ziggurat]

OK, SDG, let's do a challenge. Let's crunch some numbers. We can make this an unrealistic flashlight (ie, it could output more power than any actual flashlight), so science fiction which obeys basic constraints is OK.

So here's what we do. Pick an initial mass for your flashlight (the mass will change over time). Pick a size (specifically, we need distance from center of gravity to the side which emits). Pick a power output. Pick a velocity for our moving reference frame. Crunch the numbers, see what the recoil pressure is, see what the recoil acceleration is, see what the torque you calculate in the moving frame is, calculate what the angular momentum should be in the moving frame.

And we will see whether or not the flashlight really spins.

 

[SDG]

Before we do numbers we need to understand what is happening in the cavity.

There is the electric field between anode and cathode in -Y direction.

This is straight from Einstein's 1905 paper:

X, Y, Z - electric field in the rest frame
L, M, N - magnetic field in the rest frame
X', Y', Z' - electric field in the moving frame
L', M', N' - magnetic field in the moving frame
\Beta - Lorentz factor

The rest frame has all values 0 only -Y electric field.

We apply transformation from the rest frame to moving frame and we get -Y' electric field in the moving frame, but there is also +N' magnetic field in the moving frame.
This moving frame magnetic field is coming towards us, arrow coming out of the computer screen, right between the anode and cathode.

Do you agree?

 

[Ziggurat]

Before we do numbers we need to understand what is happening in the cavity.

What exactly do you imagine the cavity even is, and what do you think is happening in the cavity? Because in the context of an LED, a cavity isn't what you seem to think it is. It is not a gap between the cathode and the anode.

You are way out of your depth here.

 

[SDG]

What exactly do you imagine the cavity even is, and what do you think is happening in the cavity? Because in the context of an LED, a cavity isn't what you seem to think it is. It is not a gap between the cathode and the anode.

You are way out of your depth here.

Right, I agree, I have not designed my own LED
Having said that:
https://www.photonics.intec.ugent.be/download/pub_1708.pdf

The diagram I showed is simplified but close enough.
There is a possible design; anode and cathode as I showed it.